Time and Phase Coherent Loudspeakers
|
Silver Member Username: KbearCanada Post Number: 896 Registered: Dec-06 | I thought I'd kick off a discussion on this topic, since it doesn't get much. I've done a bit of reading on time and phase aligned speakers. There is a nice, easy to understand overview on Thiel's website: http://www.thielaudio.com/THIEL_Site05/Pages/FAQs/faqtimephase.html. I know there are many skeptics who do not believe these two ideals are important. I believe they cite tests where listeners were not sensitive to phase and time errors, and John Atkinson of Stereophile indicates that poor time and phase alignment has not prevented many speakers from being recommended by his publication. There are many other key aspects to good speaker design, which must be executed properly. If they are not then no measure of great coherence will save a speaker. However, he goes on to state that phase and time coherence does not hurt if a speaker is designed well otherwise. Then there is the effect of the room and seating position, which will affect time and phase alignment and will probably prevent true time and phase alignment most of the time. However, I still find it curious that more speaker makers do not use these design theories. The fact that higher frequencies travel faster in itself is a logical reason to move the tweeter further back. I believe to accomplish this all you need to do is employ an offset tweeter, with the woofer on top. Of course, some speaker makers do this. Most do not. I would think if the speakers are toed in and aimed at the listener, much of the sound will be direct (not reflected) and thus will reach the listener based in large part on how the drivers are situated. But can we hear this? I don't know. Like JA said though, it can't hurt. I can only imagine that if high frequencies reach us before they are supposed to that the timing relationships of the various instruments is thrown off, even if ever so slightly. A speaker might still sound great, but the question becomes would the same speaker produce an even more cohesive sound if it was remade to be time aligned? Could you just compare it to a different speaker that was time aligned? You could, but then other things change as well. To my mind you'd almost need to have the exact same speaker, only one version is time-aligned and the other not. And certainly, the best measured step responses (as per JA), which should theoretically help yield better sound, are from brands like Reference 3a and Thiel. I guess the bottom line for me is that it just seems to be one more aspect of speaker design that you'd probably want to optimize. So many details are looked after in the design of a hi fi speaker, so why not this one? How many speakers have we all heard where the sound just doesn't grab us? Are time and phase alignment a factor? Phase-aligned speakers might cost more to implement, what with needing first order crossovers, or no crossovers at all. I gather it is the slow roll off of frequencies that necessitates the use of very high quality drivers, though perhaps there is even more than this to consider. However, as I said above, I don't see what is preventing the creation of more time-aligned speakers. |
|
Gold Member Username: HawkbillyNova Scotia Canada Post Number: 1184 Registered: Jul-07 | Like you Dan, I've read various opinions on the importance of phase and time alignment. I've seen a number of different approaches to this with multi-driver speakers (sloped baffles, duel concentric drivers (like your DC6's), recessed tweeters, etc.....but the simpliest approach many have chosen is to avoid multiple drivers altogether and go with a single full range driver. No right or wrong necessarily, just different points of view. I've read some articles that claim that phase alignment is critical in the core midrange frequencies, and less so in the upper frequencies.....which I suppose accounts for some designers chosing to xover at higher frequencies (Zu for instance). Now, take those same Zu speakers (the Essence for example) that so many people have raved about (I haven't heard them personally) but JA noted several concerns in its measured performance.....yet they are almost universally lauded as a best-in-class speaker. Sometimes measurement factors don't indicate how much success I design will have......for whatever reasons. Like everything else in speaker design, there are compromises. Single driver speakers have limitations that multi-driver speakers don't, and vise versa. I can't imagine that phase alignment isn't a factor in what the listener perceives, although they may not be able to link what they hear to phase issues or the lack thereof. The question is, does it affect how the listener enjoys what they are hearing. Perhaps phase and time coherance is less influential on their experience than transient response, bass impact, or macro/micro dynamics. Ultimately, the listener will prefer whatever they prefer, and in most cases without knowing why. I guess if you study and read around the hobby enough, you can see a pattern. Perhaps the speakers you liked over the years tend to be a certain design. If that's the case, then you might be able to piece together what specifically is important to you. I've done that I think, at least to a degree and when I upgrade speakers it will likely be to another single driver, or a fullrange + super tweeter design. Not that I couldn't happily live with other types of speakers, but those tend to make music best for me....at least in my listening experiences to date. Perhaps it is the lack of phase issues, perhaps it is other factors. |
|
Platinum Member Username: ArtkAlbany, Oregon USA Post Number: 13779 Registered: Feb-05 | Actually this topic was beat to death by a former member here. It was the hottest topic here several years ago and one of the most fiercely debated. If I never saw the term again I'd be thrilled. |
|
Gold Member Username: Stu_pittStamford, Connecticut USA Post Number: 4287 Registered: May-05 | Mauimusicman! There's been several clowns who amused me over the years, who are no longer with us. Paul was one of my favorites for a laugh, either laughing with him or at him. I guess at the end of the day, it doesn't matter as long as I've had a good laugh. Sorry to derail the topic. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15738 Registered: May-04 | . Two things - at least - to get straight right off the bat; don't believe everything you read from a manufacturer's webpage and, in audio I know of no single benefit that doesn't come with at least two tradeoffs. If I give you one thing, I'll take away two other things which might be as or even more important to your perception of music. The time/phase/frequency response issue is very difficult to discuss as all three parameters are thought of as separate qualities in the design of a crossover and the complete speaker system. Each is a distinct parameter yet each affects the other two. Very much like enclosure volume, bass extension and sensitivity are all discrete but interdependent parameters of speaker design; change one and you affect the other two in some way. Consider the illustration for "time correction" in the Thiel article, it is too simple to cover what actually occurs in a real world speaker system and therefore leads you to a conclusion which cannot be achieved in a real world speaker system. Drivers are illustrated as delivering off axis response - which is true - in a sloped front, time aligned system such as the Thiels but no mention is made that off axis response is less smooth than on axis response and therefore requires other compensations for good frequency response which is of prime consideration when discussing "soundstage" and "imaging", the very reasons Thiel gives for using time correction. "Other compensations" include additional crossover parts which add to the complexity of the system and bring into question both total electrical phase shift and the resulting amount of power loss each capacitor,resistor and inductor present to the amplifier and to the very low level signals which make up a substantial portion of the music in an audiophile quality system. In regard to the latter only, does a speaker system sound good at only high volumes? If so, is that a good loudspeaker overall? Next up, either you will need to decrease the working diameter of each driver to raise the frequency where the dispersion of the driver begins to beam and/or you will have to produce drivers which have inherently broader dispersion characteristics. The broader the dispersion of the driver, the more the room reflections then account for what is perceived in room by the listener which will affect the preceived time arrival and frequency response of the speaker system. These are factors a speaker designer cannot control and which then place higher costs onto the buyer to ensure the room is properly treated to minimize room reflections issues. It also places further restriction on the listener as time alignment is really only going to be correct for a two way system in one location away from the speaker system. Additional drivers make that point ever more complex and the fact most listeners share a room with other domestic duties only complicates the issue of whether the listener will actually benefit from a time aligned speaker system. Even if time is accounted for in placement, what if the speaker's bass response does not accomodate that specific speaker/listener position? So far physics dictates that dispersion narrows as the wavelength of any frequency approaches and then becomes shorter than the diameter of the surface reproducing that frequency. On the other hand, all other qualities being equal, larger diameter drivers have the ability to produce lower frequencies. This becomes important not so much in bass drivers but far more so in mid and high frequency drivers where the designer is always looking for lower and lower response from tweeters and midrange drivers. The lower the response limits of, say, a tweeter in a two way system, the lower the designer can place the crossover point. Theory and practice suggests we will perceive "better" sound if we hear more of any individual instrument's sound from one coherent source - one single driver. The octave difference of a woofer is more dramatic than that of a tweeter in that the difference between 25Hz and 200Hz is three octaves yet the difference between 1kHz and 8kHz is also only three octaves. "If you've never had a chance to look at the way frequency response corresponds to the sound of instruments, you might want to note first that the divisions along the bottom line of our chart are anything but even. When most people first visualize the frequency range from 20 Hz to 20,000 Hz, they imagine a nice, linear, tape-measure span of measurement, on which the marked increments are as equal as the inch or centimeter markings on a ruler. But when you look at an actual response chart, the measures along the lateral line are definitely not an equal distance apart. In fact, the seemingly "small" span between 20Hz and 40 Hz is actually wider than the 6,000 Hz of difference between 10,000 and 16,000Hz. That's because the vibrations of the heavy-hitting bass instruments of music are ponderous and far apart, while the successively higher pitched instruments going up the scale vibrate faster and faster, and closer together. The frequency scale of music (and all sound) isn't linear but logarithmic -- which is probably why mathematics and music often seem to go so well together; http://www.psbspeakers.com/audio-topics/The-Frequencies-of-Music The fundamental frequencies of classical music played through acoustic instruments ceases to exist above roughly 5kHz and for most "popular music" much lower than that. As that same article states, "Harmonics are what let you tell instruments apart. Without them, similar instruments that played the same frequencies would sound the same." Due to the size of the driver required to produce satisfactory bass response that same driver begins to beam its frequencies rather quickly as the response moves from omni-directional, long wave lengths (http://www.soundoctor.com/freq.htm of bass instruments into the much shorter frequencies found in the upper bass/lower midrange. This beaming affect alters the in room "power response" of the driver as the shorter, more directional frequencies are not bounced off the room surfaces in equal proportion to the broadly dispersed lower midrange frequencies reproduced by the mid/tweeter. (Remember, in a first order crossover with a -6dB per octave slope quite a bit of energy is shared between the two drivers which are both reproducing signals at and beyond the specified crossover point.) Actual in-room frequency response then becomes more ragged as the listener perceives a distinct dip in the power response of the speaker system; http://www.linkwitzlab.com/design_of_loudspeakers.htm Add to this the poor off axis response of the drivers being "time corrected" on a sloping baffle and you have more problems with good in room measured and perceived response while solving only a small portion of the real world issues of "time" in a loudspeaker. To avoid beaming from drivers as much as possible the low pass to high pass crossover point for a two way system needs to be low as possible with the high frequency driver carrying the bulk of the frequency range. This necessitates a fairly small driver which in turn cannot reproduce lower frequencies without further compromise. To have the most coherent response of any single instrument's sound the crossover between drivers should be placed either as high or as low as possible. Placing the crossover point in the middle of the human vocal range as has been the custom for most loudspeakers due to the physical demands of the drivers is, in most opinions, the absolute worst location for splitting the "sound" of the human voice (to which the human ear is the most sensitive). In other words, the designer must treat the woofer as a near full range driver and allow it to produce as much as the first eight octaves of the frequency bandwidth bringing in the tweeter only at the highest frequencies to supplement the decreased dispersion and ragged response of any driver meant to produce 40Hz. Alternately, the designer can place a crossover as low in frequency as possible but this compromises response in that most high frequency drivers simply do not have sufficient working diameter to cleanly reproduce even 1.5Khz and, if they do, then they might be so large as to beam at their upper limit. Now take that driver struggling to make good clean sound at 1.5kHz and set it off axis to the listener's ears where its response and distortion product are at their worst. Consider also that with the lower the frequency requirements of such a driver, the lower the ultimate power handling of the driver in real world terms which then compromises the ability of the speaker system to accomplish real world listening levels without break up and distortion or even failure. The designer can compensate for these issues by increasing the diameter of the driver's working surface but that will only lead to the tradeoff of narrowed dispersion at ever lower frequencies. In today's market where home theater applications demand good sound quality across a very broad seating pattern, designers have been forced to make ever more widely dispersed response the norm which has for better or for worse moved away from the traditional audiophile speaker with a very narrow sweetspot. Broad dispersion and good in room power response across a very broad seating area bring their own tradeoffs to speaker system design most of which are antithetical to the goals of time correction by physical means; i.e, sloping baffles. That bit alone deals with about 1/3 or less of the major issues a designer must consider when building a "time corrected" speaker system. And we haven't even come close to discussing phase correction and how it plays with time correction and frequency response in a speaker system used in a real world room. While Thiel promotes their first order crossovers as the "right" way to achieve "phase correct" performance, most designers would argue with that conclusion as any first order crossover will result in a 90° electrical phase shift of voltage/current. Any additional crossover components forming additional frequency response correction filters (notch filters) used in such a crossover to compensate for the problems of broad dispersion/frequency response drivers will increase the difficulty in acheiving a truly phase correct output from two drivers. Add a third driver to solve a few problems such as driver beaming and reduced power handling and overall the problems become expotentially more difficult. Additionally, in order to discuss "phase" in a louspeaker system, we would first have to ensure we were all on the same page regarding "absolute phase", "relative phase" and 'electrical phase" and then discus how each is affected by the music, the recording and design choices and how each affects the listener's perception. From the Thiel article; "Your ear is set up for, and is much better at, determining location from time information rather than loudness information." This takes us back to our primitive ancestry and the fight or flight perception training (which still exists in present day humans) to perceive the location of a predator or food as we take in the location of the sound by the time differences we perceive between our two ears. Once we begin to discuss perception (what transpires after our ear drums) we get into a territory of cognitive perception which is really only now being minimally understood by researchers. All of this should give you pause in considering the benefits vs. tradeoffs of time and phase alignment in loudspeaker systems in use in real world rooms. This is an area of research, discussion and disagreement which has existed in audio for decades and is still only now being somewhat understood. Each benefit has, as should be obvious from the above simplifications, multiple tradeoffs which will affect the listener's perception of music reproduced through any specific loudspeaker in any particular room. Add to this the fact most modern recordings do not attend to time and phase relationships and you find yourself in a morass of contradiciting ideals once you even scratch the surface of the topic. For now consider one more basic fact of audio; in a higly competitive market place where literally thousands of speaker and amplifier designs exist for the choosing it pays to set yourself apart from the crowd by presenting a good story which most potential buyer's do not fully understand and which would take considerable time to unravel. If time and phase correction are what you choose to present as your "story" - and they are good stories no doubt about that, you have a good chance many listeners will not take the time to research all of the potential benefits and tradeoffs your story might mean to them in the real world. In short, one of the very first lessons most salespeople will learn is simple, sell the sizzle and not the steak. Hope that helps. . |
|
Platinum Member Username: ArtkAlbany, Oregon USA Post Number: 13780 Registered: Feb-05 | At least we get to start this one off with an intelligent post relative to the topic. Thank you, Jan. Yeah it was Maui and it was not amusing as Paul tended to be. Maui was bound and determined to prove that his speakers and those like them were the only good sounding speakers. He's still out there on other forums and has tempered his approach quite a bit so hopefully if he sees this he enters as Jan did. Paul was just funny. His Yammie and CV's. When I first arrived Paul and Edster were at war...it was highly entertaining and highly toxic. Sorry, carry on. |
|
Gold Member Username: GavdawgAlbany, New York Post Number: 1610 Registered: Nov-06 | I thought Maui had SDAT |
|
Gold Member Username: GavdawgAlbany, New York Post Number: 1611 Registered: Nov-06 | Let me put it this way... buy what sounds good to you! I have heard speakers that measure awful that sound wonderful, and ones that measure well that sound awful! +/- 3db doesn't really tell you the full picture. |
|
Gold Member Username: Stu_pittStamford, Connecticut USA Post Number: 4290 Registered: May-05 | Tawaun had SDATs. Maui had Green Mountain Audio Europas, I believe. Or at least that was what he was claiming. That guy had some pretty hallarious posts. I think more were intentionally idiotic than many people realized. Sorry again... |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15739 Registered: May-04 | . All of that was before registration was manadatory on this forum. How many different people posted as Maui? Three, if I remember right. All were equally nuts. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1244 Registered: Oct-10 | I've heard a lot of interesting arguements when it comes to speaker design. I stumbled across a website once where someone insisted that ALL drivers in a speaker should be cone type and the cones AND the coil formers must be made of paper. I've seen arguements for thick cabinet walls, thin cabinet walls, aluminum cones and cones of various other materials. I've heard that speakers should be aimed directly at the listening possition. I've tried that in every room I've had a system in with different speakers, etc. This position never sounded good to me. Speaker wire manufacturers have argued in favor of the time correcting benefits of bi-wiring for years and now they claim that it extends to interconnect cables. So if you have seperates, time correction (according to Monster, Audio Quest & others) is applied from source to preamp, from pre to power and power amp to speakers. IF there is any validity to time correction within wires, then my concern with doing so at every stage is the possiblity of over doing it to the point where bass arrives before treble. The smaller the room, or shorter distance from speaker to ear, the more likely this is to become a problem. Most of what companies in general say about their products is said to make you think their equipment is better than that of the competition. