The answer is most certainly yes, I have and continue to view the reproduction problem as one of a fractal nature, although I use the term entropy as its more appropriate. I have hypothesized that the ear likes a certain amount of entropy in its signals because whenever we increase the entropy the ear seems to like it.
Jean-Micheal is correct in that the foam is a fractal structure and will impart a small degree of entropy to the signal (in addition to its absorptive characteristics). This may be why it sounds so good when in fact the measurements indicate only a small change.
I had tried to interest several driver manufacturers in making an increased entropy loudspeaker along the lines of a fractal structure, but alas, to those guys "no change is a good change".
I have gone a ways down this path, but I'm not prepared to discuss it at length in a public forum.
Jean-Micheal is correct in that the foam is a fractal structure and will impart a small degree of entropy to the signal (in addition to its absorptive characteristics). This may be why it sounds so good when in fact the measurements indicate only a small change.
I had tried to interest several driver manufacturers in making an increased entropy loudspeaker along the lines of a fractal structure, but alas, to those guys "no change is a good change".
I have gone a ways down this path, but I'm not prepared to discuss it at length in a public forum.
gedlee said:
With all due respect, that seems as about as narrow minded as the view that horn loaded systems can never sound good. Granted, about 95 percent of car stereo sounds like crap, but I have heard some aftermarket systems done by pros using very capable drivers and who have no biases about judicious use of EQ guided by RTA analysis--the results would likely give many systems here a strong run for the money.
As miserable as a car may be acoustically, its a precisely known quantity with only a couple of sweet spots that need to be addressed. And as good as home audio can be, there is something about an open stretch of road and good tunes/capable playback that can not be matched by any home system, period.
Ironically, some of the best horn loaded systems I have heard were in cars. 😀 One even did away with the front seats and put the driver rear center. 😎
NewMexicoNoob said:
Well I don't see much respect shown in that statement. And of course waveguide systems can sound good. But having spent about 20 years in car audio and the same or more in home audio, I think that it is fair to say that the very best cars cannot compete with the very best home systems. Sure there is some overlap between the poorer home systems with car systems, but in the end the car is simply never going to be able to match a great home system. There are way too many constraints.
No disrespect intended, just taking exception with the blanket statement with implied condescension about laughter re "great car stereo."
I have heard what I consider to be great car sound, albeit while parked and the motor off. 😉
What I object to is the kind of snobbery I see from time to time--that car audio aficianodos are some lesser breed only concerned with high SPL shoot-outs. Some of the more interesting technologies such as DSP corrected bass emerged from car audio IIRC, and the innovation and effort involved in these showcase efforts has much to be admired. I'm glad that many young people are passionate about audio these days and certainly don't blame lousy car stereos for the dearth of good recording technique/practices. I do take exception with the only good if your ears bleed camp or the 190+dB shoot-outs that have no utility whatsoever.
Rather see that energy put in making the cars faster like we old timers did in the old days. 😀
I have heard what I consider to be great car sound, albeit while parked and the motor off. 😉
What I object to is the kind of snobbery I see from time to time--that car audio aficianodos are some lesser breed only concerned with high SPL shoot-outs. Some of the more interesting technologies such as DSP corrected bass emerged from car audio IIRC, and the innovation and effort involved in these showcase efforts has much to be admired. I'm glad that many young people are passionate about audio these days and certainly don't blame lousy car stereos for the dearth of good recording technique/practices. I do take exception with the only good if your ears bleed camp or the 190+dB shoot-outs that have no utility whatsoever.
Rather see that energy put in making the cars faster like we old timers did in the old days. 😀
Hi Earl,
About standard audio systems, and cars as listening environments, I hear you. Completely agree.
Hi NewMexicoNoob,
I was involved in car audio from the start, and had a 12.5 sec '67 Olds Cutlass built. I was afraid of that car, 450 real horses instead of make believe noises.
When those "Mind Blowers" were out, I had a Uher 4400 Report in my car with a pair of Grundig speakers on the parcel shelf. Sounded pretty good at the time. I installed a pair of 12" low compliance woofers in the back where the rear seat used to be. Pretty efficient. They died when the MGT 2100 came out (coupled to a Concord HPL-200). Killer audio, the MGT was 50 WPC with +/-33 VDC rails. 😎
Anyway, Earl must be killing himself laughing by now. My final system was a Nakamichi system, PA-300 II and PA-350, TD-700 and Denon DCC-8900. I still have it. Throughout time I've done warranty service on Nakamichi, McIntosh, Alpine, Yamaha, Linear Power, Audio Control, Carver, Denon and probably some I forgot.
