Dispelling Myths About EQ

[sfogg]

I use very narrow band EQ to deal with room resonance problems.

EQ on the top end driver to deal with its mass rolloff.

EQ in the crossovers to make their slope dramatically sharper to reduce combing/lobbing between drivers within the speaker.

And even tone/tilt controls on occasions for recordings that can benefit from it.

Gasp.. must need to rip out everything and start over again cause someone somewhere said EQ is 'bad.'

Shawn

[mas]

I use very narrow band EQ to deal with room resonance problems.

EQ on the top end driver to deal with its mass rolloff.

EQ in the crossovers to make their slope dramatically sharper to reduce combing/lobbing between drivers within the speaker.

And even tone/tilt controls on occasions for recordings that can benefit from it.

Gasp.. must need to rip out everything and start over again cause someone somewhere said EQ is 'bad.'

Shawn

Whether you think EQ is good, bad or indifferent is superfluous.

It is not effective in dealing with a multi-source(real and virtual)/multi-path non-minimum phase environment. Nor does EQ minimize comb filtering or polar anomalies due to overlapping passbands in a crossover with multiple sources. The irony is that this fact was first presented by and proven by a group that even included the original developer of the first 1/3 octave equalizer!

If, on the other hand, if you want to EQ direct minimum phase sound, have fun, but that has little to nothing to do with the use of EQ to fix room and multi-path signal nomalies. ...Unless you live in a near field reflection free zone (where you will have issues with real and virtual (read diffractive) sources of inter-driver anomalies regardless of the crossover slopes).

But it is discouraging to note how your focus on addressing the quality of the source material and the problems inherent in your equipment is confused with the behavior of multi-point sources and of the behavior of sound in a room.

The point of the article was not that EQ can never be used! The point was to recognize where it is effective and where it is inappropriate. Unfortunately, that distinction has still yet to become understood here.

It amazes me how so many here so deftly overcome acoustical physics based simply upon their 'beliefs', the origin of which is "to wish it so".

But why should anyone be surprised when more people believe in ghosts than believe we actually landed on the moon. ...It sure beats trying to figure out where Chicago is on a map!

[sfogg]

"It is not effective in dealing with a multi-source(real and virtual)/multi-path non-minimum phase environment that is known as a room."

That is what you say, others disagree.

Riddle me this... what is the minimum width a notch needs to be to be audible. You yourself have said 1/3 octave in earlier posts on the forum.

So... hypothetically if you have a room with a resonance at 37hz (and room resonances are very narrow... around 1hz wide) and you put on a very high Q filter at 37hz such that you are putting no energy into the room at 37hz the audible effects of that resonance are gone as you are no longer driving the resonance.

Yet the filter is far narrower then 1/3 octave so the notch itself isn't audible per your own posts on this. And the EQ removes the audible effects of that resonance at every point in the room.

"Nor does EQ minimize comb filtering or polar anomalies due to overlapping passbands in a crossover with multiple sources.."

Yes, it most certainly does. This is *plainly* audible on pink noise.

Or do you think crossover slope has no bearing on comb filtering/polar anomalies due to overlapping passbands in a crossover with multiple sources? Because wether you know it or not that is the position you are arguing saying that the way I use EQ in my crossovers has no effect on this.

Shawn

[mas]

"It is not effective in dealing with a multi-source(real and virtual)/multi-path non-minimum phase environment that is known as a room."

That is what you say, others disagree.

Riddle me this... what is the minimum width a notch needs to be to be audible. You yourself have said 1/3 octave in earlier posts on the forum.

So... hypothetically if you have a room with a resonance at 37hz (and room resonances are very narrow... around 1hz wide) and you put on a very high Q filter at 37hz such that you are putting no energy into the room at 37hz the audible effects of that resonance are gone as you are no longer driving the resonance.

Yet the filter is far narrower then 1/3 octave so the notch itself isn't audible per your own posts on this. And the EQ removes the audible effects of that resonance at every point in the room.

"Nor does EQ minimize comb filtering or polar anomalies due to overlapping passbands in a crossover with multiple sources.."

Yes, it most certainly does. This is *plainly* audible on pink noise.