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15740 Registered: May-04 | . http://www.rane.com/note160.html http://www.linkwitzlab.com/crossovers.htm http://www.linkwitzlab.com/frontiers_5.htm#V "Crossover transient response Good transient response means that an impulsive sound arrives at your ears at a single instant in time, as one nice solid jolt of sound pressure. Poor transient response means the system smears the sound over a few tenths of a millisecond, or into several impulses arriving at slightly different times. An excellent test for transient response is how a system reproduces a square wave. A square wave represents an extreme example of an abrupt transition in sound pressure, and contains a broad spectrum of frequencies. A sound system must reproduce these frequencies with a flat amplitude response, and with a flat (or linear with frequency) phase response, or the wave will no longer be square. This type of sound does not occur in real music. However a system that can accurately reproduce a square wave can also accurately reproduce anything that real music can throw at it. It is an acid test. [Thanks to Bob Stanton for pointing out my description of an ideal speaker response to a square wave was incorrect. See discussion in the section on Thiele-Small analysis]. The crossover network response to one cycle of a 1kHz square wave for a 1st order Butterworth shows perfect behavior [46.1kb] (this and a few other examples below are the same as those used in the Crossover Demo Section). There are four curves in the graph. The signals sent to the tweeter, midrange, and woofer are shown as the blue, green, and red curves respectively. The heavy black curve is the total response. With the assumed ideal drivers, this is simply the coherent sum of the voltages across the terminals of the three drivers. Frequencies above 50 kHz have been eliminated, causing the ripple in the response. Other than that, it is perfectly square. The responses shown are for the tree topology. The 1st-order series and dual-series topologies also produce a perfect square wave. The 1st-order parallel topology performance with the same filters (not shown) is good, but it is not perfect. It is also not perfect with Dickason's component values. Each time division in the figure is 0.2 millisecond. The human ear just begins to hear two distinct clicks when they are separated by half this difference, as discussed elsewhere. Even though ears do respond to time variations smaller than 0.1 millisecond, features closer in time than 0.1 millisecond tend to be blurred together in our perception of sound. In any case, variations within the time frame shown in the graph are perceptible, in principle. The square wave response of a 4th order Linkwitz-Riley is not a pretty picture [47kb]. A 4th order Butterworth, or the 4th order Dickason design, create a square wave response (not shown) that looks very similar. The big question is: can you hear the difference between a good and a bad square wave response? My own experiments indicate that for time-alignment the answer is yes, but for the crossover phase effects shown in these curves, no. The low-frequency response of the system (below 100 Hz) has essentially no effect on the edges of a 1kHz square wave; low-frequencies mainly effect the flat top part of the wave. At subwoofer frequencies it is shown in the section on room acoustics that, even with a perfect speaker system, the response is chaotic. I believe that transient response for woofer frequencies is almost certainly irrelevant, as far as the effect of a crossover is concerned. The enclosure can cause significant ringing, which may be perceptible"; http://www.silcom.com/~aludwig/Sysdes/Crossove_Design.htm#Transient_response . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1245 Registered: Oct-10 | I also have to add here that I seriously doubt that pushing the tweeter and midrange back a few inches will make any noticeable difference in timing. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15742 Registered: May-04 | . Then why do you suppose speaker designers do it? Thiel is not the only manufacturer designing time aligned systems. Why would they go to the trouble and expense if they didn't believe there was a benefit? Have you taken the time to read any of the links I've provided? |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15743 Registered: May-04 | . "Do you hear what I hear?" http://mixonline.com/mag/audio_hear_hear/# "How we localize sound" http://www.aip.org/pt/nov99/locsound.html "Music and the Brain" http://www.centerformusicmedicine.org/pdfs-music-and-brain/Music_and_the_Brain_B y_Norman_M._Weinberger.pdf Floyd Toole; http://www.infinitysystems.com/home/technology/whitepapers/inf-rooms_2.pdf http://sound.westhost.com/pcmm.htm http://sound.westhost.com/ptd.htm . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1246 Registered: Oct-10 | Before seeing your last two posts in this thread? Didn't have chance. I looked at them and to me it just doesn't seem like a few inches would make much difference. Wire companies make time correction speaker wire and now time correction interconnect cables. How much of this has to do with what speaker and wire companies believe and how much is about gimmicks and public perception. I can tell you from my own experience that bi-wired speakers sound better to me than single wired ones. The timing of the music is better. Why though? Is it really because the wires are making a time correction or is there another explanation? |
|
Gold Member Username: HawkbillyNova Scotia Canada Post Number: 1188 Registered: Jul-07 | " I looked at them and to me it just doesn't seem like a few inches would make much difference." Have you never noticed a difference when moving your speakers a few inches back, further apart, closer to the wall ? |
|
Gold Member Username: Nickelbut10Canada Post Number: 3399 Registered: Jun-07 | I would think it would make a difference when it came to physically moving a tweeter/midrange in a speaker design. Its not just a box with some woofers and tweets jammed in it. There is way, way more to location/layout and result to measurement with the cabinet then you would think. A lot more. Its not easy. Every little bit makes a difference. |
|
Silver Member Username: KbearCanada Post Number: 897 Registered: Dec-06 | I could see it making a difference. It's not like a non-time aligned speaker has highs that are way ahead of lows. The integration is already very good. Time alignment simply a way to fix the slight, but perhaps significant, timing discrepancy. An inch or two is enough. Bass doesn't take long to reach our ears. When you see your woofer move, to me at least, I hear the sound almost instantly. So how far back do you really want to move the tweeter? And the fact that Stereophile's measurements show step response for time and phase aligned speakers is superior, pretty much confirms that the methods used by the likes of Thiel make a difference. Although I'm sure Stereophile is doubtful of it's ultimate importance. As I think is Jan. I'm personally just curious about this subject right now. While the theory behind time and phase alignment makes sense to me, I'm certainly not trying to argue it is a necessity for good sound. For me to take on Jan would be like going into a gun fight with a plastic knife. By the way, thanks for the explanations Jan. I have yet to really read them...I'll need to take some time to do so, so I'll probably leave it till the weekend when I've got a half hour to sit down and really concentrate on it. In my readings I did come across some of those old maui posts. I didn't really read through most of those threads....maybe I should! But with all due respect to those old discussions, I kind of doubt they really ever dealt with the subject the way they could have, and the way Jan just did. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15745 Registered: May-04 | . Dan, I sold Thiel speakers for several years at one of the highest volume Thiel shops in the West and I have also sold against Thiel for even more years. You have to remember that as a salesperson, depending on the feedback I got from a client, I could argue either for or against the concepts Thiel promotes and which I do personally find less than perfect. But we were one of the highest volume Thiel dealers in the country so I could certainly sell the Thiel ideas when that was what the client had convinced themself they needed. What I discussed above was primarily a disagreement with Thiel's general philosophy. I could just as easily turn the debate around and provide many of the strong points of Thiel design if you'd like. Furthermore, the same arguments against or for Thiel do not necessarily apply to all time aligned speakers. Tradeoffs, Dan, that's what its all about, every designer and every listener should be making decisions based upon their own concepts of how live music sounds to them and not on how well written a manufacturer's white papers are done. As I've stated numerous times, IMO the most important thing is that you have priorities established and that you do not allow those priorities to stagnate. Have some concepts in your head as to what is important to you in the reproduction of music but never allow a decision to go unchallenged as you experience and learn more about music and audio. I think if you do more reading for and against time alignment, you'll find out a bit more about what is important/possible in technical terms. Whether you can then perceive those things in the real world becomes a significant challenge to your priorities. Listen to the opposing view points; listen to a Quad ESL, a Walsh driver and a Lowther to understand what is changing in the mind of the designer. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15746 Registered: May-04 | . "So how far back do you really want to move the tweeter?" In time aligned systems the general idea is to align the acoustic centers of each driver so the launching platforms for all frequencies are in the same vertical plane relative the the listener's position. Think single driver or co-incident drivers (as Thiel and KEF employ) and you'll get the idea of how drivers should theoretically be aligned in space for proper time correction. Point source or line array speakers present an alternative yet not contradictory concept of time alignment. "And the fact that Stereophile's measurements show step response for time and phase aligned speakers is superior ... " IMO correcting for electrical phase between drivers is far more important than time alignment alone. Such corrections are achieved in the design and construction of the crossover and not in the physical construction of the enclosure/baffle. However, due to the increasing inductance of the voice coils within each driver due to heat build up as a consequence of friction, true phase coherence is much more difficult to achieve in practice. . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1248 Registered: Oct-10 | So what you're saying Jan is that the idea, at least as Theil would have us believe is to combine the advantages of 3 drivers with the advantages of one driver. Is that correct? Chris, I've seen the differences in distances that you describe. However, these are differences in accoustics, not the timing of different parts of the spectrum. The tweeter being back a few inches didn't seem like a viable way to achieve time correction. Jan's explanation makes more sense provided I'm understanding him correctly. |
|
Gold Member Username: MagfanUSA Post Number: 1996 Registered: Oct-07 | My panels are perhaps a reasonable example of line source. they are 2-way and it makes a difference if the tweeters are 'in' (closer to listener) or 'out' (further away). Because the crossover is a first order hi pass for tweeter and 2nd order low pass for the woofer doesn't that make them 90degrees apart? Is that relative phase or electrical phase? Seems the tweeter closer produces better imaging and much less beaminess. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1249 Registered: Oct-10 | Hmmm! Now that's certainly interesting Leo. The tweeters being closer than the woofers is better with your speakers? |
|
Gold Member Username: MagfanUSA Post Number: 1997 Registered: Oct-07 | Yes, i have panels so the entire speaker is on a sheet of mylar. The woofer and tweeter are wired electrically out of phase and are 90degrees out of phase due to 1st / 2nd order crossover. The woofer / tweeter sections are next to one another on the same sheet. I was just floating the question that IF the woofer section were 'ahead' of the tweeter section, the arrival to the listener would be time aligned if the tweeter were closer to the listener. I simply don't know how or if that plays into the panel sound which I like. |
|
Gold Member Username: HawkbillyNova Scotia Canada Post Number: 1189 Registered: Jul-07 | "Chris, I've seen the differences in distances that you describe. However, these are differences in accoustics, not the timing of different parts of the spectrum." In my mind they are one and the same. Moving the whole speaker or moving a driver both affect acoustics. If moving a speaker an inch makes an accoustical difference (better or worse), then I have no trouble believing moving a tweeter back the same distance in relation to the woofer will have an impact, one way or the other. That's all I was getting at. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1250 Registered: Oct-10 | That puts an interesting spin on things Leo. Here Theil is putting the midrange and tweeter further back, most likely using the same order crossover for all divers. Meanwhile, panels are mixed resulting in the need for reverse compensation. This sort of goes back to what I said about time correction cables and the possibilty that if true, said time correction could go too far resulting in the bass arriving ahead of the treble. My omnis can't really be positioned with the tweeter back since it sits on top of the woofer and both drivers are top firing. In other words, if they were front firing, the tweeter would be in front of the woofer. One of the many things I like about them is their flexiblity in that room location and positioning do not effect their sound as much as other speakers I've heard. Of course these things always have some effect, but less on omnis than on most speakers. So while some places and positions are better than others, I have yet to find a bad place or position for them. Conversations like this keep me wondering how much manufacturers deal in reality and how much in what they believe the consumer's perception will be. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1251 Registered: Oct-10 | Do you have box speakers or some type you can tip backward Chris? If you do and you have a way to support them, like maybe books under the fronts of the spkrs, you could actually try this and see how well it works. Such an experiment would not work with my spkrs. |
|
Platinum Member Username: ArtkAlbany, Oregon USA Post Number: 13781 Registered: Feb-05 | Richard Vandersteen is another speaker designer who champions time and phase coherent designs and does it through physical placement of drivers. I enjoy his speakers more than Thiels. http://www.vandersteen.com/vandersteeninterview.pdf |
|
Gold Member Username: SuperjazzyjamesPost Number: 1252 Registered: Oct-10 | Wow Art! Nice to know I'm not the only early riser! I'd like to hear some Vandersteens some time. I've heard many good things about them. |
|
Gold Member Username: HawkbillyNova Scotia Canada Post Number: 1190 Registered: Jul-07 | "Do you have box speakers or some type you can tip backward Chris? If you do and you have a way to support them, like maybe books under the fronts of the spkrs, you could actually try this and see how well it works. Such an experiment would not work with my spkrs." Wouldn't work with mine either. I have single driver speakers. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1253 Registered: Oct-10 | Okay, nevermind Chris! Lol! |
|
Platinum Member Username: ArtkAlbany, Oregon USA Post Number: 13782 Registered: Feb-05 | During the work week I get up at 3:30 am for my morning walk while my wife gets ready for work and then I have my chance and it's out the door at 6:05... |
|
Platinum Member Username: NuckPost Number: 15674 Registered: Dec-04 | Nice little speakers, too! |
|
Gold Member Username: MagfanUSA Post Number: 2001 Registered: Oct-07 | Panel fans are of 2 groups. The 'tilt 'em back' school and the 'leave them vertical' school. Tilt influences time align nature of panel. Sound travels about 14" per ms. IF you can hear sounds or be influenced by sounds say.....0.5 ms apart, then yes, moving a tweeter a couple inches will make a perceptible change. from Jans 1st post in this thread:: For now consider one more basic fact of audio; in a higly competitive market place where literally thousands of speaker and amplifier designs exist for the choosing it pays to set yourself apart from the crowd by presenting a good story which most potential buyer's do not fully understand and which would take considerable time to unravel. If time and phase correction are what you choose to present as your "story" - and they are good stories no doubt about that, you have a good chance many listeners will not take the time to research all of the potential benefits and tradeoffs your story might mean to them in the real world. In short, one of the very first lessons most salespeople will learn is simple, sell the sizzle and not the steak. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1254 Registered: Oct-10 | I hear ya Art. I'm up at 4:00. My Sony boom box has single driver speakers. I used use it at work. While I'm certain that they don't compare to Chris's speakers, I was amazed at clarity these things have considering they are built into $50 unit! Single driver technology has come a long way. |
|
Platinum Member Username: NuckPost Number: 15675 Registered: Dec-04 | You can have a Mulligan for the Sony comment, JJ. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1255 Registered: Oct-10 | Thanks Nuck! Lol! I know, a Sony boom box is pretty far from high end. I haven't knowingly heard a high end, single driver speaker. I'm just saying... Ok then, obviously single driver technology has come a long way since the 60s & 70s if even one top company is making SD speakers. Is that good? |
|
Gold Member Username: HawkbillyNova Scotia Canada Post Number: 1191 Registered: Jul-07 | "Nice little speakers, too!" Damn straight Nuck. Although, I had their replacements ready to order when I lost most of my roof in the windstorm here last month. So much for new speakers. Not a problem though, I can easily live with the Lings for a while longer. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1257 Registered: Oct-10 | Ooh! Sorry to hear about that Chris! Most of your roof, how bad was the rest of the damage? I'm guessing the speakers will have to be replaced some time soon? |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15750 Registered: May-04 | . As usual the British do an excellent job of explaining their technology. The KEF Uni-Q driver is the technological starting point for the Thiel coaxial system; http://www.kef.com/Wiki/en/QSeries2010/html_en/kef_uni_q.html (Click on "Components" and "Technology" for drop downs.) If I remember correctly, Thiel's first systems using coincident/coaxial driver technology were built with KEF drivers and then later changed to Thiel drivers built in house. One of the issues with using first order crossovers is the ability to also use extremely well behaved drivers. With a first order (-6dB) filter, the simplest filter type available in a passive system, the electrical function of the filter alone will result in the driver having a very shallow roll out. If the designer places the cross at 1,500Hz (the crossover "point" being that frequency where both drive units have rolled out their response by -3dB), a first order electrical filter would have only dropped the frequency information reaching the tweeter by -6dB at posssibly 1kHz or lower. (Minus six decibels is where most designers consider a driver's output to be essentially unusable.) However, 1kHz is well beyond the usuable range for a 1" high frequency driver to produce clean signal information. All things being equal if the designer selects an even smaller diameter driver to obtain broader dispersion at higher frequencies, along with lowering the ultimate power handling of the system the issues of poor low frequency bandwidth response become more severe. If we click on the link Art provided to R. Vandersteen's interview we'll see on page 5 that "Filter Theory 101" states a "good" (first order) crossover will allow for flat response of the drivers one octave above and one octave beneath the actual crossover target frequency. That requirement should push the crossover frequency in our theoretical system above 1500Hz to at least 2500-3000Hz to achieve a "good crossover" with the vast majority of today's high frequency drivers. Should we choose the lower 2500Hz frequency the tweeter would still need to operate with clean output to approximately 1400-1500Hz which is about the point where the limits of modern driver technology allow. More importantly, the low pass filter is also operating at -6dB per octave which now allows the lower frequency driver to have usuable output up into the 4-5kHz range. This too is well beyond the normal clean operating range of most low frequency drivers which means by now the driver is beaming its output rather severely in the same frequency range where the upper frequency driver is (hopefully) enjoying wide dispersion. A disparity in dispersion characteristics results in less than ideal in room power response of the speaker system as certain bandwidths are given higher levels of reflected energy while other bands are provided lower reflected in room energy. While the speaker might perform well in the near field where measurements are taken, it's performance in a typical room will be somewhat ragged in perceived frequency response. Combine beaming frequencies and poor power response with poor out of bandwidth performance at the limits of both drivers which results in ragged frequency response and high distortion components then spread that over the two octaves (that region where both the lower and upper frequency drivers are sharing output) and place it smack in the middle of the frequency range where the human ear is quite sensitive to anomalies of timbre due to disparate drivers and dissimilar driver materials and you can forecast the problems which might arise as we begin to seek transparency from the system. If timing and phase are to be maintained there are several other solutions to the problems the designer has faced when using first order filters. Steeper filters can be used but steeper filters tend toward ringing which leaves the driver in motion well after the actual electrical input to the driver has ceased. "Ringing" is not a component of the original signal and can therefore be considered a distortion. Further, with each additonal step in filter action; i.e. second (-12dB), third (-18dB) and fourth (-24dB) order filters, the complexity of the filter itself becomes more of an issue as high tolerances are often the norm for crossover components even in high end designs. Maintaining a tighty toleranced, symetrical frequency response is imperative to good soundstaging and imaging along with ambience retrieval in a high quality two channel system. Higher order filters then require more control over parts selection with more parts rejected which will naturally raise the final cost. More to the point when discussing phase and timing issues would be the electrical phase shift created by higher order filters. With each capacitor and inductor the voltage component of the signal is shifted in electrical phase against the current component of the signal. (This is what you are seeing when you look at the electrical phase shift as measured by Stereophile, the 0° point is voltage and current moving together and any deviation away from 0° represents either a dominant capacitve or a dominant inductive component of the crossover combined with the inductance of the drivers' voice coil. The greater the phase shift in degrees, the more voltage and current have been shifted out of synch and the harder the amplifier must work to accomplish the task of moving the driver, thereby creating either an easy or a difficult speaker to drive.) As the electrical phase shift becomes more severe with each higher order filter type (due to the higher number of parts needed to accomplish the filter action) not only will the speaker become more difficult to drive but those shared frequencies of any instrument's fundamental/harmonic structure spread across multiple drivers will be further out of acoustic phase between the two drivers. Looking at the step response measurements of any speaker you'll usually notice a mention of whether the drivers have been connected in positive or reversed electrical phase meaning in this case whether the positive lead from the crossover has been connected to the "+" terminal of both drivers or reversed to the upper frequency driver. Using a first order filter, most designers will connect the drivers in reverse phase which will place the tweeter behind the motion of the woofer and therefore provide some rudimentary time alignment of the system without physically moving the tweeter. At least this is what should happen in theory. As you can see from the measurements taken on this Thiel speaker the type of filter employed (Butterworth, Linkwitz-Reily, etc.) will alter the "lobing" of the driver combination which will in turn change the correct listening height for the speaker; http://www.stereophile.com/content/thiel-cs37-loudspeaker-measurements The slope of the baffle and, in this case, the physical distance between driver units exacerbates the issue of correct listening height in this particular Thiel. Now we have a theoretical speaker using high order filters which will create electrical and acoustical phase shifts between drivers and since phase is imtimately linked to time the time alignment of the system is also compromised. Two "somewhat exceptions" to this rule exist. A second order filter will shift electrical phase approximately 180° and a fourth order filter will shift phase approximately 360°. By either reversing the electrical connections to the two drivers when using a second order filter or leaving the electrical phase intact at the connections when using a fourth order filter, acoustic phase will be maintained despite the fact absolute timing will lag in the higher frequency driver with each filter type. Depending on where the filters are placed in frequency (out of the bandwidth where the ear is most sensitive would be a good choice), the resulting time smear may or may not be objectionable to various listeners. The larger problem still remains that "time" when compensated for through either driver location or electrical means will still only be correct for a narrow band of frequencies when the listener is "X" distance from the speaker system. Given that most crossovers will result in some degree of lobing one of the difficulties of good speaker design can then become matching the correct time alignment position of the listener with the best frequency response position of the listener (distance away from the drivers). Obviously, all of this assumes a completely non-reflective room at this point. Let's return to the top of this post and think about an advantage of purpose built drivers for speaker systems using first order filters. One of the first considerations here is the ability of the designer to build in the mechanical roll out of the driver to minimize out of bandwidth frequency response errors. From the link to Thiel in the op in reference to drivers filtered through first order low pass/high pass crossovers; "To properly execute this system in practice requires very high quality, wide bandwidth drivers and that the impedance and response variations of the drivers and the cabinet be compensated for across a wide range of frequencies. This is a complicated task since what is important is that the acoustic output of the drivers roll off at 6 dB/octave and not simply for the networks themselves to have 6 dB/octave roll-offs. For example, if a typical tweeter with a lower roll-off of 12 dB/octave is combined with a 6 dB/octave network, the result is an acoustical output which rolls off at 18 dB/octave. To achieve a first order system in practice, the tweeter must have a very low and very well damped resonance with high output capability and the network must in fact have a complex response. Both of these requirements are expensive to implement." By building drivers in house or obtaining drivers purpose built to specific design goals the combined mechanical roll out of the drivers along with the electrical filter action can be made to be steeper than the electrical filter alone would allow which can minimize out of bandwidth errors in the driver itself. As is typical, cost becomes a determining factor in how well this solution can be implemented. http://www.kef.com/Resources/Current%20Products/Loudspeakers/Q%20Series%202010/_ brochures/q_tech_explained_010910_en.pdf http://thielaudio.com/THIEL_Site05/PDF_files/PDF_tech_papers/CS3_7premtech.pdf . |
|
Gold Member Username: MagfanUSA Post Number: 2004 Registered: Oct-07 | Jan, if you choose 2.5k for crossover and use 1st order for hi and low pass, Electrically the speakers will -6db 1 octave above and below crossover. However, if the driver when fed a 'flat' signal is mechanically rolled off, you add that to the electrical rolloff. Doesn't that help the problem? The -6db of the crossover and the additional -6db (varies per driver) natural rolloff of the drivers 1 octave above and below crossover should put that driver 'out of the picture' at that frequency. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15751 Registered: May-04 | . At this point we have a speaker system which can be time and phase aligned through various mechanical and electrical means. The speaker designer sorts through the possible solutions available and makes the decision to build a system using first order LP/HP filters with well behaved drivers and tightly toleranced parts. The issue of electrical phase is still an issue some designers might care to deal with. As explained above a shift in electrical phase indicates a shift in the voltage/current components of the signal from the amplifier. When the designer is looking to retrieve the ultra small value ambience clues and the tiny nuances of performance which set a high end system apart from the rest of the market, a shift in electrical phase represents a potential loss of information as voltage/current diverge and "less work" is done. Going back to the Thiel article in the op, "A first order system achieves its perfect results by keeping the phase shift of each filter to less than 90° so that it can be canceled with a filter that has an identical phase shift of the opposite direction. The phase shift is kept low by using very gradual (6 dB/octave) roll-off slopes which produce a phase lag of 45° for the low frequency driver and a phase lead of 45° for the high frequency driver at the crossover point. Because the phase shift of each driver is much less than 90° and is equal and opposite, their outputs combine to produce a system output with no phase shift and perfect transient response." IMO there's a bit more advertising hyperbole involved in that explanation and "no phase shift and perfect transient response" is impossible to achieve real world in a multi-way speaker system. Thiel claims a very low maximum allowable (average) electrical phase shift for their current speaker; systems this is, however, achieved by placing filters against one another and the resulting average phase shift appears to be what Thiel would like you to pay attention to. As has been common with Thiel speakers for decades, due to the high parts count of a typical Thiel crossover the minimum impedance of the system is quite low and the combined impedance/phase angle of the system will require a substantial amplifier to perform well. "My panels are perhaps a reasonable example of line source. they are 2-way and it makes a difference if the tweeters are 'in' (closer to listener) or 'out' (further away). Because the crossover is a first order hi pass for tweeter and 2nd order low pass for the woofer doesn't that make them 90degrees apart? Is that relative phase or electrical phase? Seems the tweeter closer produces better imaging and much less beaminess." "Richard Vandersteen is another speaker designer who champions time and phase coherent designs and does it through physical placement of drivers. I enjoy his speakers more than Thiels." " ... John Atkinson of Stereophile indicates that poor time and phase alignment has not prevented many speakers from being recommended by his publication." While time and phase alignment are popular qualities to be aimed for they are seldom achieved in the real world. As should be obvious, there are as many ways to think about speaker design as there are designers. Vandersteen has been a popular brand since roughly the time of Thiel's inception. Also employing first order filters to achieve a reasonable facsimile of time and phase coherence, Vandersteen prefers not to build a sloped baffle. Instead Vandersteen uses what might be considered a "baffle=less" system where each driver is enclosed in its own box and the grill cloth sock is held up by a series of dowels; http://www.stereophile.com/floorloudspeakers/284/ Spica was another popular brand in the 1990's which built time and phase aligned speakers and their top model was the Angelus. Unlike Thiel and Vandersteen, Spica was not an adherent to only first order crosssovers. "In the time domain, the Angelus's impulse response on the tweeter axis (fig.5) appears time-coherent, due to the combination of its crossover's performance--first-order high-pass to the tweeter, fourth-order Bessel low-pass to the woofer--and the time alignment of the drive-units due to the sloping front baffle. This is confirmed by the step response (fig.6), which shows an excellent right-triangle shape due to the same-polarity outputs of the tweeter and woofer arriving at the microphone at the same time, spoiled only by a leading-edge overshoot associated with the treble plateau noted in fig.2.; http://www.stereophile.com/content/spica-angelus-loudspeaker-1993-measurements (Take the time to read the entire Spica review as there is a substantial amount of useful information to be found there.) A fourth order (-24dB/360°) low pass filter and the tweeter connected in "correct" electrical phase to achieve phase and time correction and excellent step (transient) response. The front baffle of the Angelus was sloped backwards placing the acoustic centers of the two way system in vertical alignment as with Thiel and Vandersteen but, in this case, intentionally sending the on axis response of the drivers towards the ceiling for maximum perception of "ambience" and "space" while using carefully controlled drivers with good off axis response for the listening position. Additionally, the Angelus had a very wide baffle for the tweeter and a pinched in baffle for the woofer which gave each driver a different spactial loading in which to operate; http://cgi.ebay.com/SPICA-Angelus-speakers-Great-Condition-/120652495115?pt=Spea kers_Subwoofers The front baffle of the Angelus was covered in a thick felt blanket to minimize reflected energy due to diffraction (somewhat similar to the Vandersteens) which gave the speaker a narrow sweetspot (unlike the Vandersteens). So, in the 1990's Thiel, Vandersteen and Spica all tried to achieve many of the same goals in speaker design yet all three went about the process of design with different decisions being made. Of the three, Vandersteen and Spica were relatively easy loads on any amplifier while the Thiel required a substantial amplifier due to its phase/impedance. The Thiel line has always tended toward reviews which emphasize careful selection of associated equipment while the Spicas and the Vandersteens were/are comfortable with a wide variety of high quality amps. Vandersteen's house sound is characterized by a "warm", free flowing musical nature with a substantial bottom. Thiel is to many listeners "bright" and to those who buy them "analytical" or "accurate" with a tight, extended bottom end response. Spica was - the company no longer exists as the designer, John Bau, got tired of the high end BS - the speaker that existed in between the other two and IMO offered a musical performance as close to that of the Quad electrostatics as any dynamic driver system I've heard. In regards to your panels, leo, the tweeter placement (in or out) is, to my knowledge, more a function of room response than of time correction. "Because the crossover is a first order hi pass for tweeter and 2nd order low pass for the woofer doesn't that make them 90degrees apart? Is that relative phase or electrical phase?" In theory, yes, they are more or less 90° shifted in electrical phase but theory gets somewhat lost in the real world of building a multi-element crossover. Looking at the graphs of impedance/phase for any speaker the electrical phase is not a stepped shift at any one frequency but rather a gradual swing or curve (or curly-que in your favorite Smith Chart) as the filters affect rising or falling frequency response. The 90° spec would be the shift in electrical phase while the acoustic phase of the system would be dictated by the positive or reversed connection of the tweeter from the crossover. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15752 Registered: May-04 | . "Jan, if you choose 2.5k for crossover and use 1st order for hi and low pass, Electrically the speakers will -6db 1 octave above and below crossover. The specified crossover "target frequency" is where the two drivers have both lost three decibels of energy. The combined energy of the two drivers will bring the overall response back up to approximately flat response in this region. Have a look at Fig. 3 and Fig. 4 here for an example; http://www.stereophile.com/content/proac-response-d-two-loudspeaker-measurements Therefore, the acoustic -6dB point is unlikely to occur at exactly one octave below the crossover target frequency. The electrical filter has an approximate -6dB per octave action but given the -3dB acoustic roll out at the actual crossover target frequency, the driver will have already lost a good bit of energy before it hits that target frequency. That's why a high pass 1,500Hz first order electrical filter probably doesn't have a measured -6dB acoustic roll off point of 750Hz for the tweeter - or one octave below the target. From the Thiel article, "This is a complicated task since what is important is that the acoustic output of the drivers roll off at 6 dB/octave and not simply for the networks themselves to have 6 dB/octave roll-offs. For example, if a typical tweeter with a lower roll-off of 12 dB/octave is combined with a 6 dB/octave network, the result is an acoustical output which rolls off at 18 dB/octave. To achieve a first order system in practice, the tweeter must have a very low and very well damped resonance with high output capability and the network must in fact have a complex response. Both of these requirements are expensive to implement." What is being implied here is the "expensive" Thiel tweeter must have the ability to roll out its energy at a controlled rate yet maintain the ability to perform well at out of bandwidth frequencies considerably beneath those of the average, cost effective driver. "However, if the driver when fed a 'flat' signal is mechanically rolled off, you add that to the electrical rolloff. Doesn't that help the problem? The -6db of the crossover and the additional -6db (varies per driver) natural rolloff of the drivers 1 octave above and below crossover should put that driver 'out of the picture' at that frequency." Yes, that's what you should probably take away from the information in the Thiel article. Include in that information though that filters are filters no matter whether they are electrical or mechanical in nature. If the electrical filter is first order and the mechanical is also first order, the combined second order filter would - at those shared frequencies - exhibit an approximate phase shift similar to an electrical second order. So while the designer might gain some advantage of lower out of bandwidth frequency errors, the phase shift errors would still exist within a narrow band of frequencies as the driver is presenting - if I remember this correctly- a fast rising, infinite impedance to the signal. The trick then, as I understand/remember it, is to make those errors exist at a point where they are inconsequential to the perceived output of the system; in other words, high enough or low enough in frequency to be "unusable" energy (more than -6db down in level). Using very shallow first order filters this becomes a more complicated issue due to the poor performance of the average driver. As Thiel suggests, expensive drivers and tight quality control over the entire production process will be required to achieve the performance target. Or you could use steeper filters but that would also present tradeoffs to the final product. Then you have to examine the results a company such as Spica achieved with fourth order filters and "correct" electrical connection of the drivers and decide which route you prefer to follow. Or you scrap the whole idea and build a single driver, full range system that is inherently time and phase correct that will operate as a single point source just as an acoustic musical instrument will. . |
|
Gold Member Username: MagfanUSA Post Number: 2006 Registered: Oct-07 | Got it. And it is slightly more complicated than just 'where is the crossover frequency'? IF you want again.....2.5k for cross over, the hi pass will start above and the low pass below the 'target' frequency. I have a print of the Magnepan frequency response here somewher which clearly shows the 'crossing' at about 600hz, with the 2 distinctly different slopes of the 1st and 2nd order portions of the crossover. I keep forgetting it is the SUMMED response which counts. In the case of Magnepan, the 2 sections are wired electrically out of phase. Gotta crash. Digging a ditch tomorrow to help drain friends yard. He had 3" and rising of standing water with no drainage. Previous work had screwed it up. I found 20' of buried and disconnected drain pipe today which had 1 end right at the low spot. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1258 Registered: Oct-10 | "IMO, there is a bit more advertizing hyperbole in that statement..." Doesn't that sum up every advertizement? Not say that there is NO validity to what they say, but it always seems that there is way more hype than real deal in ads. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15753 Registered: May-04 | . Let's take a minute to look at this statement from Thiel, "This is a complicated task since what is important is that the acoustic output of the drivers roll off at 6 dB/octave and not simply for the networks themselves to have 6 dB/octave roll-offs. For example, if a typical tweeter with a lower roll-off of 12 dB/octave is combined with a 6 dB/octave network, the result is an acoustical output which rolls off at 18 dB/octave. To achieve a first order system in practice, the tweeter must have a very low and very well damped resonance with high output capability and the network must in fact have a complex response. Both of these requirements are expensive to implement." What do you see them saying when you read those sentences? Well, first, as has been discussed, the acoustic output of the system is of primary importance and not just the electrical filter action of the crossover. As Thiel states the issue, "if a typical tweeter with a lower roll-off of 12 dB/octave is combined with a 6 dB/octave network ... ", it would appear they are implying they would be running the driver down to the limits of its useable frequency response at which point the mechanical roll out of the driver would begin to account for the "typical" -12dB level change. That would be an unusual decision on the part of the designer since this would easily place a substantial portion of the driver's out of bandwidth errors within the range of audible frequency response. So the designer is more likely to select a target frequency for the crossover a few hundred Hz higher than what Thiel suggests in that sentence. In real world useage this makes driver selection and driver quality even more difficult and even more expensive. One alternative to this -12dB roll out of a "typical" tweeter would be to alter the virtual space the driver sees or works into. Going back to the Spica Angelus you can see the tweeter has been mounted at the broadest horizontal section of the baffle and is offset from the centerline to ensure the two halves of the radiating pattern do not have similar distances to the edge of the baffle and therefore do not have similar frequencies where the half sphere pattern of the baffle disappears into open space radiation. Mounting the tweeter on a wide baffle in this way is similar to placing your speaker systems flat against a wall and taking advantage of the gain provided by the reflective surface. (For now we'll ignore the baffle step compensation required for a typical box enclosure mounted against a wall and focus only on the effect of the baffle on the radiation pattern and subsequent frequency response of the driver; http://sound.westhost.com/bafflestep.htm) In the Spica Angelus the wide baffle at tweeter height provides several "bounces" of the driver's lower, longer frequencies which represent the lower operating limit of the driver. Using these reflections much as you would use a near-wall placement to add a few db of bass extension and gain to a woofer, Spica has effectively altered the useable lower frequency limit of the tweeter and in the process added a few extra dB of gain to the overall level of the tweeter (which aids power handling) and a few hundred Hz of useable lower end response (which makes the driver more well suited to first order, high pass filters). This substantially aids the use of the first order, shallow filter action of the crossover by achieving useable response a few hundred Hz lower than a similar tweeter mounted in free space or, as is the case with the "typical" Thiel tweeter and those used by Vandersteen, where a very narrow baffle immediately presents open or "whole" space radiation to the driver. In today's market (home theater use often dictates that) baffles have shrunk to the physical limits of the drivers making the wide baffle/large footprint concept of the Spica far less desirable in a domestic situation. However, the narrow baffle design presents still more tradeoffs for companies such as Thiel and Vandersteen when they try to combine time and phase alignment in first order filtered systems. Looking again at how each designer has chosen to deal with the problems presented by a driver's need for a mounting baffle we see three different decisions being made. Vandersteen chooses to mount his drivers in separate enclosures just large enough to support the driver and then stacked on top of each other. The very small baffle for the mid/tweeter will reduce the potential gain obtained from the wide baffle of the Spica and thereby reduce the total efficiency of the driver neccesitating a driver selection with still higher output which in turn can compromise power handling and raise the expense of the system. The hard, sharp 90° edges of the enclosures can result in some refraction components in the speaker's response; http://www.stereophile.com/content/vandersteen-2ce-signature-loudspeaker-measure ments Thiel has for decades rounded the front edges of their baffle to "ease" the radiation pattern's leap off the front baffle plane into open space. In general, measurements for Thiel systems indicate no real issues with refraction from the baffle in the upper frequencies; http://www.stereophile.com/content/thiel-cs37-loudspeaker-measurements Power handling and efficiency issues are similar to those found in the Vandersteen when used with first order filters. Finally, Spica's solution meant covering the wide baffle with a very thick blanket of felt damping material which effectively absorbed much of the energy of those frequencies travelling across the baffle with each subsequent bounce and then champhering the final edge of the baffle to smooth the transition into whole space for those frequencies which survived the trip. The felt damping material further narrowed the sweet spot for the Angelus as the listener was hearing primarily on axis vertical response from the drivers.. If you've checked out the link to baffle step compensation, you'll find another issue designers deal with in making choices regarding the unknowns they face once their product is in the hands of the buyer. How much compensation will be required is dependent upon the placement of the speaker relative to the surrounding reflective surfaces. Ignoring the manufacturer's suggestion for correct placement either by choice or by neccesity might result in a very large and audbile peak in response which again would not show up on near field measurements. http://www.stereophile.com/content/revels-kevin-voecks . . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1259 Registered: Oct-10 | Busy day! Finally got a chance to read and re-read all the posts and links. I guess time and phase correction are far from being exact sciences! |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15775 Registered: May-04 | . Since Von Schweikert has come up in another thread I thought this might be a good time to revisit this thread with more designer concepts on time and phase response. As with any other manufacturer's white papers you need to take into account the concept of selling what the designer believes to be vital to the reader "getting" their sound and technology. Von Scweikert has probably done as much research into the field of psychoacoustics as any other working speaker designer and far more than most - who have generally done nothing. Like John Bau with Spica, Von Schweikert has taken to heart the writings and thoughts of Richard Heyser and bases much of his work on Heyser's theories regarding hearing vs measurements. Any contact with Heyser's writings would IMO be time well spent if audio is of concern to you. "FIRST ORDER VERSUS FOURTH ORDER NETWORKS Experiments validated the concept of consistent (not the same as coherent) phase vs. frequency linearity in a 180-degree arc around the speaker system, and appeared to work far better than phase coherency limited to the axial tweeter response. As is commonly known, first order crossovers have severe problems with driver overlap, which lead to an effect called lobbing. This problem is related to the fact that the drivers can sum perfectly only on one very narrow axis, since the path length from the drivers to all other axes cannot sum to unity, in either frequency, phase, or transient response! This not-surprising effect is due to the mathematics governing wave transmission and is easily verified by simple experiments or "doing the math." Thus the measured polar vertical off-axis response, for instance +/- 180 degrees, of speakers using first order crossovers will typically exhibit amplitude dips and peaks of up to 18dB caused by the lobbing effects caused by uneven path lengths and will have severe phase distortion as well. The ear/brain hearing mechanism can easily hear this effect, due to reflected response from the room boundaries even though the listener may be seated on the perfect axis. Not amazingly, the ear is far more critical than any type of test equipment yet devised, so these effects cannot be ignored on a psychoacoustic level, especially in a normally reverberant living rooms where the off-axis response dominates the perceived frequency and phase response. OFF-AXIS PHASE VS. AMPLITUDE CONSISTENCY I have termed my method of enabling consistent phase vs. frequency behavior Global Axis Integration tm, since my design constructs a consistent polar response both in the amplitude and time domains, both horizontally and vertically. Not only does this radiation pattern enable the listener to perceive well-balanced frequency and harmonic integration from almost anywhere in the listening side of the room, but also enhances sound-stage imaging over a 180 degree axis horizontally and 70 degrees vertically. This is especially important psychoacoustically, since the ear/brain hearing mechanism responds favorably to this reconstructed sound wave pattern. This Global Axis Integration method consists of a carefully engineered radiation pattern created by front and rear driver arrays. Proprietary circuits form steep 24 dB acoustic crossover slopes at specially selected frequencies without the penalties of induced ringing and excessive phase delay. These slopes are necessary to limit lobbing effects and non-linear off-axis response, and actually enable the consistent phase behavior necessary between drivers. The architecture of the circuitry resembles first and second order filters combined with Zobel conjugate compensators in parallel. By using a minimum of high quality parts in series with the drivers, the sound remains transparent, yet the control over phase and amplitude can be corrected with the paralleled Zobel circuits."; http://www.vonschweikert.com/techspecs/2.php If you happen to be unfamiliar with Von Schweikert's designs, you might want to spend some time with his web site too. "I guess time and phase correction are far from being exact sciences!" I would disagree. The science of time and phase alignment is rather well codified, designers well understand the functions of the various filters and how soundwaves move through air. The phase response of capacitors and inductors has been understood for decades. It is not that any of this is a mystery, it is that the science of speaker design is vastly different than the art of speaker design. That's where the exceptional designers can be seen to stand out from the average. The decisions being made and how to work a series of decisions into a successful system are what define the art of design. In short, knowing how to listen. The emerging field of psychoacustics has brought a new understanding to the many failures and discrepancies of measurements vs perception. In that regard Heyser designed an amplifier which measured as "perfect" as was possible at the time and found the amplifier simply didn't sound very good when reproducing music. This had little to do with typical measurements we discuss when we say we each hear differently and therefore have our own preferences. We spend little to no time discussing psychoacoustics on this forum, which is probably just as well, but they are very likely to represent the next major direction high end audio pursues. . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1287 Registered: Oct-10 | Okay Jan, then how are differing room sizes and accoustings, listening distances and distances between speakers accounted for with regard to phase and time correction? In other words if 2 people buy the same pair of speakers and person A has a 10 x 15 room and listens at 12 feet away with the speakers by a 10 ft wall with speakers 6 ft appart, while person B has a 20 x 28 room and sits 18 ft away with the speakers by a 28 ft wall with the speakers 12 ft appart, wouldn't phase and time correction work better in one room than in the other? Or is it possible for these factors to be reconciled just as well in both rooms with same speakers? I ask because it seem as though there would be at least some discrepency between the two arrangements. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15778 Registered: May-04 | . Electrical phase is consistent to the listener no matter the room or the listening distance since it is accounted for within the speaker system. There's nothing about the room that alters the order of the crossover filter or the +/- connection of the drivers. Acoustic phase perceived by the listener is very complex and is dependent upon many factors, a few of which have to do with the room and the listening distance away from the speaker system. Certainly the resulting acoustic phase of the crossover filters plus the mechanical roll out of the drivers is consistent at any distance away from the speaker in any room. There is no one answer to the issue of acoustic phase as perceived by the listener due to room reflections and, therefore, there are too many answers to try to provide any sort of definitive answer in this forum other than to say the electrical/acoustic phase of the speaker system at the speaker system itself first determines the relative acoustic phase of the speaker system as soundwaves leave the drivers. This in phase/not in phase quality at the speaker system obviously has some effect on how those soundwaves arrive at the listening position. Just as with on axis frequency linearity in a speaker system if it is not correct as it leaves the drivers or as it sums at a specific distance away from the drivers, then it stands very little chance of being correct at any other time or at any other distance. "Time" is the most fragile and the most fungible of all in that if you assume higher frequency soundwaves are less dense and therefore travel at a slightly higher rate of speed than would lower, denser soundwaves, you also have to conclude that those changes in speed are not stepped in frequency like the crossover. If the change in relative speed is constantly changing with frequency, then there can only be a handful of frequencies which will be "in time" with each other upon arrival at a listening position any distance away from the drivers. This can be plotted and specified by the designer but the designer has no real control over how their product is used by the consumer. Add in the time differences of reflected soundwaves and the concept of time alignment as a real world fact is hardly a realistic possibility in the average listening room. Well thought out and well executed room design and/or treatments will make a preceptible difference in "time alignment" in any real world listening situation but the designer can't count on that being the case with every speaker installation. That is one reason many designers feel an accurately phase aligned system is more important than a so called time aligned system. Most designers feel you cannot have a time aligned system unless the system is first electrically/mechanically phase aligned if there is more than a single driver involved. If you noticed Von Schweikert places a good deal of emphsis on the phase alignment of his systems yet, while he is likely to align the acoustic centers of the drivers in an array, he doesn't place as much emphasis on strict time alignment as will Thiel who have made this a signature of their designs since their inception. However, if you do as Thiel attempts and you manipulate the electrical/mechanical phase of the entire system and you use coaxial drivers as Thiel does, in theory you could have a correctly time aligned, virtual point source system at the driver's face while the entire system would be electrically phase aligned to less than 45°. For testing purposes most would consider that arrangement to be technically about as good as it gets unless a single driver were being used. Moving the crossover frequency higher up the range makes for better integration of time in most cases but also introduces other, more restrictive trade offs for many designers/listeners. Von Schweikert in at least one current system places his target crossover frequency at 8kHz high pass to a ribbon tweeter while using what he refers to as an "augmented one way" system. Keep in mind perceived time alignment of the total signal is dependent upon the perception of the listener and it is not completely understood as to how each listener registers the quality of "time" in their brains. As is usually the case, several different methods for time alignment can be used with each getting different marks for accuracy from different listeners. Your question, however, shows why speakers are not designed in average listening rooms, there are too many variables to contend with as a designer. Speaker designers target certain goals that can be spec'd and then measured as their beginning reference point. Once those goals have been achieved in the testing facility most good designers will then perform listening tests in a variety of known rooms. The goal there is to have real world performance that has its basis in the measurements taken in environments no person would inhabit and with material more complex than sine waves. If the two situations do not provide consistent results, then any good designer would probably begin to search for why they do not. If you've read the literature, you'll see that was what Heyser did and that is the genesis for much of what Von Schweikert has done and why his speaker systems are something everyone should pay attention to whether you can afford them or not. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15779 Registered: May-04 | . http://www.aes.org/aeshc/docs/jaes.obit/JAES_V35_5_PG413.pdf http://karlnordstrom.ca/research/?p=299 http://forums.klipsch.com/forums/p/94684/959304.aspx (Highly recommended reading, each sentence informs you how much there is to know and how little of it you actually even think you know.) Seen in Stereophile April 2008. From Time Delay Spectrometry, the Audio Engineering Society's collection of Richard Heyser's papers for the AES and Audio magazine. Let me put this another way. You out there, Golden Ears, the person who couldn't care less about present technical measurements but thinks of sound in gestalt terms as a holistic experience. You're right, you know" http://www.audioenz.co.nz/forums/archive/index.php/t-5707.html http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=stereophile%2frichard%20 heyser&type= . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15780 Registered: May-04 | . "Speaker designers target certain goals that can be spec'd and then measured as their beginning reference point. Once those goals have been achieved in the testing facility most good designers will then perform listening tests in a variety of known rooms. The goal there is to have real world performance that has its basis in the measurements taken in environments no person would inhabit and with material more complex than sine waves." Take this and compare it to the comments made in this link I had placed earlier in the thread; http://www.stereophile.com/content/revels-kevin-voecks Revel is part of the Harman Group which has as their Director of Benchmarking & Acoustic Research, Harman International one Sean Olive. Olive is a died in the wool objectivist who believes what the numbers tell him first and what he hears second. His "bible" is the double blind listening test and he is, IMO, the very antithesis of Heyser. http://seanolive.blogspot.com/2009_01_01_archive.html http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=sean%20olive&type= . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1288 Registered: Oct-10 | Thanks Jan! I have further questions, but I haven't had a chance to look at all of your links yet (hopefully over the next 2 days). I look at those, then get back to you. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15781 Registered: May-04 | . This Global Axis Integration method consists of a carefully engineered radiation pattern created by front and rear driver arrays. Proprietary circuits form steep 24 dB acoustic crossover slopes at specially selected frequencies without the penalties of induced ringing and excessive phase delay. These slopes are necessary to limit lobbing effects and non-linear off-axis response, and actually enable the consistent phase behavior necessary between drivers. The architecture of the circuitry resembles first and second order filters combined with Zobel conjugate compensators in parallel. By using a minimum of high quality parts in series with the drivers, the sound remains transparent, yet the control over phase and amplitude can be corrected with the paralleled Zobel circuits.; http://www.vonschweikert.com/techspecs/2.php Filter Types The most common filter responses are the Butterworth, Chebyshev, and Bessel types. Many other types are available, but 90% of all applications can be solved with one of these three. Butterworth ensures a flat response in the passband and an adequate rate of rolloff. A good "all rounder," the Butterworth filter is simple to understand and suitable for applications such as audio processing. The Chebyshev gives a much steeper rolloff, but passband ripple makes it unsuitable for audio systems. It is superior for applications in which the passband includes only one frequency of interest (e.g., the derivation of a sine wave from a square wave, by filtering out the harmonics). The Bessel filter gives a constant propagation delay across the input frequency spectrum. Therefore, applying a square wave (consisting of a fundamental and many harmonics) to the input of a Bessel filter yields an output square wave with no overshoot (all the frequencies are delayed by the same amount). Other filters delay the harmonics by different amounts, resulting in an overshoot on the output waveform. One other popular filter, the elliptical type, is a much more complicated filter that will not be discussed in this text. Similar to the Chebyshev response, it has ripple in the passband and severe rolloff at the expense of ripple in the stopband.; http://www.maxim-ic.com/app-notes/index.mvp/id/1795 The Butterworth filter is a type of signal processing filter designed to have as flat a frequency response as possible in the passband so that it is also termed a maximally flat magnitude filter.; http://en.wikipedia.org/wiki/Butterworth_filter a Bessel filter is a type of linear filter with a maximally flat group delay (maximally linear phase response). Bessel filters are often used in audio crossover systems. Analog Bessel filters are characterized by almost constant group delay across the entire passband, thus preserving the wave shape of filtered signals in the passband.; thus preserving the wave shape of filtered signals in the passband.; http://en.wikipedia.org/wiki/Bessel_filter Linkwitz--Riley (L-R) filter is an infinite impulse response filter used in Linkwitz--Riley audio crossovers, named after its inventors Siegfried Linkwitz and Russ Riley, which was originally described in Passive Crossover Networks for Noncoincident Drivers in JAES Volume 26 Number 3 pp. 149-150; March 1978. It is also known as a Butterworth squared filter. An L-R crossover consists of a parallel combination of a low-pass and a high-pass L-R filter. The filters are usually designed by cascading two Butterworth filters, each of which has -3 dB gain at the cut-off frequency. The resulting Linkwitz--Riley filter has a -6 dB gain at the cutoff frequency. This means that summing the low-pass and high-pass outputs, the gain at the crossover frequency will be 0 dB, so the crossover behaves like an all-pass filter, having a flat amplitude response with a smoothly changing phase response. This is the biggest advantage of L-R crossovers compared to Butterworth crossovers, whose summed output has a +3 dB peak around the crossover frequency. Since cascading two nth order Butterworth filters will give a 2nth order Linkwitz--Riley filter, theoretically any 2nth order Linkwitz--Riley crossover can be designed. However, crossovers of higher order than 4th may have less usability due to their increasing peak in group delay around crossover frequency and complexity.; http://en.wikipedia.org/wiki/Linkwitz-Riley_filter Pay attention here to the fact various filter orders and even types can be combined - cascaded/paralleled - to achieve a desired result which might often be able to compensate for specific deficiencies of each other, say, ringing in higher order filters or another type of filter choice all together. And it might be a good idea to return to Elliot for a brief reminder; http://sound.westhost.com/lr-passive.htm http://sound.westhost.com/pcmm.htm http://sound.westhost.com/ptd.htm And, finally, for this post; http://www.clear.rice.edu/elec301/Projects00/elec301/ "The thing we call time in audio measurements and the things we call frequency are different coordinates for describing precisely the same signal."; http://www.stereophile.com/asweseeit/572/index.html . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1289 Registered: Oct-10 | New question popped up. Sounds travel at slightly different speeds according to their frequencies. If you saw a car hopped up with racing cams and cherry bombs and a loud thumping stereo pass by with it's windows open never getting closer than 100 feet away all the sounds coming from this car are going to reach your ears with the highest pitched first and in decending order there after. If you are watching the same thing happen in a movie through time aligned speakers, won't the time alignment take away from the realism? |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15782 Registered: May-04 | . You're getting too esoteric and you're missing a vital part of the equation which was mentioned by Von Schweikert. The concept of high frequencies travelling "faster" than low frequencies is not a real concern in your car example. For practical purposes what we perceive in nature is all frequencies travelling at the same rate with variations in rate of travel being more dependent upon the density of the air or a wide divergence in temperature. A more dense, humid environment will slightly alter the rate of speed vs an extremely dry climate but it will do so more or less equally to all frequencies. But to our ears, in either case, the all the soundwaves arrive intact and "in time". Read the Thiel material again and then read Heyser's Klipschorn review. The K'Horn suffers from dislocation of the acoustic centers of each driver - as do many speaker systems. The high frequency driver is mounted at the throat of a very shallow horn while the midrange driver is set back further away from the listener in a longer, larger horn. Their acoustics centers being disimilar accounts for those sounds which exist in both drivers to be dislocated in time at the listener's ear with the high frequencies travelling at essentially the same rate as the mids but with the high frequencies arriving at the listener's ear first due to the shorter distance those sound waves had to travel. Electrical phase issues could account for either the lower or upper frequencies being ahead or behind the other but for practical purposes assume all frequencies traveling at the same rate of speed. Reversing the phase to the tweeter would affect its time value or, say, using a fourth order filter would place the signal arriving at the tweeter one wavelength behind the signal leaving the woofer - it would be out of time by one wavelength. That is where the designer gets their "time alignment" in the electrical mode. In the acoustic mode the alignment of the acoustic centers or the location along the vertical axis determines which frequencies will arrive at what "time" in regards to the listener or the microphone. And in your example the microphone would be the missing portion of your equation. Let's say high frequencies did actually travel at a noticeably faster rate of speed than do low frequencies. To get the reproduction of the car's sound into the loudspeakers you would first need a microphone to capture the sound. The microphone would record the rate of arrival for all frequencies and that would become the order in which the playback chain would then translate them to the speaker. As Von Schweikert points out the loudspeaker needs to be the acoustic inverse of the microphone, the mirror image of what came into the mic being what comes out of the speaker; ACOUSTIC INVERSE REPLICATION Additional research led to my further discovery that recording microphones encode the musical signal with their overlaying pickup response patterns. After making a series of recordings, using several different microphones, it was obvious during playback that the mics not only had tonal differences related to frequency response errors, but also created different types of imaging patterns. The perception of depth and space was not only dependent on the recording environment and mic placement, but also on the mic's off-axis polar response. For this reason, I decided to engineer an adjustable ambience retrieval system radiating from the rear of the VR speakers, in able to recreate the space and depth heard in the concert hall when the spaced omni method of recording is used. Thus, a correctly designed speaker system should project the inverse of the mic signal, acting as a decoder to translate the original sound field. I have termed my design for this decoding as Inverse Acoustic Replication tm, and the Virtual Reality series of designs was developed from several important concepts related to microphone pick-up patterns. These concepts are based on the consistent phase/frequency relationships in the polar response pattern of the mics, which was later reverse engineered into the VR speaker systems. Unless the recording mic were a "two way" microphone with the sections spaced apart in time, the sound waves would arrive at the mic capsule intact just as they would to our ear if it was placed in the same location. We would recognize the car as being in time. But for all practical purposes in a consumer loudspeaker high frequencies do not travel at a rate substantially faster than do low frequencies in the typical domestic environment, they only arrive earlier at the ear due to dislocation of the drivers' sound source. In playback we perceive the sound of the car as all the frequencies arriving at our ear at the same time but we must correlate the different timing cues which relate to a real world environment in which reflections and out of phase signals also are a large portion of what we perceive. How those timing cues are recorded is to a large extent depenedent upon the pick up pattern of the microphone and our brains will make adjustments according to the information received as to whether the sound is realistic in nature or artificial. Now you're discussing the art of recording and microphone selection and why a cardiod or a uni-directional microphone would have been a poor choice to realistically capture the sound of that car travelling down the street. If the car had been recorded inside an anechoic chamber, then there might be some issues with the time arrivals of the directly radiated soundwaves of the car - not much but possibly some just for the sake of argument. But it wasn't and it won't be any more than a symphony orchestra would perform inside an anechoic chamber. So, for practical purposes of a time aligned loudspeaker, don't worry about the impossible. Assume all frequencies travel at the same rate of speed in a typical environment and electrical phase and the alignment in a vertical array of the launching platforms for all sound waves are the primary goals for phase and time aligned speaker systems. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15783 Registered: May-04 | . See if this helps; Effect of frequency and gas compositionThe medium in which a sound wave is travelling does not always respond adiabatically, and as a result the speed of sound can vary with frequency.[12] The limitations of the concept of speed of sound due to extreme attenuation are also of concern. The attenuation which exists at sea level for high frequencies applies to successively lower frequencies as atmospheric pressure decreases, or as the mean free path increases. For this reason, the concept of speed of sound (except for frequencies approaching zero) progressively loses its range of applicability at high altitudes.:[4] The standard equations for the speed of sound apply with reasonable accuracy only to situations in which the wavelength of the soundwave is considerably longer than the mean free path of molecules in a gas.; http://en.wikipedia.org/wiki/Speed_of_sound#Effect_of_frequency_and_gas_composit ion http://www.physicsforums.com/showthread.php?t=238802 . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1291 Registered: Oct-10 | Between your posts and your links you've put up quite a bit of info and I appreciate that. However, in all fairness some of it is likely to get missed eh? I have read all of your posts twice each and 5 of your links once each which is considerably more than I expected to get to today. The reason I posted the car example is because I am trying to understand the hows and whys of this need for correction. It's abundantly obvious that when listening to recorded sounds, the timing and phase can get out of alignment. Otherwise there would be nothing to correct. So you and the people who wrote the articles are saying this missalignment is recording issue caused by the microphone. If that's the case, if a drummer hits a crash symbol and bass drum at the same time, hearing it live, the sound of two should arrive at the listener's ear at the same time and in phase regardless of distance there from. When it's recorded, the issues of phase and time arise due to the short coming of the mic(s) used correct? So then unless a perfect mic is invented (snowball's chance...), the burden of compensating for these problems will always rest on the shoulders of the people designing playback gear, especially speaker designers, correct? |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15784 Registered: May-04 | . I'm sure some of the material I've linked to will be missed or not fully explored. There's only so much time to devote to these issues and they are in the scheme of things still rather small beans compared to other problems we might encouter. Understanding someone like Heyser is not a function of merely clicking on a few quotes from third party sources. IMO Heyser is one of the most widely acknowledged true genius personalities of the last half century in audio. As Atkinson notes, every minute spent reading Heyser can provide ten minutes or ten years of thinking about what Heyser has stated. I can tell you from my very shallow understanding of the topics Heyser explored that coming back to anything "Heyser" every few months/years is worth the efffort. Most of the links I've supplied are to writers who have far more to say than any single link can provide while a few are one hit wonders who got lucky and struck a nerve. Some people prefer only to have the equipment or only the music be their reason of interest in high end audio while others have a desire for the why's and what not's of the affairs of audio as a holistic experience. And, for many the divide between their interests and those of another might represent an unbridgeable gulf where neither speaks the other's language and each has no interest in what the other has to say at all. http://www.stereophile.com/thinkpieces/165/index3.html " It's abundantly obvious that when listening to recorded sounds, the timing and phase can get out of alignment." OK, I'll ask, at this point what is abundantly obvious to you about this matter? I ask because this, "So you and the people who wrote the articles are saying this missalignment is recording issue caused by the microphone", still seems to miss the larger point of why time and phase are important to the speaker designer and to the listener. "Add to this the fact most modern recordings do not attend to time and phase relationships and you find yourself in a morass of contradiciting ideals once you even scratch the surface of the topic."; https://www.ecoustics.com/cgi-bin/bbs/show.pl?tpc=1&post=1931777#POST1931777 . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15785 Registered: May-04 | . Is there anyone else who is still with this thread? Anyone else with anything to contribute? |
|
Gold Member Username: MagfanUSA Post Number: 2044 Registered: Oct-07 | Jan, there is a daunting amount of information here. Just grazing the links is several hours reading. Do my panels help or hurt the time / phase mess? On the reproduction end of things, my panels have a 1st / 2nd order x-over and the single mylar sheet does all the radiating...though it does have separate sections for HF / LF. Given that recordings are generally not time/phase coherent, does it matter? I took an absolute phase test and scored 50%. IOW, I couldn't tell. at least using my EarBuds. I didn't repeat the test using the computer speakers. Have you ever taken such a test? If not, would you care for the link? Who knows, you may be good at it! |
|
Gold Member Username: MagfanUSA Post Number: 2045 Registered: Oct-07 | It occurred to me that some may be better at hearing this sort of thing than others. To that end: http://www.delosis.com/listening/home.html They play a musical phase...then repeat. identical or NOT? in 2 tests of 30 items, I got back-2-back 21s. Better than chance? Yes. Stellar? nope. How'd you do? |
|
Gold Member Username: SuperjazzyjamesPost Number: 1293 Registered: Oct-10 | I don't know Jan, I think they figured you've got it and you do. The abundantly obvious statement was poor wording on my part. I meant that when I first heard about phase and time correction, whether pertaining to speaker design or wiring, it was immediately obvious TO ME (something I should have made clear) that recorded sounds had issues in these areas. The question I had was: How does this happen? Is it because real world sounds have P & T issues or is it something else. You and the gentlemen you are quoting are saying that short comings or pickup patterns of mics cause this. Okay, then my next step is to read the rest of your links when possible. I was hoping to do so between today and tomorrow, but my sons want me to watch playoff games w/them today and take them hiking tomorrow since all 3 of us have off for MLK. IMO, all businesses should be closed tomorrow. MLK is a hero to anyone who stands for racial harmony because that's what he stood for. So I will read and probably re-read your links as I can. Thanks for all your help in this matter Jan. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1294 Registered: Oct-10 | I agree Leo. In addition speakers, room and position of speakers, I think some people pick up on P & T issues than others. My wife for instance wouldn't notice such issues unless there was a serious problem that stuck out like a sore thumb. Then again, she'd rather listen to music on the $50 boom box than go into the music room and hear it on my system. Most of the people I know just aren't very concerned with sound quality and certainly don't pay attention to phase or timing issues. |
|
Gold Member Username: MagfanUSA Post Number: 2046 Registered: Oct-07 | OOOPS! in my post above, i meant 'phrase', not 'phase'. Please 'take the test' and post results. I'm curious. The 'absolute phase' test was on another site. I'm looking. If I can figure out how to get it to my stereo....so much the better. Ear Buds? I don't know. Computer speakers? Awful. |
|
Silver Member Username: KbearCanada Post Number: 918 Registered: Dec-06 | I still plan to read through this thread, but daunting is right, leo! I just haven't found the time to sit down and work through it yet. I'm hoping the next week will be slower for me and I can do that. However, my questions were probably answered sufficiently with Jan's first few posts. I'll see if I want to get into all of it. Will take the test too leo, but first I have to install a new sound card (my old one is causing lots of static). |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15791 Registered: May-04 | . "Do my panels help or hurt the time / phase mess? On the reproduction end of things, my panels have a 1st / 2nd order x-over and the single mylar sheet does all the radiating ..." According to Stereophile they are neither time nor phase correct; "Despite its flatness, the MG1.6 is not a time-coherent design, at least not on its tweeter axis. The step response on this axis (fig.5) reveals the tweeter to be connected in inverted acoustic polarity, and its output leads that of the woofer by a small fraction of a millisecond." http://www.stereophile.com/content/magnepan-magneplanar-mg16qr-loudspeaker-measu rements First, if you've read the material on crossovers at all, you should see that different crossover types will yeild results which are either "transient perfect", "amplitude perfect" or some combination of the two or in a few cases neither. So, you'd need to know more about the type of filter architecture employed than just knowing the filters are first or second order. However, one criticism of designs such as the Magneplanars is their radiation patterns are unlike most conventional dynamic systems as both the low/mid and the mid/high sections function largely as line sources. Measurements will not provide an exact picture of what happens in a typical listening room since on axis measurements are taken at specific distances from the drivers. That distance might not reflect the distance where the two drivers output will sum into a more integrated pattern or, with two lines operating on a wide panel, it's possible they will never fully integrate as one time aligned system since one line source will by necessity always be a slightly different distance away from the listener. The issue becomes even more complicated by the radiation pattern of the panels since there will be a very large amount of rear wave output that is within the room at the listening position and part of the set up for panels is a decision made to allow some information to be more out of phase at the listening chair while other information is more in phase. You might want to concentrate on the Von Schweikert information as most of his designs employ a rear facing driver which provides an in-room radiation pattern more similar to your panels than would a Thiel. As far as phase alignment is concerned due to the radiation pattern of the system I doubt tremendous effort was put into such concerns but I really don't know. Panels overall have oftentimes seemed to be systems where measurements and theory finally give way to what simply sounds best. . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1297 Registered: Oct-10 | One question comes to mind here Jan. I know you won't all of the specifics of Leo's room accoustics without paying him a visit, but what kind of affect do you think his listening room has being that it's octogon shaped with these particular speakers as opposed to the same speakers in a square or rectangular room? I'll understand if this difficult to answer. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15792 Registered: May-04 | . " took an absolute phase test and scored 50%. IOW, I couldn't tell. at least using my EarBuds. I didn't repeat the test using the computer speakers. Have you ever taken such a test? If not, would you care for the link? Who knows, you may be good at it!" http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=audio%2fabsolute%20phase &type= "Absolute phase" and "absolute polarity" are slightly different things which tend toward being interchanged in many conversations. In a single instrument recording they can generally be considered identical but once more than one microphone is in use (not at all uncommon on even a single instrument such as guitar, violin, piano, etc.) there are no guarantees which say absolute phase has been maintained even though absolute polarity might still exist. Both terms, however, tend towards "controversy" between those with "golden ears" and those "rational beings" who believe the tooth fairy comes to visit all the pitiful "audiophools" who buy into such crap each and every night. Certainly in most modern recordings absolute polarity has little to no meaning for the vast majority of recordings you'll encounter. Go back to those mono 78's where everyone was in the room and playing together into one single mic and that's where you'll find absolute polarity. Today with a dozen mics on a single drumset and 128+ channels of single track recordings the engineers don't concern themself with the niggling worries of a very small percentage of the music buying public when only an even smaller percentage of such claim to be sensitive to affairs of polarity and phase. I took the test, leo, the first time I've experimented with absolute polarity in years. My question would be, how did you find out the tests referred to absolute phase? I didn't see any indication of what I was judging other than an indication the two samples might be different or the same. I would prefer not to know which quality I'm listening for - and I still do as I also saw no indication after taking the test what exactly I was judging other than "same/different". The test does IMO demonstrate one of the classic flaws in testing audio in that I was told there "might" be a difference between the two samples. And while the samples didn't actually represent "music" in a conventional manner they did have musical values. Once I was told I should be aware of the possible differences between the two samples an adjustment of just how I perceived the music was made and I was listening explicitly for any possible difference. That certainly isn't how I listen to music for enjoyment. As most astute and experienced listeners will point out when double blind or ABX testing comes up as a challenge to how well a subjectivist can detect changes in, say, the direction of a cable, what is happening is exactly what leo asked - "How'd you do?" The test is then a thing that challenges the listener's hearing acuity under stressful or uncommon conditions and tends not to test how the listener actually listens to music. Tremendous amounts of material has been written on audio DBT's and ABX's and it can be a semi-interesting topic to explore on a rainy afternoon when you're already wearing your waders. I was somewhat pressed for time when I listened to the samples and went from one sample to the next rapidly and without pause. By the time I was about half way through the second test I was aware of some fatigue and some confusion and would have preferred to have taken more time with a greater pause between each set of samples. Not an excuse, just an explanation of how I listened. I received a result of 24/30 on the first set and 21/30 on the second. I noted the samples were exceptionally close mic'd which makes differentiation of absolute phase/polarity a bit easier (if that's what the tests were actually all about) and I used headphones which alleviates any phase and room issues in anything other than mono-pole single driver loudspeakers in the extreme nearfield. Two things about what I perceived during the test; first, the samples I labelled as "different" sounded almost as if the bit rate of the recording had changed going from possibly a FLAC to a lower rate MP3 sample. Additionally, I would say the "difference" I experienced and most often latched onto was one that tends towards proving Heyser's concept of sound as being five dimensional and most especially considering the "when" aspect of Heyser's concept. According to Heyser, the real aspects of the concrete framework that supports the two abstractions are at least five-dimensional. Sound has a "where," which covers three dimensions by itself. It has a "tone," which includes pitch and timbre, themselves independent variables. Its intensity, a "how much?" that varies with time, represents dynamics. It has a "when" aspect in that the listener's instantaneous perception of musical values depends very much on what has gone before. And which of these aspects is the most important when assessing quality will be different for each listener.; http://www.stereophile.com/asweseeit/398awsi/index.html Depending upon whether the tonal sequence of the notes in each sample was primarily ascending or primarily descending the difference I most often preceived was one of a tonal quality which was not quite right in comparison to the previous sample's order. A trained musician or someone with good ear training in musical values might identify such differences when ascending/descending in a scale and hitting a flatted (minor) third rather than a natural (major) third*. (Most of us just hear it as the foundation of rock music.) The differences I detected were not as severe as a half step tonal difference but more as if the tone had been "bent" by about 1/8 step slightly away from it's "correct" tonal quality in the ajoining sample. Often, when absolute phase/polarity is bothersome to a listener keen to such qualities the music will sound slightly off key or off pace/time and slightly "flat" as if the standard A/440Hz tuning had not been followed. Additionally, in most classical music where multiple instruments are playing simultaneously, the openess and the air in the venue and between musicians will seem to be less noticeable. Of course, the recording will have had to have some attention paid to such matters or have an origin in the days before extreme multi-tracked recordings. If you care to test your system and yourself for acuity to such matters, you need to reverse both speaker cables in one location. So "+" to "-" and "-" to "+" at either the amp or the speaker but not both. The simpler the recording technique the better for these tests. Listen to a few minutes of music and then make the switch and repeat the selection. Unfortunately, this rather tedious method also tests your aural memory. *This is a fairly large difference in most music and the differentiation between major/minor tones is one of the foundations of Western music heard in virtually every rock and roll/blues/jazz performance you can find. The addition of a minor to a major shift in tone value (particularly when played against the backdrop of a major key chord progression) in a guitar solo is a trademark of many musicians and can commonly be found in the playing of B.B. King and Eric Clapton. . |
|
Gold Member Username: MagfanUSA Post Number: 2050 Registered: Oct-07 | MY BAD. The test I just linked was for 2 musical phRases played one after another.....Same or Different? My score was 68% correct which is sufficiently enough above chance to be valid. In 2 sets of 30 trials. Some months ago, I linked an absolute PHASE test, which I failed to distinguish at better than 'chance' rate of 50%. I took this test with earbuds. Maybe I'd score better on my stereo...perhaps not. I'll look again for this site. It had other tests of audio acuity which may be of interest. This is much easier and faster than swapping speaker wires around. My pre amp has an absolute phase switch which makes a difference on only a few recordings...and than of solo instruments only. I leave it alone. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15793 Registered: May-04 | . Well, that makes more sense and less sense. I did hear the tonal values shift - sometimes rather obviously more than a 1/8th step - but I was still aware of what sounded like bit rate changes. Was I just hearing what I thought might occur with a phase change or was there more to it? This would seem to be a constant issue with audio tests of any sort, once you plant an idea the listener might detect some sort of difference the test subject begins to listen for differences above all else. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15794 Registered: May-04 | . "You and the gentlemen you are quoting are saying that short comings or pickup patterns of mics cause this." That's not exactly what's being said. Von Schweikert has made mention of the polar patterns of various microphones. Thiel and Vandersteen to my knowledge have not. Let's keep in mind then that Von Schweikert is selling his concepts along with his speakers over someone else's concepts and someone else's speakers and a good story of how the designer came to see the light of audio Nirvana always helps in that regard. Microphones are chosen by a recording engineer for a raft of reasons and a major studio engineer is unlikely to use the same microphones in the same manner as a low production quantity "audiophile" recording engineer might. This tremendous variability of recording techniques makes the task of designing any component all the more difficult as the final component can only be judged a success when it replicates as closely as possible what is being fed into it. For this reason alone many audio manufacturers have become interested in making their own recordings to use as "known" references. This allows them to put into the recording those same qualities they are trying to extract from the recording. That those audiophile qualities seldom exist in major label productions and why they do not begins with microphone selection and set up and that is really what Von Schweikert and many other designers are saying here. You cannot get something out of a recording if it was not first intended to be placed on the recording. The issues of time and phase, however, are those that exist in the speaker system. Those are the constants the designer can control. "One question comes to mind here Jan. I know you won't all of the specifics of Leo's room accoustics without paying him a visit, but what kind of affect do you think his listening room has being that it's octogon shaped with these particular speakers as opposed to the same speakers in a square or rectangular room? I'll understand if this difficult to answer" I have no answer for that. Squared and circular rooms are the worst starting points for good sound reproduction. The further away from either of those two you can get the better the sound quality should become. But there's far more to it than general shape. The point of the thread I would say is the time and phase characteristics within the speaker system remain the same no matter the room. The designer controls what they can and hopes for the best with all the rest. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15795 Registered: May-04 | . http://www.youtube.com/isomike#p/u/2/P7e-WRojZDM http://www.soundfountain.com/amb/mercury.html http://www.humbuckermusic.com/acguitrectec.html . |