My point. Good car audio can sound very good indeed. But it can't hold a candle to a good home audio system. They play with DSP at home too!
-Chris
About standard audio systems, and cars as listening environments, I hear you. Completely agree.
Hi NewMexicoNoob,
I was involved in car audio from the start, and had a 12.5 sec '67 Olds Cutlass built. I was afraid of that car, 450 real horses instead of make believe noises.
When those "Mind Blowers" were out, I had a Uher 4400 Report in my car with a pair of Grundig speakers on the parcel shelf. Sounded pretty good at the time. I installed a pair of 12" low compliance woofers in the back where the rear seat used to be. Pretty efficient. They died when the MGT 2100 came out (coupled to a Concord HPL-200). Killer audio, the MGT was 50 WPC with +/-33 VDC rails. 😎
Anyway, Earl must be killing himself laughing by now. My final system was a Nakamichi system, PA-300 II and PA-350, TD-700 and Denon DCC-8900. I still have it. Throughout time I've done warranty service on Nakamichi, McIntosh, Alpine, Yamaha, Linear Power, Audio Control, Carver, Denon and probably some I forgot.
My point. Good car audio can sound very good indeed. But it can't hold a candle to a good home audio system. They play with DSP at home too!
-Chris
Chris,
Hey I'm a DEQX owner myself, so I have heard of DSP. 😀
But ten years ago I was going crazy when all there was in the home arena was the megabuck and functionally limited Meridian system--meanwhile the auto jockeys had these outrageous long excursion woofers, DSP bass EQ, and clean outputs that rivaled a Saturn-V. Home audio innovation seemed dead in its tracks innovation wise and apart from hawking more channels and adding a TV to the mix, seemed headed nowhere.
Re quality, I'm only saying that the best of show in quality reproduction categories these days is a far cry from a couple of Jensen 6x9's! These systems have nuance and punch and are flatter than most home systems. 10 to 30K is a lot of nice components in a confined space! Is it sane,? Heck I don't know if any of this is, but if you spend more time in the car than a small apartment, there is some logic...
Hey I'm a DEQX owner myself, so I have heard of DSP. 😀
But ten years ago I was going crazy when all there was in the home arena was the megabuck and functionally limited Meridian system--meanwhile the auto jockeys had these outrageous long excursion woofers, DSP bass EQ, and clean outputs that rivaled a Saturn-V. Home audio innovation seemed dead in its tracks innovation wise and apart from hawking more channels and adding a TV to the mix, seemed headed nowhere.
Re quality, I'm only saying that the best of show in quality reproduction categories these days is a far cry from a couple of Jensen 6x9's! These systems have nuance and punch and are flatter than most home systems. 10 to 30K is a lot of nice components in a confined space! Is it sane,? Heck I don't know if any of this is, but if you spend more time in the car than a small apartment, there is some logic...
Why so defensive?
I spent 20 years designing sound systems for cars so I know whats possible and whats not. Car audio has come a long way since the Jensen 6x9 yes, but its now up against a wall which is not going to move. That wall is the listening environment. Even in cars it is size that determines just about everything. The best systems that I heard were in bigger vehicles (Lincoln Navigator) - DAAA! So is it any wonder that the home has the natural advantage just as an auditorium has an advantage over a home? No one is criticizing what can be done in a car, we're just saying that the potential is limited.
Nothing in terms of DSP, etc. ever started in automotive, it all started in Pro and really good home systems have used pro gear for a long time. What did start in automotive were some concepts in the EQ of small spaces etc. which were first used there. Virtually everything else has followed Pro.
I spent 20 years designing sound systems for cars so I know whats possible and whats not. Car audio has come a long way since the Jensen 6x9 yes, but its now up against a wall which is not going to move. That wall is the listening environment. Even in cars it is size that determines just about everything. The best systems that I heard were in bigger vehicles (Lincoln Navigator) - DAAA! So is it any wonder that the home has the natural advantage just as an auditorium has an advantage over a home? No one is criticizing what can be done in a car, we're just saying that the potential is limited.