Or do you think crossover slope has no bearing on comb filtering/polar anomalies due to overlapping passbands in a crossover with multiple sources? Because wether you know it or not that is the position you are arguing saying that the way I use EQ in my crossovers has no effect on this.

Shawn

You sound like the folks who debated this back in the late 80's based upon an understanding dating back prior to the mid 70's.

And this has become a very tired debate with so many basing their suppositions upon belief rather than demonstrated physics. But hey, some are content to believe the world is flat.

So EQ resolves comb filtering and polar anomalies? really? I can cite a
LONG list of authorities who maintain, and have demonstrated, that EQ
does not resolve non-mininum phase signal anomalies.Signal alignment
in the time domain does. EQ does not. It is also easily reproduced and demonstrated.

And so you must drive a tuned pipe with the specific frequency to drive a resonance at that frequency? Where did you take physics? And the Q of resonances are typically equal in width to 1 Hz? Wow, thats news. And so the typical Q of summed resonances of the various modes is equal to a width of 1 Hz? Really?

The point is that these issues have been addressed ad nauseum and the debate resolved over 20 years ago, except in the realm of audiophilia who still live in the realm of RTAs.

One wonders what you think causes comb filtering and polar anomalies, especially as you think that EQ can resolve this issues. .

But have fun, as we believe what we choose to believe. Unfortunately acoustical physics doesn't care what we believe. And physics doesn't depend upon poles or a consensus of belief. Go figure.

But what would be nice would be if a number of folks became familiar with current developments and thought in acoustics so that the ramifications of such understanding could be further pursued, rather than the repeated debates over such old topics as whether EQ can fix time domain issues. Anyone up for a debate over whether there exist 'little invisible bugs' that can cause infection and disease rather than spells of witches?

And in the spirirt of quoting those whose beliefs are without question: "To assert that the earth revolves around the sun is as erroneous as to claim that Jesus was not born of a virgin." Cardinal Bellarmine

If only faith were as effective in acoustics...[]

[sfogg]

Are you going to discuss what I posted or just randomly choose things to rally against?

"So EQ resolves comb filtering and polar anomalies? really? "

Read what I wrote.

"Signal alignment in the time domain does."

You keep thinking 'signal alignment in the time domain' is some sort of panacea. It is not the end all be all of signal processing. Is it helpful, certainly. Does it cure all comb filtering issues and polar anomalies... not for non-coincident sources. To claim otherwise is overly simplistic.

You can't time align non-coincendent drivers for all points within a room. Where do you time align the signal for? At best they can be aligned for one plane in the room. By definition everything outside of that plane is not time aligned and therefor still has interference between drivers. Align a pair of vertically arranged drivers. Now move the microphone up 10 feet and measure again... the drivers are not in the same time alignment. Simple geometry shows that to be the case.

"EQ does not. "

Asking the same question again are you arguing that crossover slope has no effect on comb filtering and the resultant polar response anomalies?

Shawn

P.S. Instead of your endlessly tiring insults and such wouldn't a more productive approach be to simply discuss the topic at hand? Your SNR is very poor....

[mas]

Are you going to discuss what I posted or just randomly choose things to rally against?

"So EQ resolves comb filtering and polar anomalies? really? "

Read what I wrote.

"Signal alignment in the time domain does."

You keep thinking 'signal alignment in the time domain' is some sort of panacea. It is not the end all be all of signal processing. Is it helpful, certainly. Does it cure all comb filtering issues and polar anomalies... not for non-coincident sources. To claim otherwise is overly simplistic.

You can't time align non-coincendent drivers for all points within a room. Where do you time align the signal for? At best they can be aligned for one plane in the room. By definition everything outside of that plane is not time aligned and therefor still has interference between drivers. Align a pair of vertically arranged drivers. Now move the microphone up 10 feet and measure again... the drivers are not in the same time alignment. Simple geometry shows that to be the case.

"EQ does not. "

Asking the same question again are you arguing that crossover slope has no effect on comb filtering and the resultant polar response anomalies?

Shawn


P.S. Instead of your endlessly tiring insults and such wouldn't a more productive approach be to simply discuss the topic at hand? Your SNR is very poor....