|
Gold Member Username: MagfanUSA Post Number: 2051 Registered: Oct-07 | Jan, When I took the 'phrase' test, I ignored what I thought were digital or transmission (latency.....etc) artifacts and concentrated on the 'tune' and the rhythm. A few obvious 'glitches' caught my ear. So, How'd 'ya do? I wish others here, especially the musicians, would test themselves this way. I'd expect good results. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1300 Registered: Oct-10 | "That's not exactly what's being said" Am I at least in the ballpark? Afterall, if a speaker is to be the inverse of a microphone, then isn't the speaker designer, Von Swelkert at least trying to correct what happens when sound enters the mic? Not that I've closed my mind to the possibilty that there are other issues than just those of the mics, but that seems to be the heart of phase and time issues. In other words, if a mic picked up sound the same way the average ear does wouldn't the phase and time issues be greatly diminished? As for the question regarding Leo's room, that's about as good an answer, in fact a little better than what I expected. I simply asked in case you could answer it. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15797 Registered: May-04 | . "Am I at least in the ballpark? Afterall, if a speaker is to be the inverse of a microphone, then isn't the speaker designer, Von Swelkert at least trying to correct what happens when sound enters the mic?" Well, no, you're not in the ballpark with that assumption. IMO the overall goal of the high end industry, if you were to state it as a mission statement or something everyone strives toward as a whole given the divergent ideas each member represents, would be stated as simply as "first, do no harm". The overriding concepts of transparency, accuracy, neutrality or what have you speak to the realization there should be no component which attempts to "correct" for what is being put into its inputs. The age old adage of a "straight wire with gain" has guided most designers with a certain sense of wisdom for over a half century. The concept of system building which places the source as primarily responsible for the quality of everything that follows is based upon the "garbage in = garbage out" adage from the computer sciences. There is nothing a component can do to retrieve information which has been lost in an earlier stage or to correct for what will come next without adversly affecting those signals which do not require identical correction or replacement. If a designer makes any attemnpt to correct for, say, the polar pattern of a cardioid microphone, how would that same system accurately respond to the polar pattern of an omni mic? If the designer assumes all recordings will be made with a crossed or conicident pair, how would the system respond to a Decca Tree with any degree of accuracy? http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=understanding%20micropho ne%20patterns&type= http://www.audiomasterclass.com/?p=archive&a=stereo-microphone-technique-the-coi ncident-crossed-pair Go back to Von Schweikert's statements regarding microphones and read more about microphone patterns and in-room radiation patterns for loudspeakers and pay attention to which pattern he selected to be his "ideal". Then, IMO, it would be a good idea to find out a bit more about that pattern and why a recording engineer might select that or another pattern. Certainly what Von Schweikert has chosen for his "inverse" source is a microphone pattern that does not reflect how the vast majority of non-classical recordings are being made and which actually has little to do with how many classical recordings are captured when cost considerations continually creep upward. Go back to the earlier links to the various microphone patterns and pay attention to the recordings made for the Mercury Living Presence series in the 1950-60's with Cozart Fine. Then you'll need to sort through the marketing aspect of Von Scweikert's stated concepts to find what is a useful take away from the technical information he has presented. IMO most technical papers presented by manufacturers have quite a sufficient amount of useful information contained in them to make their story plausible. But like most partisan papers at times the truth of the matter is not always made completely clear to the reader in hopes the reader will simply accept what is being said as a high minded reflection of what they had always themself thought but could not clearly state. Even when the story is factually correct and fairly presented there is always the chance someone with equally good facts but a different outlook on the topic will disagree and present their own conclusions as a counter. The job of the informed consumer, if that's what you wish to be, is knowing enough to have a good idea when the manufacturer is presenting useful information and when the manufacturer is simply telling you why you should buy their product instead of someone else's product. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15798 Registered: May-04 | . "In other words, if a mic picked up sound the same way the average ear does wouldn't the phase and time issues be greatly diminished?" But that's impossible, there is no microphone which displays an identical polar pattern to the human ear. Even if there were such a device, what we "hear" is not what we "perceive". A microphone lacks a brain. This takes us into the area of the cognitive sciences which are a vast mystery in many ways as no one can prove or disprove what another person perceives. At the extremes of this thinking would be the person who hears voices no one else can distinguish. How do you decide that one person is wrong and not the entirity of those who do not hear those voices? Closer to accepted "normalcy" is tinnitus - a ringing in the ears. It has only recently been considered a real malady by the medical feild and yet reports of similar strange noises perceived by reputabe individuals goes back centuries. It's impossible to prove tinnitus even exists, more importantly it has generally been accepted there is no clear explanation for what those who suffer from tinnitus actually suffer from or respond to. If you're unaware of "the Cocktail Party Effect"; http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=the%20cocktail%20party%2 0effect%20refers%20to%20the%20tendency%20for%20people%20to&type=, do some reading and you'll quickly realize the vast differences between what a microphone will pick up and what the human ear/brain connection perceives. Finally, if you've heard a recording made of a simple conversation within a room, you might have noticed the resulting sound played back through a speaker or even headphones doesn't really resemble what we would have "heard" had we been present at the recording. Depending upon the mic pattern chosen the recording is likely to have included the reflected sounds of the room in which the recording took place. Play those back through a speaker in even the same room and you are overlaying those recorded sounds with the sound of the room where the recording is played back. Listen to the recording through headphones to eliminate this overlap and you'll hear the room sounds of the recording quite distinctly. Had you been present during the recording there's a good chance you wouldn't have been aware of those reflections in the same way the microphone captured them - or, at least, you would not have perceived them in the same manner since your brain would have discriminated the reflected sounds through the filters we use everday to make sense of the world and you wouldn't have noticed those reflections existed until you heard them captured on the recording. Strangely though, what you could have done without thought at the time of the recoring is now impossible to process while listening to the recording. If you have a recording device, record the sound of your speakers playing back a recording you feel "sounds good". Pay attention to what the mic captures that you do not perceive in the same detail when listening through your speakers. When you hear or read a discussion of "psycho-acoustics", this is largely where it all begins. Once again, go back and read Von Schweikert's thoughts on how his wife perceived the sounds she heard in another room from where the speaker sat. This is important to stuff you should begin to grasp about sound reproduction and even more so as it relates to speaker design. . |
|
Gold Member Username: MagfanUSA Post Number: 2054 Registered: Oct-07 | Jan, Could you please address a technique I've seen where an 'artificial head' is used with microphones in place of ears? And another technique I've seen is the old Carver Sonic Hologram. I don't know if I can explain that one. Seems that sound intended for one ear has a component on the other side which you can 'subtract' so each ear gets only what is intended for it. I liken it to how sound canceling headphones work, by introducing a signal opposite in polarity to ambient noise, only by L/R channel. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15799 Registered: May-04 | . "Could you please address a technique I've seen where an 'artificial head' is used with microphones in place of ears?" What do you want me to say, leo? . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15800 Registered: May-04 | . To begin here, leo, what you're referring to are "binaural recordings" made using two omni-directional microphones placed at a position which would represent the location, and most often within the shape of a human ear, as it sits on a dummy head. The intention is to capture the "head related transfer function" which exists at the heart of all human perception. The ancestral imprinting on our brain is to perceive location of sound by processing the time, phase and amplitude differences between what is perceived in each ear. Due to the specific shape of the human ear and the difference in distance between each ear humans detect the micro differences in those three qualities as they arrive at each ear and process the information to determine a location and a distance for the source of the sound. This informs the brain and begins the process of assigning the fight or flight reaction to either finding food or being food. http://en.wikipedia.org/wiki/Binaural_recording http://en.wikipedia.org/wiki/File:BinauralPaper.ogg http://en.wikipedia.org/wiki/Dummy_head_recording http://www.virtualbarber.org/page.php?3 Most of the information you'll need to better understand binaural recording and playback will be found in those links along with a few expamples of such recordings. It's important to notice that binaural recordings can, first of all, only be played back through headphones (and it's better to have phones designed specifically for the binaural process) if you want to experience what the process is all about. Since the intent of binaural recording is to take into account the time, phase and amplitude differences between the two ear positions on an average shaped head dummy, playing back the recordings through loudspeakers not only provides no "binaural" effect but actually makes the recordings sound oftentimes less impressive than a well produced traditional stereo recording. I've only once heard a real demonstration of a binaural recording played back through headphones designed for binaural playback and that was back in the mid '70's. The spatial immersion possible with a good system of binuaral recording/playback is impressive in many ways but ultimately more as an experiment in the psychoacoustics of human perception. Listening to the recordings provided above will give a good enough example of what to expect from a binaural recording. Possibly you'll have a different experience than I do but I never forget I'm still wearing headphones and listening to a recording. More traditional microphone techniques have followed the overall scheme of the binaural process. Engineers commonly use either omni-directional or cardioid patterned (heart shaped) microphone capsules spaced roughly at a distance similar to or greater than those used in binaural recording depending on the event being recorded and the amount of stereo separation desired. A spacing similar to the head-width difference of binuaral recordings is made with the ORTF technique; http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=ortf%20micing&type= Baffles between the two mic capsules can be added to enhance the separation effect and this is at the heart of the Kimber "Iso-Mic" procees I've linked to above. The Iso-Mic process is best played back through an Ambisonic decoder which interprets the the data leading to a long overlooked method of creating highly realistic spatial information in a two channel system; http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=ambisonic%20recording&ty pe= Experiment with the various microphone placement techniques in virtual hardware on a few sites; http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=spaced%20omni%20micropho ne%20technique&type= Here's a good explanation of spaced vs crossed or coincident capsules operation and the benefits/disadvantages of each; http://www.prosoundweb.com/article/print/better_recording_microphone_techniques_ to_produce_warm_spacious_stereo Unfortunately, it appears the audio examples have been deleted from use in that article. I would encourage anyone to do some legwork - or thumbwork depending on your choice - and follow a few links to learn more about how recordings are made at the very source as this should provide a better understanding of what you are hearing through your system. Here's an interesting comparison of how to mic a drumset; http://recordinghacks.com/2010/04/03/drum-overhead-microphone-technique-comparis on/ If you'll also read the text to that article you will notice a reference to "mono compatibility" of various techniques as some stereo mic positions will leave a large hole in the middle of the mix due to time/phase cancellations between capsules when they are mixed down to a mono signal. Take that and compare the effect to what might happen with a time and phase speaker system as the "inverse" to a microphone's polar pattern. You'll also notice the author mentions "wet" and "dry" mixes which generally refers to how tightly focussed is the soundfield. A very dry mix is focussed almost exclusively on the singular sounds of the mic's placement while a more wet mix is going to have more ambience and a higher level of room sound in the final mix. These are all considerations which begin the recording process and influence what the listener will perceive. Do a bit more work and you'll see that in most modern pop recordings the overhead mics placed to capture the overall sound of the drumset are often only the beginning of the mic'ing technique and additional microphones placed within the kick drum, under the crash cymbal, just to the side of the snares, etc. will be added which will allow the engineers to adjust the mix for any particular focus or spatial size of the soundfield they desire. Then, in reference to time/phase aligned loudspeakers, consider the effect of a single drumset being recorded in real time by a dozen microphones placed around, in and above the set with each mic capturing not only the sound of, say, a single snare or the above the set sound of the overall drumset but also inevitably also picking up some of the sound from the other sections of the set and the room and then all of that being folded into the final mix with slightly out of phase/out of time information from each mic capsule. To wrap up this section of the post, now go back and re-examine the microphone techniques of the Mercury Living Presence recordings. Three omni-directional microphones were spaced across the front of the proscenium arch to capture and entire orchestra in the best of the best of the Mercury recordings. Achieving the perfect mix of instrument and room sounds which represented what the engineers were hearing at the time of the recording took hours of set up and adjustment. Going back to the spaced omni's mentioned earlier, you might remember the fact that a too wide spacing between capsules would result in a hole in the center of the soundfield. To eliminate the effect the Mercury - and later the RCA "Living Stereo" - engineers placed the third channel mic in the center and to accommodate the more popular two channel systems of the time this mic's information was folded into the stereo mix. It was popular at the time of stereo's introduction to use two very widely spaced speakers for left/right and to add a center channel for the third channel fill found on the Mercury/RCA and other early two channel recordings. The format wars between tape and disc found favor with the three channel system while the eventual adaptation of a two channel stereo disc with a 45/45 groove wall made the convenience of the stereo LP the most popular choice and the format wars ended soon after. The Living Presence and Living Stereo recordings have, since their introduction, been hallmarks of high quality recordings against which all others can be compared. To add to their audiophile mystique the early recordings were made in a manner similar to that of the "direct to acetate" cylinders and 78's and without the use of tape splices, edits or overdubs; if a mistake was made in the performance, the recording would be stopped and the entire performance would start again from the beginning with a fresh roll of tape. Long out of print in vinyl, they fetched astronomical prices on the used market until the SACD/DVD-Audio reintroduction of the recordings in the digital format. Intended to be played in multi-channel mode the new discs for the first time allowed most listeners to actually hear the recordings in the three channel format in which the engineers originally intended. Whether you have the ability to playback the discs in the SACD/DVD-Audio formats or not, if you want to get a glimpse into how excellent stereo recordings began and to have a concept of how all that followed either adhered or devolved from that ideal, listen to any of the early Mercury or RCA recordings. Do some research first into which discs are which as the three mic/direct to tape process was eventually abandoned in favor of less time/money consuming methods. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15801 Registered: May-04 | . "Carver preamplifiers had Sonic Holography, a fancy handle for interaural crosstalk cancellation; http://www.audio-ideas.com/interview/carver.html http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=sonic%20hologrpahy&type= . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15802 Registered: May-04 | . When you go searching for "binaural recordings" be careful what you are finding. The binaural recording process is one thing and one thing only. However, out of the research into the psychoacoustics of the sound/brain connection there exist what are called "binaural beats". Not to go into extreme detail here but the concept is to use the two ears as two discrete conduits to the workings of the brain as it is divided into hemispheres. By playing, for example, two beats with each assigned to one ear using headphones a third phantom beat will be perceived by the brain. So, playing 1400Hz in the right ear and 1407Hz in the left ear will provide a phantom singal of 7Hz as perceived by the brain. Why is this useful? Once again to keep things brief for now, the brainwave frequency at which humans operate determines their level of awareness/consciousness. In a Theta brainwave state there is a hyper-focus on what we perceive while also placing the brain into a very relaxed state; http://www.formulaformiracles.net/brain-waves.html You might want to consider this as being similar in function to that point where you've listened to music for awhile and possibly imbibed in a few other ... "activities" and you realize the music sounds far better than it did at the beginning of the evening. In many cases you'll soon drift off to sleep as your brainwave frequency continues to drop. http://search.yahoo.com/search?p=binaural+beats+meditation&fr2=s b-top&fr=slv8-hptb5 Binaural beats have been used in "new age" meditation and other fields and have become popular in areas of use outside of the traditional meditative process for their proven effects on how we perceive what is around and within us. One of the more controversial products to come out of this research is the Schumann Resonance device first marketed in audio circles by Acoustic Revive; http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=acoustic%20revive%20rr%2 077&type= The essential concept here for the audiophile is to introduce a signal into the listening room which shifts your brain "entrainment" into a Theta frequency which allows for a fully awake version of those moments just before you drift off thinking how good your system sounds. This is a further development of the binaural techniques used with headphones and extends the concept to Isochronic tones which do not require headphones. While appearing a bit touchy-feely for many, I would encourage you to take advantage of the many free downloads of binarual and Isochronic samples available on the web. By using a recording to possibly reduce a headache or to bring you into a restful state, you might get a better idea of where this discussion has drifted and how the field of psychoacoustics has developed. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15805 Registered: May-04 | . More on tinnitus and the brain's sound receiving/processing network; http://discovermagazine.com/2010/oct/26-ringing-in-the-ears-goes-much-deeper |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15806 Registered: May-04 | . http://www.delosis.com/listening/home.html OK, leo, I hope you don't tell me this was a test for some other "difference" than whether the two musical phrases are identical or not. Just following your link I still see nothing to tell me what I'm listening for. Which IMO brings up an interesting point of auditory testing, the test subject tends to listen for qualities they have been told they should listen for. This time I paid no attention to any similarities in phase and had none of the fatigue I experienced earlier when that was what I was focussing my attention on. This time I was unaware of anything that could have related to an in phase/out of phase situation as my focus had shifted strictly to musical/tonal values. Was I only hearing what I thought I should be hearing in the earlier test? Probably. How does that influence a test for audible qualities if the mere suggestion some phrases can be in phase/out of phase (when they probably have not been altered in such a manner) draws the listener's attention to just those qualities they assume will be influenced by in phase/out of phase? It was a totally blind test in that I had no idea whether any of the phrases had reversed phase/polarity. Yet, I heard mostly what I thought I should hear. I did detect shifts in the musical patterns and the tonal values which confused me during the first test - why would they change the order of notes or the timing of a phrase if that was not the focus of the test? - but they were not what I was "supposed" to be aware according to the instructions I was provided. What then does that say about such double blind tests and the validity of their results? If you give the test subject any idea what they should be aware of, are you not planting in their head the idea they should hear a difference or, depending upon how they feel about such tests and the audibility of matters such as cable direction for example, that they should detect no difference between the examples? If you provide nothing more than the suggestion "there might be a difference", then isn't the subject open to too many variables and you are only testing the listener for how well they might guess at what is different? How do such tests then relate to the manner in which the listener generally listens to and responds to music when they are not in a test situation? It's an interesting concept that comes up often in audio discussions as most of the objectivists who rely strictly on test bench measurements and what they think should be true when interpreting those data - those who generally refer to subjective listeners as "audiophools" - believe DBT's and ABX's will answer any and all questions concerning issues of things such as cable direction or even whether cables themself can make any difference in the quality of the sound (not the music, mind you, just the sound). So, given the results of the two very different tests I have taken with the exact same test material, what does that say about DBT's and ABX's and those who believe in their absolute value? Listening only for whether the two phrases were identical or dissimilar in musical values I scored a 27/30 on both tests and IMO had the ability to point to which phrases I probably missed. 27/30 represents the tests' statistical peak (28/30, 29/30 and 30/30 fell rapidly in scoring) and is just slightly higher than their individual averages of roughly 25/30. http://www.delosis.com/listening/summary.html http://www.stereophile.com/asweseeit/487awsi/index.html . |