Nothing in terms of DSP, etc. ever started in automotive, it all started in Pro and really good home systems have used pro gear for a long time. What did start in automotive were some concepts in the EQ of small spaces etc. which were first used there. Virtually everything else has followed Pro.
Dr. G,
The Pro audio point is a good one--I guess part of what I was reacting to is the very conservative nature--almost Luddite--of home audio to embrace technological innovations as they come down the pike. Four to five years post DEQX and TaCT I still see little penetration in to the market, even in niches such as DIY where education and proclivity should make it such products a natural. That the younger generation seems more willing to use EQ, bi & triamping, etc in their cars I see as a positive. So maybe I am defending car audio on that basis from what I perceived as a somewhat holier-than-thou attitude.
I have been curious based on my last car stereo (Acura TL with DVD-A 5.1 sound system) whether a good set up with highly directive drivers migfht sound fairly extraordinary with a properly recorded multichannel source. For all the many pitfalls, there are some potential advantages--somewhere between headphones and a trypical HT setup say where there are a lot of reflections from the direct channels that compete with ambience info from the rears. I am only guessing as there is such a dearth of such recordings I never had the opportunity to play, and the stock speakers were all conventional drivers. I have also wondered whether a small line source mounted in the A pillars might have beneficial effects re the rapid falloff and relative mismatch between channels.
I guess I'm the type who wonders whether there are some unique opportunities in such an environment that might be exploited, rather than simply believing its hopeless. It may be, but having heard some what I consider to be some pretty mind blowing auto systems, I have just always wondered what if anything could be done to take it to that next level of realism.
But what I don't want to do is hijack this thread on a temporary tangent dealing with a topic that you didn't want to discuss in the first place. I'm new here, and want to become a regular member.
The Pro audio point is a good one--I guess part of what I was reacting to is the very conservative nature--almost Luddite--of home audio to embrace technological innovations as they come down the pike. Four to five years post DEQX and TaCT I still see little penetration in to the market, even in niches such as DIY where education and proclivity should make it such products a natural. That the younger generation seems more willing to use EQ, bi & triamping, etc in their cars I see as a positive. So maybe I am defending car audio on that basis from what I perceived as a somewhat holier-than-thou attitude.
I have been curious based on my last car stereo (Acura TL with DVD-A 5.1 sound system) whether a good set up with highly directive drivers migfht sound fairly extraordinary with a properly recorded multichannel source. For all the many pitfalls, there are some potential advantages--somewhere between headphones and a trypical HT setup say where there are a lot of reflections from the direct channels that compete with ambience info from the rears. I am only guessing as there is such a dearth of such recordings I never had the opportunity to play, and the stock speakers were all conventional drivers. I have also wondered whether a small line source mounted in the A pillars might have beneficial effects re the rapid falloff and relative mismatch between channels.
I guess I'm the type who wonders whether there are some unique opportunities in such an environment that might be exploited, rather than simply believing its hopeless. It may be, but having heard some what I consider to be some pretty mind blowing auto systems, I have just always wondered what if anything could be done to take it to that next level of realism.
But what I don't want to do is hijack this thread on a temporary tangent dealing with a topic that you didn't want to discuss in the first place. I'm new here, and want to become a regular member.
NewMexicoNoob said:
I've said it before, but I'll repeat it since you say that you are new. I don't see DEQX or TaCT or other DSP controllers as having any real benefit in home audio. I don't use them and I don't recommend them. EQ, no matter how sophisticated, cannot solve acoustics problem and, in general they make them worse. Electronic EQ can only ever work at a single point in space and the acoustic field is three dimensional. It can be shown that correcting one point in space will virtually always make all other points worse. They do create really nice looking curves - AT THAT POINT! Only acoustic solutions can fix acoustic problems. I use no EQ in my systems at all (beyond the passive crossovers of course.)
As to there being a unfound "holy grail" for car audio, there have been a lot of attempts by some pretty competent people, but in the end it always comes down to simple competent design and attention to details. In my 20 years I never saw any magical solutions that could make the problems go away. I don't expect that one will ever be found. But there I go again being closed minded and negative.
And no one said that cars were hopeless, they can be made to be very good. They just can't be made to compare with a good home system.
Now, back to waveguides, OK?