Signal alignment does not resolve comb filtering and polar anomalies for all points in a room for non coincident drivers.No one to my knowledge has ever claimed otherwise.The nice thing is that this is not required in a finite listening space. And this requirement varies with the our focus outside of the near field where inter-speaker spacing dominates over inter-driver spacing. Although each level of scale contributes to the issue and should be optimally be addressed.

And the amazing thing is that you seem to think that this somehow invalidates its use for this purpose. Nevertheless, it is the only technique that does, unlike the manipulation of the signal within the frequency domain which is utterly ineffective to resolve non-minimum phase anomalies on all accounts. And likewise, in the far field where the speaker itself begins to act as a point source as its interdriver wavefront ceases to vary as it does in the near field, signal alignment again remains the only practical technique that can minimize the effects, whereas the traditional application of EQ fails. And by practical, this means that the average person is not going to design and apply digital signal processing techniques.

Minimizing the degree to which multiple sources reproduce he same passband does minimize the degree of comb filtering. But this is simply to say that it does cause comb filtering and polar anomalies, but we minimize it simply by doing less of it. It in no way eases to result in it. And virtual sources such as diffraction act as virtual sources on the micro level and reflections within the room further aggravate the issue on the macro level.

This issue has been discussed at length in the past. And no, I
am not about to enter into the entire discussion again just because you
you show up with the usual objections. Sources from folks as diverse as
Danley, Davis, Sabine, Berkow and many others have been cited to this
effect.

So you can debate whether it is possible to mitigate all of the phase effects of multiple sources reproducing a common passband whereby varying positions result in varying relative interdriver spacing and distances by signal alignment, or we can debate the original issue which was that EQ is not effective in resolving any of these effects!

Bottomline, EQ is ineffective in resolving these issues in non-minimum phase applications. Rather, the only practical method of resolving and minimizing these effects is signal alignment, although this will not simultaneously resolve all issues at all positions simultaneously. But this technique could still be used to resolve the issue for any of the given points that you wish to address. The point is that EQ cannot. And that was the original issue.

And as usual, you fail to appear during the body of any discussion and opt to offer anecdotal attempts to maintain that EQ can resolve such fundamental phase issues and attempting to invalidate effective methods by pointing out that signal alignment for non-coincident sources varies with position, which in no way invalidates the use of the technique within a particular frame of reference. (although signal alignment within the time domain by such techniques as co-axial design can certainly do this!) while ignoring established fundamentals of modern acoustics that are available to anyone has either the practical application experience or who choses to avail themselves of the current literature and research in the field. .

Your objections are akin to someone, after saying that EQ can solve non-minimum phase errors, showing up to object to signal alignment just as one might object to food being used as a solution to starvation based upon the objection that some may be allergic to wheat or peanuts. It still does not justify the inappropriate application of EQ, a technique that is ineffective in all cases in a non-minimum phase environment...And like the article acknowledged, it is important to understand the limited number of situations, namely minimum phase environments where it can be used effectively.

And the problem remains that too many still believe that EQ can resolve non-minimum phase signal errors and as a result inappropriately apply a technique that only exacerbates a situation.

[sfogg]

"And the amazing thing is that you seem to think that this somehow invalidates its use for this purpose. "

No, the amazing thing is how badly you either misread or intentionally distort what others write. I never said it invalidates it, I even flat out said it was useful.

I wrote the following....

"You keep thinking 'signal alignment in the time domain' is some sort of panacea. It is not the end all be all of signal processing. Is it helpful, certainly. Does it cure all comb filtering issues and polar anomalies... not for non-coincident sources. To claim otherwise is overly simplistic."

Which buried in your latest post you agree with.

"And likewise, in the far field where the speaker itself begins to act as a point source as its interdriver wavefront ceases to vary as it does in the near field, signal alignment again remains the only practical technique that can minimize the effects, whereas the traditional application of EQ fails."

Nice backpedal with the 'traditional application' of EQ. Is this a sign you might have actually stopped for a second to consider how I'm using EQ in the crossover? Naahhhhhh..........

"Minimizing the degree to which multiple sources reproduce he same passband does minimize the degree of comb filtering."