|
Gold Member Username: MagfanUSA Post Number: 2056 Registered: Oct-07 | Thanks Jan, and Well Done. I suspect your years of experienced listening paid dividends. Testing? I think it can be important, but maybe not for regular reasons. Testing may, under 'lab' conditions help determine absolute limits. Stuff you'd never think was important, like the cockpit voice in fighter aircraft is female. Male pilots pay better attention to a female voice. Other phenom can be isolated in a lab setting as well. Can you hear absolute phase differences? What are frequency limits of people? (last would be statistical in nature) Do sounds or frequencies beyond the upper limit contribute to musical enjoyment, even if beyond what you can 'hear'? All this phase / timing / perceptual stuff is difficult to study when 'lumped' together in a single test. Later, when some data is taken for single phenom, than combinations can be tested. Are people more sensitive to hi frequency or lo frequency phase differences? Is the ear more sensitive to time of arrival differences at hi or lo frequencies. You could come up with a couple dozen of these kinds of tests or questions in just a few minutes. Depending on how the information was presented and to what level of detail and if it could be distilled into some easily applied principles, no telling who could take advantage of it. Better rooms? Better speakers? Something un anticipated? A better version of 'Sonic Holocaust'? The cool part about research...and the 'basic' variety is my favorite, is that you never quite know where it'll lead. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15810 Registered: May-04 | . "Are people more sensitive to hi frequency or lo frequency phase differences? Is the ear more sensitive to time of arrival differences at hi or lo frequencies." Due to the length of the soundwave, low to mid frequency phase and time issues are more noticeable by even the "untrained" ear. I believe the less time spent on Sonic Holography the better. If forced into making a poor choice, I'd prefer to be subjugated into living with a vintage bucket brigade "digital" delay system. "The cool part about research...and the 'basic' variety is my favorite, is that you never quite know where it'll lead." Henry Kloss once said something to the effect of, "Of course it's research. If we knew what we had found, it wouldn't be called research any longer." Kloss was a student of Edgar Vilchur's at MIT before they founded Acoustic Research together in the 1950's and then Kloss proceded to originate KHL, Advent, Kloss NovaBeam and Cambridge Soundworks and spin off a dozen other companies from those where he had primary responsibility; Boston Acoustics, EPI, the original Genesis speaker company, Tivoli, etc. They were the essence of "East Coast Sound" in the age of JBL, Altec and Bozak. He worked with Roy Allison and he gave Tomlison Holman (THX anyone?) his initial big break in audio when TH designed the phono pre amp for the Advent 300 receiver. Between the two men their list of accomplishments are legendary and not the type found in the average "researcher's" resume. IMO there is a vast distinction between those who prefer to use "data" as a path to new thinking and those who use "data" as a weapon with which to cudgle anyone with a dissenting opinion. I know this thread is extremely dense with material to research but those interested in audio and its foundational ideas could do no worse than to spend a little time with Kloss and Vilchur; http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=hnery%20kloss%2f%20of%20 course%20its%20research&type= http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=edgar%20vilchur&type= If you don't find something interesting in any decent article on either man, you probably don't belong in this hobby. . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1307 Registered: Oct-10 | Maybe THIS weekend, I'll get a chance to read everything...... |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15814 Registered: May-04 | . Not to single you out, james, but reading is only half the story. Understanding is the difficult part. I've been at this hobby for over four decades, worked in the industry for twenty five years, and there's still just a surface wound to what I have scratched away at regarding all of this. . |
|
Gold Member Username: SuperjazzyjamesPost Number: 1309 Registered: Oct-10 | "....reading is only half the story, understanding is the difficult part " Agreed! That's why I read everything as many times as possible, look up words I've never seen and study visual aids when ever possible. I can't help but wonder when the people studying and writing it all for the rest of us eat, sleep, shower, use the bathroom, etc. |
|
Gold Member Username: John_aLondonU.K. Post Number: 4837 Registered: Dec-03 | I've just caught up with this interesting thread. Could I just go back to the first post? "The fact that higher frequencies travel faster in itself is a logical reason to move the tweeter further back." I disagree. The tweeter reproduces high frequencies. If you move the tweeter back, you delay arrival of these frequencies at the listener's ears - agreed. However, if high frequencies do indeed travel faster, then they must do so when they are components of the original sound just as much as when they are components of the reproduced sound. So moving the tweeter back will introduce an artefact - an unnatural delay in perception of the high frequencies reproduced by the tweeter. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1313 Registered: Oct-10 | Actually John, if you read what Jan told me when I asked him, sounds in the real world travel at about the same speed. Read all of Jan's posts and links, my questions, his answers, etc. more than once each and you should start gaining understanding a little at a time. That's what I'm doing right now. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15818 Registered: May-04 | . John, sound travels at the same rate of speed for all frequencies when we are discussing a consumer loudspeaker. The variables for any differentiation between rate of speed for various frequencies will not enter into the picture in a domestic system. You are correct, however, as I pointed out in this post; https://www.ecoustics.com/cgi-bin/bbs/show.pl?tpc=1&post=1933486#POST1933486 that should high frequencies travel at a faster rate of speed, the speaker must still be an inverse of the microphone's pick up. The "logical" reason for moving a driver fore or aft is to vertically align the acoustic centers - the point of origin for all sounds - of each driver. Doing so provides one sort of compensation which takes into account all frequencies travelling at the same speed. Go back and read Heyser's review of the K'Horn for a better idea of why this might be important. Doing this vertical alignment ensures all signals from the speaker system will arrive at the listener's ears in a more "time aligned" fashion than with any other physical manipulation of the signal. Time and phase coherence in a multi-way speaker system in the absolute exist as impossibilities. Driver location and filter manipulation can lessen the difference to a minimum as with the coaxial drives in the KEF and Thiel systems. Look at the "step response" of such systems in a good review to see the near miss of such theories. How the individual outputs of the various drivers in a multi-way system sum at certain distances from the driver will be a consideration in building a time/phase coherent system. However, IMO, the only way to get close to actual alignment of all frequencies is with a single driver, full range system. At this time, however, SDFR's still have too many trade offs for most listeners who require "realistic" listening levels to complex material. In general, they kinda suck at Mahler. If Mahler at 110dB is not your cup o'tea, then possibly a SDFR is a good choice from which to begin any search for time/phase aligned speaker systems. I'm not sure how this idea that high frequencies travel "faster" than low frequencies became a common myth in audio. I think this all originated in cable marketing which asserted (at about the same time that time and phase became popular issues in loudspeakers) "skin effect" should be a consideration in audiophile cable design. Skin effect then became a point of disagreement as no physics textbook will indicate skin effect has any relation to how electrons are transmitted through a conductor until the frequency reaches the supersonic range. Trying to make their case for their cables designers began to construct all sorts of theories which were speculative in nature and then the selling and buying public took a misunderstood proposition and turned it into something it is not. With litle hestitation the idea of "time" became an entirely misunderstood buzzword by the retail/buying public and its "alignment" was projected onto everything in equal doses whether a cable is similar to a speaker system or not. I don't know for sure if this is the case but I do think "time" has become one of the most widely abused principles in audio over the last few decades. Maybe we should all step back and pay attention to Peter Walker's, Lincoln Walsh's and the original Lowther/Voight system, eh? . |
|
Gold Member Username: John_aLondonU.K. Post Number: 4838 Registered: Dec-03 | "...sounds in the real world travel at about the same speed" Yes, they do, James. Thanks. I agree, and also defer to Jan every time on all things audio. So the problem in Dan's post, which I quoted, disappears anyway. Even if it didn't, moving the tweeter back would not solve it. That's my point. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1314 Registered: Oct-10 | John, if really defer to Jan, then read every question I asked him and his answers and links, again, more than once each and you will see that he answered your question as well as is humanly possible. You're just asking him what I already did. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15822 Registered: May-04 | . Nothing to worry about, james. It's a complicated subject that may require asking questions more than once. As you can see by his registration date, John's a long time forum member who hasn't been around lately. He's a good guy with lots of information and experience to share and, more importantly, he's a friend. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1315 Registered: Oct-10 | Ok, nuff said |
|
Gold Member Username: John_aLondonU.K. Post Number: 4839 Registered: Dec-03 | Thanks, Jan. Thanks, James. Apologies, Jan, for not reading you post timed 12.57 until now. My post of 12.58 was written in reply to James's post of 12.21. And thanks also, Jan, for the link back to your earlier post. Indeed my point is covered already in your post and in Von Schweikert's "Inverse Acoustic Replication tm" and I did not see that. I do, now! This is an interesting topic. Behind "Inverse Acoustic Replication" is the assumption that our aim is to reproduce the original sound. I think we agreed on this, though I recall that others don't. You will remember you started what turned out to be a contentious thread "Do you listen?". I often think it is good starting point to imagine a real sound source in place of the speakers, and then to ask how well is the whole system doing in (re)creating that real sound. If we think of it this way, "correcting" for things inherent in both the original sound and the reproduced sound makes no sense. Maybe that's just saying the same thing in different words. But the principle applies to a lot of things in audio. For example, I still don't get it when I see equalizer pre-sets labelled, say, "Jazz", "Rock" or "Classical". I've always thought time and phase coherence are high priority. So I latched onto this thread. I didn't intend to take it off track! You've all written a lot, and provided good, clear questions and answers. You did right to respond, James. All the best. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1316 Registered: Oct-10 | Thanks John! I didn't want to see Jan repost everything (a lot of work) or you to feel dumb for having missed it. Equalizer pre-sets are just plain silly. Most of us here don't have much use for equalizers to begin with. The only use I've ever had for any type of tone control at all is to reduce certain portions of the spectrum when moving the speakers to another spot doesn't help, is not an option or introduces another problem. |
|
Gold Member Username: John_aLondonU.K. Post Number: 4846 Registered: Dec-03 | Thanks, James! I agree on equalizers. But we are in a minority! Just for fun, here's the equalizer preset for "Piano" in iTunes.  What is it supposed to be about a piano that makes microphones dip out at 4 kHz, so playback has to introduce gain there, in order to compensate? And there's a whole slew of these presets. Beats me what people are thinking when the choose any of 'em. Here's my preferred pre-set.  Off-topic, I know. But it strikes me as the same problem as moving back tweeters, even if sound travels at different speeds at different frequencies. Which it doesn't. All the best! |
|
Gold Member Username: SuperjazzyjamesPost Number: 1317 Registered: Oct-10 | They make me laugh with that. Pre-sets completely ignore room accoustics, speaker performance and a host of other things that can't be predicted by someone making up pre-sets in a lab somewhere. Not that I've ever noticed a dip at 4 KHz with a piano, but what if you're listening to jazz with a piano as the lead instrument? Do you put it on piano or jazz? A lot of such EQs will add as much as 6 (db?), well, 6 of something to the lowest band, usually 31 Hz. Good way get an amp clipping and blow some speakers! Gee thanks Mr. EQ designer! Now you can buy me a pair of B&W 802s! |
|
Gold Member Username: MagfanUSA Post Number: 2067 Registered: Oct-07 | I had an equalizer and 'calibrated' microphone a LONG time ago. It was not a 10-band or any of that nonsense. I bought it specifically to help the bass of my ported 3-ways. It had 5 bands only. 30hz, 60hz, 120hz and 250hz in the bass and 10khz treble. I used it ONLY to reduce the bass port bump and increase the bass at 30hz. At that, it worked fine. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1318 Registered: Oct-10 | To get back to Jan's questio, "Do you have a recording device?......" I do not currently have such a unit that would be suitable for what you are proposing. However, I have made audio and A/V recordings in the past where background music was playing and I have an idea what you are getting at. |
|
Silver Member Username: KbearCanada Post Number: 944 Registered: Dec-06 | Jan, I understand I guess a decent amount of what you've posted, and the links as well (haven't even read half of them). However, there's a decent amount that is just way over my head. To even hope I could get a real good understanding of matters like time and phase alignment, I need a better understanding of the basics. The basics of audio design, and heck, the basics of things like circuits, electricity, and physics. I'm a business major! I didn't take anything like this in school. So with that in mind, are there any really good books that you know of, that can provide a solid foundation for understanding? I can probably find them myself, but in case you've got one or two off the top of your head I would really appreciate it. I do have one question though... That is where the designer gets their "time alignment" in the electrical mode. In the acoustic mode the alignment of the acoustic centers or the location along the vertical axis determines which frequencies will arrive at what "time" in regards to the listener or the microphone. What is there to be said then, for a company as renowned as ProAc (not to mention Reference 3a, Castle, and DeVore) who all chose/choose not to line up the acoustic centers of their drivers along the vertical plane? Is their design wrong, or does the decision to use an offset tweeter have some benefit? I can only guess a benefit may have to do with interactions of the tweeter's soundwaves with the baffle. |
|
Silver Member Username: KbearCanada Post Number: 945 Registered: Dec-06 | I've just caught up with this interesting thread. Could I just go back to the first post? "The fact that higher frequencies travel faster in itself is a logical reason to move the tweeter further back." I disagree. The tweeter reproduces high frequencies. If you move the tweeter back, you delay arrival of these frequencies at the listener's ears - agreed. However, if high frequencies do indeed travel faster, then they must do so when they are components of the original sound just as much as when they are components of the reproduced sound. So moving the tweeter back will introduce an artefact - an unnatural delay in perception of the high frequencies reproduced by the tweeter. Nice catch, John! Even though not an issue in home audio (due to frequencies traveling at basically the same speed), I see the fallacy in my original line of thought. |
|
Gold Member Username: MagfanUSA Post Number: 2078 Registered: Oct-07 | The theory I heard proposed for setting the tweeter and mid back was to LINE UP the voice coils in a vertical line. That was supposed to take care of 'it'. Whatever 'it' was. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15849 Registered: May-04 | . "So with that in mind, are there any really good books that you know of, that can provide a solid foundation for understanding? I can probably find them myself, but in case you've got one or two off the top of your head I would really appreciate it." There aren't many general audience books that are actually written with audio in their purpose. Those that are tend to focus on specific areas such as sound reinforcement, speaker design or microphone techniques, etc. Those are all useful things to know but you'll go through a lot of material to find what's relevant at any stage of your knowledge development. I have general purpose books on electronics and radio that I keep as a reference though there are chapters I've never even reached in all of them. If you have a used book store in your area, browse the shelves. There's always Hartley's book which is a good all 'rounder and written by someone who is involved in high end audio; http://www.amazon.com/Complete-Guide-High-End-Audio/dp/0964084945 Click through the "also bought" section for other sugestions. When I worked as the Master Eletrician/Sound Technician at a professional theatre, this was a good reference for the basic, practical side of many hardware based issues; http://www.amazon.com/Sound-System-Engineering-Third-Davis/dp/0240808304 For speaker building I began with this; http://www.amazon.com/Loudspeaker-Design-Cookbook-Vance-Dickason/dp/1882580109 For digital, this is where it all began for practially everyone; http://www.amazon.com/Principles-Digital-Audio-Ken-Pohlmann/dp/0071348190 On line you can find useful information from these sites; http://sound.westhost.com/index2.html http://lenardaudio.com/sitemap.html http://passdiy.com/projects.htm http://www.decware.com/newsite/homepage.html http://www.vacuumtubes.net/How_Vacuum_Tubes_Work.htm http://www.worldtubeaudio.com/ http://www.bcae1.com/spkrmlti.htm That should give you a start. Follow the links on the various pages and you'll spend the rest of the next five years reading stuff. IMO you have to decide what you need to know and what you don't. At first I was someone who wanted to know everything. Then I later realized it's not important to "know" everything because you never will, it's only important to know where and how to find everything when you need to know it. IMO you begin to know stuff when you begin to realize how little you actually know. As a salesperson I didn't need to know everything in depth, I needed to know an amount where I could carry on an intelligent conversation in most areas. (Want to know about a CrossField PermAlloy record head in a Tandberg TD 20?) YMMV. In practical terms a first year student in electronics knows more useful stuff than I do though they all too often tend to think what they know is always right. That's what too many teachers and too many books do their students, they drive out the curiosity and teach an either/or/my way or the highway methodology. The original "Audio" magazine and the original British magazine "HiFiNews and Record Review" were at one time excellent sources of information. John Crabbe at HFNNR was as good as they get when it came to presenting the practical engineering based viewpoint and then telling you why it was being challenged by the observations of the audiophile. "Audio" is sadly gone (though, if you know what you're looking for, you can still find a few articles on line) and HFNRR has changed so dramatically that it has become laughable. Ken Kessler is an interesting writer in audio and still, to my knowledge, works for HFNRR which is now just "HiFi News" (but not really). You could certainly do worse than to put the names of a few well known designers and theoreticians in audio into a search engine. Start with the above links and follow their links. As I've said before, develop your priorities (in learning too) and don't be afraid to have anything you think you know be challenged by new information. The information you'll find here is spread across divergent views and where you will find one author claiming this, another will often claim that. What is considered to be known about electricity and circuits has been well established for decades but it's not uncommon for someone to challenge the theories with new information as you can see in the posts regarding microphonics in solid state components. Quite often the challenge is meant to sell a product. I'd suggest you spend a little time reading articles that are also just meant to get you thinking about audio, sound, music, perception, etc. as it is often the empirical observation which drives the theory. More than anything else, don't read anything. Go listen to live music. It will make you think and it will make you a better person and it will seldom get in an argument with you over whether you're right or wrong. Learn to play your guitar, you'll be way better off if you can make music on your own than if you can just talk about the stuff that reproduces it. . |