Not to be a contrarian--this is your thread after all and I am not looking to argue, but I feel that some of the above comments are potentially misleading:
First, DEQX corrects time domain problems as well as frequency domain, and the improvement in coherence varies between subtle and astonishing, and when done properly, always to the better. The cost being that one needs to operate in the digital domain (anathema to many given issues with pre-ringing, etc) and some signal delay. Group delay plots are illustrative, but listening makes it obvious even the most casual of listeners. In other words, it makes sense to have as coherent a wavelaunch at the source as practical. I don't think that goal can be argued, even if we haggle endlessly over the means by which to best accomplish it.
The Tact unit OTOH is more along the lines of what Dr Geddes describes and takes issue with, but even so, many find the improvements laudable--at least within a sweetspot. Neural nets/artificial intelligence were tried by Denon to expand the sweetspot--how successful is a matter of opinion.
Personally, having owned a DEQX now for close to 4 years I would never go back to chasing the tweak of the month.
But different strokes for different folks is what makes audio such a fascinating passion and pasttime. I believe waveguides are very important and am here to learn more about them as power response has been proven to be highly important as well--and the directivity issues are only improved digitally (if I understand correctly) through removal of lobing around XO points.
First, DEQX corrects time domain problems as well as frequency domain, and the improvement in coherence varies between subtle and astonishing, and when done properly, always to the better. The cost being that one needs to operate in the digital domain (anathema to many given issues with pre-ringing, etc) and some signal delay. Group delay plots are illustrative, but listening makes it obvious even the most casual of listeners. In other words, it makes sense to have as coherent a wavelaunch at the source as practical. I don't think that goal can be argued, even if we haggle endlessly over the means by which to best accomplish it.
The Tact unit OTOH is more along the lines of what Dr Geddes describes and takes issue with, but even so, many find the improvements laudable--at least within a sweetspot. Neural nets/artificial intelligence were tried by Denon to expand the sweetspot--how successful is a matter of opinion.
Personally, having owned a DEQX now for close to 4 years I would never go back to chasing the tweak of the month.
But different strokes for different folks is what makes audio such a fascinating passion and pasttime. I believe waveguides are very important and am here to learn more about them as power response has been proven to be highly important as well--and the directivity issues are only improved digitally (if I understand correctly) through removal of lobing around XO points.
Well I take exception to your statements.
"DEQX corrects time domain problems as well as frequency domain"
Any correction of one domain simultaneously corrects the other. This is the definition of the Fourier transform.
BUT this correction is only valid exactly at the point of correction, the sound field is INCORRECT at every other point - there is no global correction with any electronic system.
The fact that it sounds good to you, or that it might in fact improve the sound of your speakers is not the point. The point is that a correctly designed acoustical system cannot be improved with DEQX and IMO the correct acoustical solution - which is globally correct - will always sound better. To wit my own systems which use no electronic EQ at all.
"DEQX corrects time domain problems as well as frequency domain"
Any correction of one domain simultaneously corrects the other. This is the definition of the Fourier transform.
BUT this correction is only valid exactly at the point of correction, the sound field is INCORRECT at every other point - there is no global correction with any electronic system.
The fact that it sounds good to you, or that it might in fact improve the sound of your speakers is not the point. The point is that a correctly designed acoustical system cannot be improved with DEQX and IMO the correct acoustical solution - which is globally correct - will always sound better. To wit my own systems which use no electronic EQ at all.
And I to yours sir. For instance, there are creatures known as all pass filters which have a flat frequency response, but shift phase angles all over the map. And if it is not true that there is some merit in launching music from the source with a flat frequency response and an intact phase relationship between the various drivers when constraints dictate multiple drivers be used, there has been an awful lot of wasted effort over the past century in loudspeaker design.
If one can approximate this, I submit to you there is a good chance it will sound much better than one that does not most anywhere in the listening space!
The DEQX unit's number one objective is to improve such departures from linearity at the source (the loudspeaker itself) and not some arbitrary point in space. Tho it is capable of helping to tame annoying room resonanances, acoustic remedies are generally preferred whenever possible, and frequency tailoring should be resrticted to the bottom few octaves.
This I think is a fundamental difference between it and most DSP based devices that is crucial to understand. But if you do already, there is little point in arguing about such design goals. I'm a happy camper living with its warts and wonders.
If one can accomplish these goals w/o any such digital sophistry, I'm all for it. Minimalism is good. But most systems that manage to do so that I have heard also cost an arm and a leg. That cost no object approach is not something I can afford, nor even appreciate.