Then like I posted originally...

"EQ in the crossovers to make their slope dramatically sharper to reduce combing/lobbing between drivers within the speaker. "

Which after several posts claiming that this is wrong and that basically I'm an idiot for suggesting it it in fact turns out you agree with this too since I am in fact using EQ to dramatically reduce the passband of multiple sources which minimizes the degree of comb filtering.

"So you can debate whether it is possible to mitigate all of the phase effects of multiple sources reproducing a common passband whereby varying positions result in varying relative interdriver spacing and distances by signal alignment, or we can debate the original issue which was that EQ is not effective in resolving any of these effects! "

ROFLMAO. Hello strawman... MAS would like a word with you.

I never said it mitigated all of the effects. I said it reduced it. Just like signal alignment doesn't mitigate all the effects either, it reduces it as you have now also confirmed.

"Bottomline, EQ is ineffective in resolving these issues in non-minimum phase applications. Rather, the only practical method of resolving and minimizing these effects is signal alignment, although this will not simultaneously resolve all issues at all positions simultaneously."

You have already confirmed that sharper slopes help to reduce comb filtering and polar response problems. So obviously signal alignment is not the only practical method available. Therefor EQ is also a practical method since it can be used to increase the slope of a crossover.


"But this technique could still be used to resolve the issue for any of the given points that you wish to address. The point is that EQ cannot. And that was the original issue."

You are correct in that EQ won't just reduce this issue at any given point you wish to address. The reduction in overlap helps reduce comb filtering everywhere. While OTOH time aligning the system to a given point will in fact make the signals further out of alignment at other points which means they could comb worse, not better there.

Shawn

P.S. How many times are you planning on going back and editing old posts during this discussion? Someone that has to keep adjusting their earlier posts is someone that is reaching to far.....

[mas]

Strawman? Read the original article!

Huh? Do what?

What did they suggest for minimizing polar anomalies? Active crossovers with higher slopes and signal alignment. Passive crossovers are still a problem. So now, according to your view of the world, crossovers are equivalent to equalizers. And I guess you use your equalizer as a crossover. And now you call a crossover 'EQ'. Whatever. LOL

"You are correct in that EQ won't just reduce this issue at any given
point you wish to address. The reduction in overlap helps reduce comb
filtering everywhere. While OTOH time aligning the system to a given
point will in fact make the signals further out of alignment at other
points which means they could comb worse, not better there."

Nope. Aligning arrays and distributed systems in the time domain which are by definition greatly out of alignment and bringing them closer to alignment within their overlapping coverage regions will NOT result in greater anomalies. And using an EQ in lieu of signal alignment will only aggravate the situation further in all cases.

What is amazing is thata there is quite a bit of information posted originally in the article, which was the focus of the thread. The thread is not based upon your perception of reality. You posted within a context begun by the issue that Heyser addressed in the KHorn review and which is further reflected in the leading article. You offer nothing new, and any points you now try to make were already posited and the use of signal alignment rather than EQ was already established.The problem is that you failed to read the article and to garner that they already addressed any points you tried to post as being contrary to the article!

But thanks for arriving late to the game and thinking that you are posting something novel as a defense of EQ despite the issues having been already addresses in the article, and thinking that they someohw validated the USE of EQ where issues such as using EQ to resolve phase anomalies within the room and in speakers persist.

So you choose to post out of context, ignorant of issues already posited and think you are offering something new? Your entire post is a strawman. The original points stand.And unfortunately you offer nothing new, and your points do not negate any points already made. And as far as crossovers functioning as EQ. Now that's new. So I guess we now have to qualify what we mean when using the terms 'EQ' and 'crossover'. Yeah... right.

No one has positied that there are no legitimate uses for EQ. I have certainly never debated their use in minimum phase direct sound applications. Unfortunately, such uses were not being debated. But you don't run for an EQ to fix phase anomalies in non-minimum phase environments.In other words, EQ cannot correct for out of phase destructive interference. And in all cases the use of EQ was referring to an inline device commonly refered to as an "EQ". If this confuses you, query a MI website and search for "equalizers" to discover to what the term commonly refers. That was the original issue, and the point remains.