|
Silver Member Username: KbearCanada Post Number: 948 Registered: Dec-06 | Thanks Jan! Will definitely start looking at those suggestions. I'm planning to go to some live shows as well. |
|
Gold Member Username: SuperjazzyjamesPost Number: 1322 Registered: Oct-10 | Don't try to grasp it all at once either. Read sections, look up words you don't know, read the section again. You might never GET all of it, but taking it in parts at a time helps. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15852 Registered: May-04 | . "I do have one question though... That is where the designer gets their "time alignment" in the electrical mode. In the acoustic mode the alignment of the acoustic centers or the location along the vertical axis determines which frequencies will arrive at what "time" in regards to the listener or the microphone. What is there to be said then, for a company as renowned as ProAc (not to mention Reference 3a, Castle, and DeVore) who all chose/choose not to line up the acoustic centers of their drivers along the vertical plane? Is their design wrong, or does the decision to use an offset tweeter have some benefit? I can only guess a benefit may have to do with interactions of the tweeter's soundwaves with the baffle." I think you answered your own question in the op to this thread, "I know there are many skeptics who do not believe these two ideals are important. I believe they cite tests where listeners were not sensitive to phase and time errors, and John Atkinson of Stereophile indicates that poor time and phase alignment has not prevented many speakers from being recommended by his publication. There are many other key aspects to good speaker design, which must be executed properly. If they are not then no measure of great coherence will save a speaker. However, he goes on to state that phase and time coherence does not hurt if a speaker is designed well otherwise." Read the conclusion to Martin Collom's review of the early Wilson Grand Slam; http://www.stereophile.com/content/wilson-audio-specialties-x-1grand-slamm-louds peaker-system-page-8 Then look at the time alignment measurements taken of the speaker system; http://www.stereophile.com/content/wilson-audio-specialties-x-1grand-slamm-louds peaker-system-measurements-part-2 IMO JA was being exceptionally "gracious" to the point of some dishonesty in his technical assessment of the Wilson's performance. Refer now to Fremer's review of the later generation Wilson Maxx2; http://www.stereophile.com/content/wilson-audio-specialties-maxx2-loudspeaker-me asurements Read the review and consider that Fremer, after his reviewing session and who owns a $100,000+ turntable, purchased the review pair as his long term reference. Two different reviewers, both familiar with the sound of live music, but with different priorities, different rooms and system set up and different goals. When reading any review of equipment you first must have a good grasp on the reviewers's priorities. Do they agree with your own? Can you even define what you believe they hold as primary in their listening? If you can and you agree with many of their perceptions, then you'll probably find their reviews to be of use to you. If you can't or you don't agree with the aspects of fidelity a certain listener holds as important, then they are of no real use to you other than to inform you how many buttons and switches are on the unit under review. It's not that the reviewer is being untruthful or improperly doing their job but more likely they are just not speaking the same language you want and need to actually hear. Here's another danger of the internet in audio. You think you want another pre amp so you go searching for reviews of the unit under consideration. All you can find are those reviews written by someone completely unknown to you and who is not clearly expressing opinions about your priorities. But you're reading about ocean's wide soundstages and Grand Canyon like depth paired with Tiffany-ish details. You've learned nothing but you think you've found the review that provides the proof required to invest in some piece of gear you have never even set eyes upon. " I can only guess a benefit may have to do with interactions of the tweeter's soundwaves with the baffle." Go back up and read the descriptions I listed for the Thiel, Vandersteen and Spica speakers' approach to baffle design and the implications they have on advantages/disadvantages to the overall speaker sound. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15853 Registered: May-04 | . "The theory I heard proposed for setting the tweeter and mid back was to LINE UP the voice coils in a vertical line. That was supposed to take care of 'it'. Whatever 'it' was." Where's the voice coil in your ribbon tweeters, leo? If you use the generic term "acoustic center", you are referring to the point on a driver's face from which the soundwaves are launched into free space. Not that the small differences matter much but taking just the voice coil itself into consideration would mean you would have to decide where on the vc you wanted to specify as the "center". Depending on the driver, that could place the acoustic center somewhat out of vertical alignment with another driver(s). On a cone type driver, where the vc connects to the driver is its acoustic center. On a dome type driver, where the vc attaches to the driver is its acoustic center. . |
|
Silver Member Username: KbearCanada Post Number: 953 Registered: Dec-06 | Thanks Jan. I re-read the section you suggested. I notice that the offset tweeter in a Castle or ProAc speaker will, like the Spica, provide different distances before the sound jumps off into free space. Like the Spica, I imagine this will ensure different frequencies where this happens. But how much different? Unlike the Spica, there is no felt material to absorb energy, and the baffle is much more narrow so you won't get the same degree of travel along it. With relatively narrow baffles, the amount of gain will surely be smaller. I notice that some models of Castle and ProAc do not use rounded edges on the front baffle. Some do. It seems this would always be a benefit. Actually, even the ProAc D2 does not employ rounded edges, which is a $3,000 monitor. My Castle Avon does employ rounded edges. I was looking at some of the crossover specs for my speakers. My Tannoy has a woofer and tweeter in dual concentric form. No other drivers exist. Yet Tannoy lists two crossovers in their specs: "1st order HF, passive low loss 2nd order LF" Would it not be one or the other with only one woofer and one tweeter? Another thing I note is the crossover position of my speakers. Tannoy is lowest at 1.8kHz, then Quad at 2.2kHz, and then Castle at 3.3kHz. Not sure what order crossover Quad uses (I could not find this online). Castle employs a 2nd order crossover. It would seem to me that 3.3kHz is a relatively high crossover point, but also as you suggested is good, is further out of the range of the human voice. It is almost double the crossover point used in the Tannoy, isn't it? From my reading, Castle is known as being a great speaker for presenting the human voice, and I'd imagine it is due at least in some part to the high crossover point. I guess I will hear what vocals sound like on each speaker as I go along. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15855 Registered: May-04 | . " ... much greater improvements can be made, just by changing the position of the driver. In addition, rounded edges will cause less refraction than square or raised edges, further improving the overall response, and particularly at the higher frequencies; http://sound.westhost.com/bafflestep.htm Unlike the Spica, there is no felt material to absorb energy ... A few speakers still use a felt damping material to minimize diffraction/refraction effects but, for the most part, such baffle treatments almost always narrow the listening sweet spot. This is largely considered undesireable if the speaker is to do HT duty. "With relatively narrow baffles, the amount of gain will surely be smaller ... " It's a series of trade offs again. Vandersteen dosn't employ a baffle of any significance on his upper frequency drivers, preferring instead to have an immediate trajectory into open space. If the bafffle is wide enough to allow about 2 1/2 bounces of any frequency, the average gain will be as good as if there were an infinite number of bounces. Obviously, if the designer is to use the baffle in order to gain efficiency of the driver, the crossover frequency must be taken into account to determine the necessary baffle dimensions as lower frequencies have longer wavelengths. "Would it not be one or the other with only one woofer and one tweeter?" Not necessarilly. Normally, there will be a high pass filter for the tweeter and a low pass for the woofer to keep out of bandwidth information from reaching each driver. You can design a two way system with just a HP on the tweeter but you'd need a fairly well behaved mid/woofer to pull it off successfully since the designer will be relying exclusively on the mechanical roll out characteristics of the driver. "Castle is known as being a great speaker for presenting the human voice, and I'd imagine it is due at least in some part to the high crossover point." Castle was originally a British owned company which designed what would be called quite typical "British/BBC" sound. They were a licensee for the original LS3/5a (though my memory says they never actually produced a product for that design). As with a vast majority of British speaker companies which sprang from the BBC design studios, their sound was classically focused upon getting the mids right first and foremost. As the British market bent a bit in the late 1980's/early '90's to serve the tastes of the much larger and much more lucrative US market, most British speakers lost some of that "midrange first" quality in favor of a snappier, bigger and balls-ier balance but were never quite so obvious about their intentions as many of the original US speaker companies. You might want to research something on the order of "east coast/west coast loudspeaker sound" for more on how American speakers developed. Each of the three companies; Quad, Castle and Tannoy were originally British companies. Each had its own distinct sound, particularly in the mids. So which is the most "midrange first" accurate? Or, which is the most "midrange first" tangible? Or, which is the most mids-first desireable? Those are questions you'll need to ask yourself and answer as best you can. One thing to remember is the overall speaker's tonal balance is typically perceived not as the speaker would measure but as the speaker would sound in a typical room. The 3/5a was designed around the sound of the human voice by engineers whose job it was to capture, amplify, record and reproduce the human voice 95% of the time. One of the bits of magic which made the 3/5a successful was its balance which had little deep octave bass extension under about 50Hz but yet also had mids and highs which were in step with the overall balance of the system. This is where many speakers fail for many listeners, they have a sonic balance that tilts one way or the other while most successful speakers have a balance where both the bass and the treble respect the mids. I hope you understand that concept, it is a long standing technique in speaker design and, if you capture the right balance, it almost ensures a long term relationship with the listener even if there are other disadvantages to the speaker system. This is more complicated than just being "accurate" or "transparent". The 3/5a's had this quality, the original Qaud ESL's had this, as did the original KLH 9's. While both the 3/5a and the Wilson Watt were specifically designed as mobile monitor speaker systems, the 3/5a had this balancing act down pat while the Watt went for a different performance. . . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15856 Registered: May-04 | . "Another thing I note is the crossover position of my speakers. Tannoy is lowest at 1.8kHz ... " The dual concentric placement of the drivers means the high frequency unit is placed inside and typically at the center of the sloping sides of the low frequency driver. This affords the advantage of the tweeter operating in a horn loading technique. (If the low frequency driver has not straight but curved sides to its driver face, then the horn would become an "expotential horn" shape; http://search.yahoo.com/search?ei=utf-8&fr=slv8-hptb5&p=expotential%20horn%20loa ding&type=) When properly designed and executed, horn loading can not only increase the efficiency of the driver but also slightly extend the low frequency cut off beyond the driver's free air resonance. With a touch of horn accommodation the tweeter in the Tannoy's might be using this technique to extend the range of the high frequency system this driving down the HP crossover point which should allow more of the information to exist from one driver. So, taken against the Castle's higher crossover frequency, there are trade offs to be found in both designs. The downside to this type of concentric placement is the upper frequency driver is also always operating within the throat of a moving horn which makes for less than ideal horn loading of the driver. Not only does this tend to make for ragged low frequency extension but it complicates the phase relationships between the two drivers. You'll often see a good deal of comb filtering happening in coincident or concentric drivers. Read the Thiel and KEF lit above for more on this phenomenon. Of course, the radiating surface area of the two drivers could also account for a change in crossover frequency range. Changing the radiating surface dimensions of the high frequency driver would also impact dispersion characteristics which then might also be further hindered by the horn's propensity toward directional beaming of the upper frequency band. As with the felt damping on the Spica, this would narrow the sweet spot for evenly balanced in-room frequency response and good perceived stereo imaging and soundstaging as less early reflections from nearby surfaces are the result of a more directional dispersion. This might be good or bad depending on the designer's intent and the user's positioning and expectations of the speaker system. . |
|
Silver Member Username: KbearCanada Post Number: 954 Registered: Dec-06 | I know that Reference 3a uses just one capacitor to stop low frequencies from reaching the tweeter. That is it, the low frequency driver is left to roll off naturally. I thought I read this described as a crossoverless system, but perhaps it isn't. It is intriguing nevertheless. Perhaps one speaker that would highlight what can be achieved by avoiding a complicated crossover, but not going outright to a one driver solution. So does the Tannoy employ two crossovers, or would it be regarded as one? In any case, it seems as if the highs roll off by 6dB per octave, while the lows by 12dB per octave. In the case of the Castle, I would guess both drivers roll off at 12dB per octave. And yes, I think I know what you mean by saying the highs and lows respect the mids. If I read into your line that most speakers are tilted one way, they will either have plenty of high or low energy. Whereas one that isn't that way will probably have more energy in the midrange, with highs and lows that provide the necessary support but do not call attention to themselves at the expense of what is happening in the mids. Very interesting info about the dual concentric, Jan. |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15857 Registered: May-04 | . I editted the last post, go back and catch a few things you might have missed while posting. The Reference 3a is not unlike the Gallos 3 series in that the midrange drivers roll off mechanically and only a simple first order filter is used as a HP filter. I think both companies bill their designs as "crossover-less" but that is really just a marketing term at that point. It's all but impossible to design a driver small enough to act as a real world high frequency system that can also handle the large amounts of power applied to the woofer. Tweeters seldom see more than a few watts at most across their terminals. Woofers, on the otherhand, can consume 10's to 100's of times the amount of energy directed to the high frequency driver. "So does the Tannoy employ two crossovers, or would it be regarded as one? In any case, it seems as if the highs roll off by 6dB per octave, while the lows by 12dB per octave. In the case of the Castle, I would guess both drivers roll off at 12dB per octave." The Tannoy employs two filters - a LP and a HP - which, in traditional speaker design, would comprise one two way crossover. Don't get hung up on the numbers. You should want to know the type of filter used (Bessel, Butterworth, etc.) to be informed about the technical apsects of the crossover. However, at this point in your learning curve, none of this should really matter to you other than to say you looked to see what the manufacturer claimed. Back to square one and all that really matters are the HxWxD plus the weight. Once you've borrowed the neighbor's truck to get those big and heavy things home, you sit down and judge the music. . |
|
Platinum Member Username: Jan_b_vigneDallas, TX Post Number: 15858 Registered: May-04 | . "And yes, I think I know what you mean by saying the highs and lows respect the mids. If I read into your line that most speakers are tilted one way, they will either have plenty of high or low energy. Whereas one that isn't that way will probably have more energy in the midrange, with highs and lows that provide the necessary support but do not call attention to themselves at the expense of what is happening in the mids." Not necessarilly more energy in the mids, just a "balanced" sound. If the speaker cannot reproduce deep bass with a facile, lucid preformance, then designing it to have extreme high frequency extension will tend towards making the system sound unbalanced. IMO ProAc is the epitome of this sort of unbalanced sound with many of the Thiel's a close second and the Paradigms giving a good chase for the ring. The shoebox sized 3/5a introduced a slight hump in the midbass around 80-120Hz which subjectively gave the impression the speaker was capable of much deeper bass extension than the enclosure/driver would physically permit with ease and efficiency. Therefore, the listener - hearing the slightly "unbalanced sound" with the hump - accepted the bass as being deeper and more powerful which allowed the engineers to extend the upper frequency response to balance the overall system sound. In turn the upper frequency repsonse made the perceived bass response necessary to balance the speaker's sound.Psychoacoustics! When MW first heard my 3/5a's in my room, he wanted to know where the subwoofer was. Of course, there was no sub used with those speakers but the size he saw in the 3/5a did not compute with the perceived bass response he heard. Since that time many engineers have used a similar trick to balance their speakers tone in both large and small enclosures. If the speaker is to balance extreme high frequency response, there must be a sufficient extension and amount of bass energy perceived by the listener. Too much mid energy will tend towards making for a honky, forward speaker system which tends to lend a sameness to all music - a Klipsch. It's the same idea as a balanced car, enough horsepower and torque along with well placed gear ratios to facilitate the handling characteristics without getting you into trouble with severe under or oversteer. Same as baking a cake, too many eggs will make the batter rich but also make the cake taste spongey unless you cook it for a longer period of time which risks making the final results too dry. Same as making a marinara sauce and having too much peppery, raw garlic taste vs the parsley and basil for the time of the season when you picked the tomatoes. This is the art of speaker design that cannot really be taught but must be learned through experience. You really have to hear this to understand it but once you catch on to the basic concept it should make sense every time you hear a speaker which obeys or breaks the rule. You'll then begin to notice this same psychoacoustic "trick" must also be inherent in the electroncis you pair with a well balanced speaker. Not that the amp needs a hump in the midbass but that the same rules apply to an amp, preamp, CD, cartridge, etc,; they must sound appropriately musically balanced. A too bright amp will always be a too bright amp. A dull cartridge will always be a dull cartridge and is only good for covering up mistakes elsewhere. A balanced amp, cartridge, etc. will always be a balanced amp, cartridge, etc. and will most often offer the most long term satisfaction as the musical material rolls through. Again, this is not just about being accurate, transparent or neutral. This is about being "balanced". . |
|
Silver Member Username: KbearCanada Post Number: 955 Registered: Dec-06 | I see. The midbass hump allows for more extension up top. I'm guessing you'd need the mids to be in line in order to maintain balance. I know some speakers have a tipped up bottom end and a tipped up high end, to make the sound more exciting overall. Of course, how that translates when you listen to music cannot be seen from looking at measurements. My old Monitor Audio RS5 had way too much treble energy. I liked what it was doing across the bass and the mids, and I thought it was an overall exciting speaker to listen to. But the highs too often made it difficult to hear the mids and bass. Songs that had parts where highs were not prominent sounded great...when they were there, they stuck out. My Quad, especially on my new stands, is the opposite. I find the lows to be almost overwhelming sometimes. Amazing what that little speaker can achieve. Highs are nice, if a little closed in, and the mids seem further back in the mix. The Tannoy seems to have an overall nicely balanced sound. That is, I should say, relative to the other two speakers. How they'd fare against other speakers I can't say. Back to square one and all that really matters are the HxWxD plus the weight. Once you've borrowed the neighbor's truck to get those big and heavy things home, you sit down and judge the music. Agreed! I guess I'm just naturally curious. It's as you have said, different speaker and amp designs are a series of trade offs, and it depends what trade offs are acceptable to the listener, which can't be determined just by studying specs. |
Main Forums
Today's Posts- Home Audio Forum
- A/V Receivers Forum
- Amps Forum
- Cassette Forum
- CD Players Forum
- CD Recorders Forum
- DAC & Transports Forum
- DVD-Audio & SACD Forum
- Equalizers Forum
- Integrated Amps Forum
- iPod Docks Forum
- MiniDisc Forum
- Mini Systems Forum
- Digital Music Systems Forum
- Phono Forum
- Preamps Forum
- Speakers Forum
- Subwoofers Forum
- Tuners Forum
- Home Video Forum
- Home Theater Forum
- Car Audio Forum
- Accessories Forum
- All Forum Topics