I haven't heard any of your systems so I have no idea as to how successful you have been; judging by the popularity, you indeed must be doing a lot right, and again I am here to learn about wave guides, not to argue engineering theory.
If one can approximate this, I submit to you there is a good chance it will sound much better than one that does not most anywhere in the listening space!
The DEQX unit's number one objective is to improve such departures from linearity at the source (the loudspeaker itself) and not some arbitrary point in space. Tho it is capable of helping to tame annoying room resonanances, acoustic remedies are generally preferred whenever possible, and frequency tailoring should be resrticted to the bottom few octaves.
This I think is a fundamental difference between it and most DSP based devices that is crucial to understand. But if you do already, there is little point in arguing about such design goals. I'm a happy camper living with its warts and wonders.
If one can accomplish these goals w/o any such digital sophistry, I'm all for it. Minimalism is good. But most systems that manage to do so that I have heard also cost an arm and a leg. That cost no object approach is not something I can afford, nor even appreciate.
I haven't heard any of your systems so I have no idea as to how successful you have been; judging by the popularity, you indeed must be doing a lot right, and again I am here to learn about wave guides, not to argue engineering theory.
I think that I understand the DEQX as well as anyone. I used one about 15 years ago.
There was nothing wrong with my first statement since I was talking about both the magnitude and phase of the Fourier transform not the magnitude only. In that context frequency response IS time response no matter what kind of filter or phase response we are talking about.
Please explain to me how a DEQX can correct for a faulting polar response? Or how about a hole that exists ONLY on axis? Or any number of other acoustic effects which exist in real loudspeakers. Electronic correction can only correct problems that are "lumped parameter", i.e. occur at all points in space, like a resonance in the driver itself - BUT NOT cone resonances since they have strong directional response effects. No diffraction problem could be fixed by electronic means since it is spatial.
You want to learn about waveguides thats great, but if my points here are lost on you then I don't think that you will see why and how waveguides are the best solution to the acoustics problems.
There was nothing wrong with my first statement since I was talking about both the magnitude and phase of the Fourier transform not the magnitude only. In that context frequency response IS time response no matter what kind of filter or phase response we are talking about.
Please explain to me how a DEQX can correct for a faulting polar response? Or how about a hole that exists ONLY on axis? Or any number of other acoustic effects which exist in real loudspeakers. Electronic correction can only correct problems that are "lumped parameter", i.e. occur at all points in space, like a resonance in the driver itself - BUT NOT cone resonances since they have strong directional response effects. No diffraction problem could be fixed by electronic means since it is spatial.
You want to learn about waveguides thats great, but if my points here are lost on you then I don't think that you will see why and how waveguides are the best solution to the acoustics problems.
Hi NewMexicoNoob,
If you are searching for perfection in a car, or home - forget it!
My experience agrees with what Earl has said. I find the best auto sound systems came about through acceptance of the laws of physics and competent design. Beyond that, your car is for listening enjoyment and the news. If you spend the same amount of money and care on your car system as you do in your home system, the home system has a head start and will turn out better.
I have had some complicated car audio systems, never once have I ever suggested you shouldn't try to make it sound good. However the use of DSP will always bring you the same result. The more you correct in DSP, the more other problems you will create. This also holds true for home systems. Pro systems are another matter. Sound quality is not the main goal most times. Non-destructive SPL and difficult to cross over speakers need to be stitched together. DSP solutions work well in these cases. I did a great deal of pro sound and recording studio work too. 😉 Been there - done that.
If you are going to spend $10 ~ $20 K on a car audio system, stop! Buy a better car instead. It will be quieter and support lower bass. Probably has a decent system in it anyway.
Sorry for the aside Earl, back to horns...
-Chris
If you are searching for perfection in a car, or home - forget it!
My experience agrees with what Earl has said. I find the best auto sound systems came about through acceptance of the laws of physics and competent design. Beyond that, your car is for listening enjoyment and the news. If you spend the same amount of money and care on your car system as you do in your home system, the home system has a head start and will turn out better.
I have had some complicated car audio systems, never once have I ever suggested you shouldn't try to make it sound good. However the use of DSP will always bring you the same result. The more you correct in DSP, the more other problems you will create. This also holds true for home systems. Pro systems are another matter. Sound quality is not the main goal most times. Non-destructive SPL and difficult to cross over speakers need to be stitched together. DSP solutions work well in these cases. I did a great deal of pro sound and recording studio work too. 😉 Been there - done that.