And any editing I do is immediately upon posting, when I can actually see the entire post, as I have trouble using the preview feature here. So if you have problems with that, wait a few minutes before responding.

Attached is a good synopsis of what signal alignment addresses that EQ does not.

The-Digital-Delay-Advantage..pdf

[pauln]

Question:

""Recent
polls have shown a fifth of audiophiles can't locate the saxophone in a spatial
sound field. Why do you think this is?"

Answer:

"I
personally believe... that U.S. audiophiles are unable to do so... because...

Uh... some audiophiles out there in our nation don't have signal alignment...

And... uh... I believe that our education, like such as in Stereophile and...
uh... The High End, everywhere, like, such as...

And I believe that they should...

Our sound engineers over here in the U.S. should help the U.S... uh...

Or, should help the listeners and should help the comb filtering and the polar
anomalies...

So we will be able to build up our future spatial imaging... for our
[ears]."

[sfogg]

"So now, according to your view of the world, crossovers are equivalent to equalizers. And I guess you use your equalizer as a crossover. And now you call a crossover 'EQ'. Whatever. LOL... And as far as crossovers functioning as EQ. Now that's new. ""

Laugh all you want, it is only new to you. You are blatantly exposing the limits of your knowledge. It is long past time for you to stop talking and learn about Cauer crossovers also called Elliptic crossovers.

Since you obviously have no comprehension of what they are they are a combination of a crossover *and* EQ to increase the slope of the crossover, like I have said multiple times and we have discussed numerous times in several threads.

Yet you still don't comprehend this because you constantly make the assumption that you know all and everyone else knows nothing.

You know what they say about assumptions don't you?

Take this basic fourth order crossover for example...



Now you add a couple of bands of parametric EQ and you suddenly have this...



Instead of taking nearly nearly an octave for the driver to drop 20dB it does it in 100hz.... 1/5 of an octave. That reduces the overlap band between drivers and reduces comb filtering and polar response issues as you have already agreed with.

It is the addition of EQ that had this result, something you have claimed many times now is not possible.

Specifically:

EQ1 -12dB, 607hz Q10

EQ2 +6dB, 499hz Q 7.9


"Nope. Aligning arrays and distributed systems in the time domain which are by definition greatly out of alignment and bringing them closer to alignment within their overlapping coverage regions will NOT result in greater anomalies."

If you alter the timing between drivers such that you move them closer together in one plane it also also moves them further apart in another plane. Basic geometry.

"And using an EQ in lieu of signal alignment will only aggravate the situation further in all cases."

Nope, see above.

Shawn

[pauln]

I'll bet this guy could get to the bottom is this dispute...(General Questions: Math & Physics Revised !)

http://forums.klipsch.com/forums/thread/963401.aspx

[oldenough]

As for me, i'm totally confused. But maybe i'm listening to too much MUSIC......getta life guys....you may even enjoy it.

[Marvel]

As for me, i'm totally confused. But maybe i'm listening to too much MUSIC......getta life guys....you may even enjoy it.

I agree totally. My room and equipment must be a total mess, but I certainly enjoy the music.

[mas]

"So now, according to your view of the world, crossovers are equivalent to equalizers. And I guess you use your equalizer as a crossover. And now you call a crossover 'EQ'. Whatever. LOL... And as far as crossovers functioning as EQ. Now that's new. ""

Laugh all you want, it is only new to you. You are blatantly exposing the limits of your knowledge. It is long past time for you to stop talking and learn about Cauer crossovers also called Elliptic crossovers.

Since you obviously have no comprehension of what they are they are a combination of a crossover *and* EQ to increase the slope of the crossover, like I have said multiple times and we have discussed numerous times in several threads.

Yet you still don't comprehend this because you constantly make the assumption that you know all and everyone else knows nothing.

You know what they say about assumptions don't you?

Instead of taking nearly nearly an octave for the driver to drop 20dB it does it in 100hz.... 1/5 of an octave. That reduces the overlap band between drivers and reduces comb filtering and polar response issues as you have already agreed with.

It is the addition of EQ that had this result, something you have claimed many times now is not possible.

"Nope. Aligning arrays and distributed systems in the time domain which are by definition greatly out of alignment and bringing them closer to alignment within their overlapping coverage regions will NOT result in greater anomalies."

If you alter the timing between drivers such that you move them closer together in one plane it also also moves them further apart in another plane. Basic geometry. **

"And using an EQ in lieu of signal alignment will only aggravate the situation further in all cases."

Nope, see above.

Shawn

Seldom have I encountered someone whose post is so utterly ignore-ant of the original subject as yours.

Did we ever maintain that there were no valid uses for EQ? Nope.

What is really funny is to listen to you try to make what you consider to be a new point, when nothing could be further from the truth!!!

The OBVIOUS fact is that you failed to read the initial subject attachment. Or, if you did, you utterly failed to read it for meaning.

So I re-attached it here so that you can try to read it for meaning.

Talk about a straw man. Next time, as you have a habit of doing this, PLEASE try to find a point that has not already been posited with which you vainly try to discredit.

And lest anyone object to this observation, go back and read the

original post, the subsequent comments regarding the post, and the late

to the party tirade. Onward through the fog indeed...

Read Jim Brown's comments. The fact that he already directly addressed the point you think you are newly raising regarding high slope, signal aligned crossovers in a bi-amped configuration must really confuse you.

It was one of the solutions advocated INSTEAD of trying to resolve the issue with later application of traditional inline 1/3 octave or parametric EQ added to attempt to resolve problems caused by multi-source non-minimum phase interference.

You see, the EQ to which they are referring is the use of what is typically called an equalizer, be it a 1/3 octave or a parametric equalizer, or 'EQ'. And the fact remains that this application of EQ fails to resolve non-minimum phase anomalies in the frequency domain.

And it remains that "your", yeah YOUR(sic), solution that was posited by Jim Brown from the inception pages before you decided to jump in (oh gee!) and display your complete lack of awareness as to the very point of the thread. And that position is precisely that an effective solution for the destructive phase interference cannot be mitigated by the employing an inline EQualizer, be it an inline parametric or 1/3 octave (or any other configuration) EQ.

So I know you think that you are raising an insightful new idea, but next time try reading the lead article of the thread that posited the basic assumptions and predicated the subsequent comments. Its painfully obvious that we (including my subsequent comments prior to your insightful ignorance) already posited steeper slopes and signal alignment as legitimate methods to mitigate the destructive effects. The subsequent use of inline EQ will not. That point, the very subject of the article!, remains!

Oh! But we will be careful not to confuse you too much with the pertinent facts, as it is apparent that they will simply confuse you further. So, you see, you added nothing to the discussion except for your ignorance of what the topic of the discussion was about. But its always a joy to watch you attack a position which several of us had already posited as a partial solution!

And as far as "assumptions" are concerned, you provide a seminal embodiment and prime example of exactly what the implications warn us against.

So, while you ponder your next attack predicated upon asserting what has already been addressed (swooft!), go back and FINALLY read the initial attachment that served as the foundation for the subject. And if you need help with the big words, let us know, I sure we can help, although, after the already demonstrated ignore-ance, I am not sure if we can help you comprehend what you read.

And no, EQ is still not an effective solution to non-minimum phase anomalies.

**And while you are at it, go back and retake 8th grade geometry.

[sfogg]

So to sum up...

I post this:

"EQ in the crossovers to make their slope dramatically sharper to reduce combing/lobbing between drivers within the speaker. "

You post not one, not two but four of your typical insulting messages claiming this is utterly wrong and not possible.

Then after demonstrating to you how my original post that "EQ in the crossovers to make their slope dramatically sharper" is very much possible you flip flop and say the following:

"The fact that he already directly addressed the point you think you are newly raising regarding high slope, signal aligned crossovers in a bi-amped configuration must really confuse you.

It was one of the solutions advocated"

In other words what I posted is accurate and is a method to improve the situation which is in direct contradiction to your earlier four messages. One wonders who is confused... the person who posted this in the first place... or the person who argued vehemently and obnoxiously against it for four posts before flip flopping and now confirming it is a solution?