If you are going to spend $10 ~ $20 K on a car audio system, stop! Buy a better car instead. It will be quieter and support lower bass. Probably has a decent system in it anyway.
Sorry for the aside Earl, back to horns...
-Chris
"EQ, no matter how sophisticated, cannot solve acoustics problem and, in general they make them worse."
If you can correct driver resonances by notching in the XO, why can't you correct room resonances by notching the speaker?
In any event, your statement is counter to my experience and thousands of others who are using Audyssey EQ in receivers, and who used Behringer BFD for EQ before that.
My listening area is essentially 22' square, which still had bad peaks at 25 and 75 Hz after adding four bass traps.
It's hugely improved w/Audussey, which measures up to eight different positions.
Taking energy away from the modal frequencies reduces modeal amplitude everywhere in the room, not just at the measuring position, and as far as I can tell everywhere in the room is indeed much better.
If you can correct driver resonances by notching in the XO, why can't you correct room resonances by notching the speaker?
In any event, your statement is counter to my experience and thousands of others who are using Audyssey EQ in receivers, and who used Behringer BFD for EQ before that.
My listening area is essentially 22' square, which still had bad peaks at 25 and 75 Hz after adding four bass traps.
It's hugely improved w/Audussey, which measures up to eight different positions.
Taking energy away from the modal frequencies reduces modeal amplitude everywhere in the room, not just at the measuring position, and as far as I can tell everywhere in the room is indeed much better.
You're narrowing the discussion here, we were talking about EQ well above room resonances.
There are situations where I would, and have used some EQ in the modal region of a room. But only if this is done properly, as you suggest. To EQ a room based on a single point measurement is totally wrong no matter what frequency you are talking about. But if you use multiple subs to smooth the frequency and spatial variances and still have a peak at some frequency when averaged over several mic positions THEN EQ of this peak is effective. But EQ of a dip (in the modal region) never works, because dips can only be local (points of cancellation), never global, and raising a dip at one point will cause a peak everywhere else followed by the long ringing of the boosted frequency.
And the only speaker resonances that can be fixed with EQ are those that are in the basic structure, like the magnet, or spider, etc. Resonances that work basically at a point, mostly the voice coil. Once the resonace becomes distributed, like cone resonances, or enclosure resonances, then the spatial distribution of the peaks and dips is not uniform and electrical EQ won't work right. There are some resonances in a waveguide that are lumped parameter - i.e occur at all spatial points - and I "EQ" these in the crossover. But they are very low Q and passive elements work fine.
There are situations where I would, and have used some EQ in the modal region of a room. But only if this is done properly, as you suggest. To EQ a room based on a single point measurement is totally wrong no matter what frequency you are talking about. But if you use multiple subs to smooth the frequency and spatial variances and still have a peak at some frequency when averaged over several mic positions THEN EQ of this peak is effective. But EQ of a dip (in the modal region) never works, because dips can only be local (points of cancellation), never global, and raising a dip at one point will cause a peak everywhere else followed by the long ringing of the boosted frequency.
And the only speaker resonances that can be fixed with EQ are those that are in the basic structure, like the magnet, or spider, etc. Resonances that work basically at a point, mostly the voice coil. Once the resonace becomes distributed, like cone resonances, or enclosure resonances, then the spatial distribution of the peaks and dips is not uniform and electrical EQ won't work right. There are some resonances in a waveguide that are lumped parameter - i.e occur at all spatial points - and I "EQ" these in the crossover. But they are very low Q and passive elements work fine.
Thank you, Noah at least someone else has had a similar experience with the use of DSP. And no, I would certainly not spend 10 to 20K on a car system, and my days of spending such for a home system are over--thanks to DEQX! I was merely wondering whether drivers with greater directivity might have benefits in a car environment, among other things. I wasn't advocating voodoo, or a departure from sound engineering practice, just wondering if there might still be some unexplored avenues for improving on auto sound.
And if you'll reread my post, Dr Geddes, I never advocated the use of EQ above modal frequencies except to flatten frequency response anomalies near the driver where presumably anechoic conditions prevail. Again this is fundamental philosophical difference between DEQX and other forms of DSP!
That Dr Geddes purports to have used a DEQX system 15 years ago says it all--must have been a helluva computer he had, or very limited processing.