Of course you couldn't come out and simply say you hadn't considered the case of elliptic crossovers which EQ is very much a part of. No, admitting that would not do at all. So the rest of your post is your typical MO of trying to spin that to try and deflect the topic to something away from those four misguided posts of yours.

"**And while you are at it, go back and retake 8th grade geometry. "

Do you really need me to draw this out for you? If you delay one driver it makes signal alignment better in some locations and worse in others.

Put any two points in space (woofer/tweeter) and there are locations that are naturally equidistant from each driver. Those locations are time aligned naturally.

You add delay to one of the drivers and that effectively moves one of those two points in space. You change one of those two points and the locations that are equidistant from each driver (time aligned) changes as well. The points that were naturally aligned no longer are and new points of alignment emerge.

I can't believe you are arguing over this since you already confirmed this:

"Signal alignment does not resolve comb filtering and polar anomalies for all points in a room for non coincident drivers."

Which is simply another way of saying what I am saying.

Shawn

[Marvel]

[mas]

Amazing. You still don't get it.

You can't do any of what you discuss with an equalizer.

You cannot use an equalizer to modify the slope of a crossover. And you certanly cannot perform signal alignment with an equalizer.

The fact that you did not read the article, nor the comments regarding it afterwards still renders you ignorant of that to which you were responding.

In short, you cannot use EQ to perform any of what you referred. And the destructive non-minimum phase relationship between multiple sources cannot be resolved by employing an equalizer

Sorry if even the article referred to an "equalizer" in the title bar graphic. And the reference to doing shows "without an EQ even in the rack" must really confuse you! Or perhaps not, as you seem to fail to understand the meaning and application of equalization completely.

At at no point did any person refer to taking any piece of gear apart and redesigning the passive crossover to re-engineer the unit to employ elliptic curve topologies. They instead choose to employ an external active crossover with signal alignment. And the use of a separate graphic or parametric EQ was deemed both inadequate and inappropriate to resolve the issue.between both drivers and speakers.

Your entire rant was superfluous to the refered use of a discreet EQ stage to resolve inter-speaker and room anomalies non-minimum phase anomalies.

And as far as "saying what (you) are saying"...your rant did not even address the issue that was at hand. It did not address the use of an Equalizer except to miss the original references which were fundamental to the article and the subsequent comments, and the additional preference to address problems with exisiting assemblages of equipments via the use of signal alignmnet and delay to redress multiple source destructive interference in the minimized speaker and array overlap areas. But then I guess the Sabine PDF confuses you as well.

The bottomline is that you beg the obvious that was fundamentally posited in the article as if it is some new insight, while the rest of the geometry lesson that you have attempted to restate several times was simply wrong. The next time you post spend a minute to discover to what you are
responding and the context in which you respond.

Showing up late to the
party and ranting about elliptic filters is moot.

And minimizing the inter-source distance and Increasing the alignment in one plane does NOT increase the misalignment in another plane.Your assertion that placing two drivers closer together on one plane makes them further apart in another plane is WRONG.

While it does not align the acoustic centers in all planes, their offsets are likewise reduced and the destructive interference also minimized by close spacing. ...Not increased as you erroneously claimed. But then we do have the luxury of defining the coverage areas. So we don't have to resolve all places.

So, review your geometry books...moving two points closer
together in one plane does not increase their separation in another
plane. Start by reviewing the "distance" formula. If you need help,
feel free to holler when you are not too busy saying that we were just
repeating YOUR point. You know, the one that was presented by others earlier. Then you can take on the subject of time relationships between postings and you can further explain how your objections in the form of repeating fundamental issues outside the scope of the use of EQ made after the concepts are fundamentally posited are pertinent. Sound confusing? Now you get a fpeek into the pertinence as to the context of your post.

And you might want to explain how signal alignment can be used to create alignment of the impulse response in the listening position to Klipsch in units such as the KHorn where the acoustic center offsets are so great between drivers using passive crossovers without signal alignment that the first order room reflections arrive earlier than the LF signal. Again, read the Sabine attachment to understand how signal alignment is practically employed..And just in the spirit of further confusing you regarding signal alignment of multiple units in the crossover passband, you might want to peruse the additional attachment made here. You will note that they do not employ an equalizer either.

And should we re quote Tom Danley on the subject? Sure, why not. And then you can claim that he is simply agreeing with you too. Except of course that he doesn't.

"I would say that generally
all filter shapes have an exactly corresponding phase shift (related
to the shape) with an exception being some DSP and all pass filters.

If
one has a loudspeaker who's response rolls off or has a bump, these
undesired "filters" also impose a particular phase shift,
correcting that with an EQ also corrects the phase as most electroacoustic
"things" are "minimum phase" (meaning there
is a specific phase change for a change in amplitude).

If
one wanted to EQ a cancellation notch caused by two different acoustic
paths (one delayed RE the other), one finds that this "filter"
does not have the normal phase change with amplitude and if the
two paths have equal amplitude, the notch is infinitely deep as
well. Using an EQ to put a big peak there does fill in the sides
some but while having little effect on the response, does add a
large amount of phase shift at the EQ point. Since the phase shift
is not corrected by the phase error of the problem filter, it is
left in the signal.

Bottom line is that what you hear so
far as phase usually depends a lot more on what kind of problem
you are trying to fix rather than on what type of circuit was used
to produce the filter. Non minimum phase things like multiple acoustic
paths and so on cannot be fixed with EQ.

One does have to
wonder that since loudspeakers generally do not preserve phase,
how much does that obscure the electronic phase effects.

Cheers,

Tom
Danley"

http://www.prosoundweb.com/live/labbest/pc/pc.php

So if you choose to comment on the use of a discreet EQ to resolve issues...in other words, if you choose to actually read the article and to understand the application of EQ as presented, we (that's editorial for sure!) will wait with great anticipation. Or maybe not.

And then you can make further claims that we were all busy making your point...even before you attempted to make it.

So, to return to the original issue at hand that has not changed: EQ cannot be used to resolve non-minimum phase anomalies. And to attempt to do so not only exacerbates the issue, it is an abuse of the proper use of EQ. That was the original issue, and it remains valid.

nl072_subwooferalign.pdf

[bodcaw boy]

So to sum up...

I post this:

"EQ in the crossovers to make their slope dramatically sharper to reduce combing/lobbing between drivers within the speaker. "

You post not one, not two but four of your typical insulting messages claiming this is utterly wrong and not possible.

Then after demonstrating to you how my original post that "EQ in the crossovers to make their slope dramatically sharper" is very much possible you flip flop and say the following:

"The fact that he already directly addressed the point you think you are newly raising regarding high slope, signal aligned crossovers in a bi-amped configuration must really confuse you.

It was one of the solutions advocated"

In other words what I posted is accurate and is a method to improve the situation which is in direct contradiction to your earlier four messages. One wonders who is confused... the person who posted this in the first place... or the person who argued vehemently and obnoxiously against it for four posts before flip flopping and now confirming it is a solution?

Of course you couldn't come out and simply say you hadn't considered the case of elliptic crossovers which EQ is very much a part of. No, admitting that would not do at all. So the rest of your post is your typical MO of trying to spin that to try and deflect the topic to something away from those four misguided posts of yours.

"**And while you are at it, go back and retake 8th grade geometry. "

Do you really need me to draw this out for you? If you delay one driver it makes signal alignment better in some locations and worse in others.

Put any two points in space (woofer/tweeter) and there are locations that are naturally equidistant from each driver. Those locations are time aligned naturally.

You add delay to one of the drivers and that effectively moves one of those two points in space. You change one of those two points and the locations that are equidistant from each driver (time aligned) changes as well. The points that were naturally aligned no longer are and new points of alignment emerge.

I can't believe you are arguing over this since you already confirmed this:

"Signal alignment does not resolve comb filtering and polar anomalies for all points in a room for non coincident drivers."

Which is simply another way of saying what I am saying.

Shawn

hey shawn,

does this remind of don quixote and the windmills? [8-|]

have a blessed night,

roy

[Deang]

"You cannot use an equalizer to modify the slope of a crossover."

Good grief. Shawn posts the curves to demonstrate the effect and you come back with this?

[Amy]

MAS/DRAGONFYR,

Please cease the personal insults immediately.

Thanks.