Maybe its time to take another listen? Where its done in real time with 4092 FIR filters. Assuming mind and ears are open, it may change your mind. And yes, of course, some problems can't be fixed, but using linear filters with steep slopes can very much help with lobing issues and hence polar response in some areas.
Let me say again, I'm here because I believe that polar response is important as well, not the only requirement obviously for life like reproduction. But based on psychoacoustic research that Dr. Geddes cites on his own web site if I'm not mistaken, the off axis uniformity to +/- 20 degrees or more and delaying early reflections--hence the need for controlled directivity.
Having said that, my understanding is that more or less flat on or close to on axis frequency response remains one of the most, if not the most important characteristic in judging sound quality, or so quoteth the Harmon website.
Apparently Harmon also sides with Dr. Geddes in the importance of controlled directivity with their EOS loaded drivers. I have a foamless pair in a small JBL system (S26) that sounds mighty fine indeed for a $200/pr speaker--after DEQX of course. 😀 The oblate spheroid notion of coordinate transformation was also Dr Geddes idea if I understand correctly.
I hope that in spite of my stubborn refusal to disavow DSP, I may still benefit from the good doctor's wisdom and advice.
And if you'll reread my post, Dr Geddes, I never advocated the use of EQ above modal frequencies except to flatten frequency response anomalies near the driver where presumably anechoic conditions prevail. Again this is fundamental philosophical difference between DEQX and other forms of DSP!
That Dr Geddes purports to have used a DEQX system 15 years ago says it all--must have been a helluva computer he had, or very limited processing.
Maybe its time to take another listen? Where its done in real time with 4092 FIR filters. Assuming mind and ears are open, it may change your mind. And yes, of course, some problems can't be fixed, but using linear filters with steep slopes can very much help with lobing issues and hence polar response in some areas.
Let me say again, I'm here because I believe that polar response is important as well, not the only requirement obviously for life like reproduction. But based on psychoacoustic research that Dr. Geddes cites on his own web site if I'm not mistaken, the off axis uniformity to +/- 20 degrees or more and delaying early reflections--hence the need for controlled directivity.
Having said that, my understanding is that more or less flat on or close to on axis frequency response remains one of the most, if not the most important characteristic in judging sound quality, or so quoteth the Harmon website.
Apparently Harmon also sides with Dr. Geddes in the importance of controlled directivity with their EOS loaded drivers. I have a foamless pair in a small JBL system (S26) that sounds mighty fine indeed for a $200/pr speaker--after DEQX of course. 😀 The oblate spheroid notion of coordinate transformation was also Dr Geddes idea if I understand correctly.
I hope that in spite of my stubborn refusal to disavow DSP, I may still benefit from the good doctor's wisdom and advice.
gedlee said:
I think you guys are talking past one another a little here. Maybe I can help bridge the gap.
I have great respect for Dr. Geddes knowledge, and I own and use a DEQX (just to make sure we are talking about the same device: http://www.deqx.com/products.php )
Noob, I think you mean correction in the time domain to mean time alignment of the drivers at the crossover point. By correction in the frequency domain, you mean equalization of the driver to correct for amplitude variations. The former can be done physically, in a speaker level crossover, a line level analog crossover, or a digital line level crossover. The latter can be done also at speaker level, or line level (analog or digital). I know Earl prefers passive speaker level crossovers for reasons he has stated (cost being a major one). The DEQX makes generating and testing crossovers much easier for me. I find it a great DIY tool.
gedlee said:
Equally true of passive crossover/compensation and electronic crossover/compensation - both applied to the same system.
gedlee said:
Understood, you can't correct for problems (directivity, for instance) that are a function of physical design. But it is nice to be able to balance the system easily and make changes to account for moving to a different space, for example. I'm not talking about basic acoustic issues, just things like adjusting the EQ to account for room gain or even different EQ profiles to use for different volume levels.
gedlee said:
Passive crossovers and digital crossovers are both electronic methods, in that they both only influence the signal going to the driver. As such everything stated above applies equally to both.
In summary, the acoustical system ultimately constrains the best that can be achieved. Crossover and EQ can be done either passively or actively. The choice between the two is primarily a function of process preference and considerations, such as flexibility, ease of design, passive network design experience, component cost, etc.. The same fundamental output can be achieved either way.
Sheldon
edit: composed while Noob replied
