[bpape]

What specifically?

I will be going to the factory tomorrow to do some extensive testing and documentation in our new test room.  I'll be there through next week.  I can certainly try to induce something and then show how it's dealt with and the results.

Bryan

[Nyal Mellor]

If you consider seat-to-seat sound pressure level variability as part of the design process then EQ can be applied extremely effectively. To do this speaker position, subwoofer position and seat position need to be manipulated such that seats interact similarly with the room modes. As an example of how this might go, Jeff Hedback and I worked together on a multi-seat home theater. He did the design, I did the calibration. Its a good example of how to use room correction to remove the negative effect of room modes. Because of how Jeff designed the room in particular how he placed the subwoofers I was able to EQ out 90% of the deleterious effects of room modes. It would have been near impossible to deal with the 25Hz mode without recourse to some major custom made bass trapping. Subwoofer placement would have been another option; if we'd been able to place subs on the back wall (this option was discounted at some point during the design process by the integrator and / or client) then we could have used a mode canceling arrangement of multiple subwoofers and then EQ might not have been needed at all. I did also put in a little midrange / treble EQ on the surround channels to help balance the surround field with the left center and right channels.

Also here's the link to the room correction primer that was referred to earlier in the thread.

[JohnR]

What specifically?

That acoustic treatment removes the nulls. (Without making RT60 too low).

[Tyson]

Both is good - acoustic treatment for the mids/highs and EQ for bass is how it seems to work best, speaking in broad generalities.  Of course, bass can/should be addressed in the acoustic realm if at all possible - ie, OB subs, or multiple dispersed sealed subs. 

[bpape]

That acoustic treatment removes the nulls. (Without making RT60 too low).

Will see what I can do John.

RT60 is not really applicable in small room acoustics.  Yes - we do need to be concerned with overall decay time and a balanced curve.  I believe I said that in my post on the previous page that "yes - it can be done.  The trick is not over deadening the room."

Bryan

[Jeffrey Hedback]

Five years ago I wrote an article with that covers much of this topic.  Geared toward "theater", the same basics apply to any small room & speaker situation.

You may find it of interest.  Bryan Pape did a great job in his answers/comments.

http://www.audioholics.com/education/acoustics-principles/twenty-questions-toward-a-correct-room

As obtuse as it may be (and as Nyal so very well covers in his Blog of our joint project), it comes down to minimum phase and non-minimum phase.  You can EQ minimum phase parts of the response (and not the non-minimum phase).  From this point you have to "trust" the knowledge/intent of the people writing the DRC algorithm.  They are not all the same.  Some are designed for general consumer use but fall very short of audiophile standards.

Nyal and I have privately chatted many times of the "day in which we live" where we are on the cusp of rapidly improving DRC options.

[JohnR]

RT60 is not really applicable in small room acoustics.

OK, T20 or T30 then. According to Nyal and Jeff, the lower limit is around 200 ms.

[bpape]

Will see what I can do.  The first goal obviously is to set up seating and speakers to avoid having the problems. 

Secondly, the decay time target in a room is not a single number for all frequencies.  Plus the curve range is also dependent on what the usage of the room is - 2 channel, multi-channel, classroom, etc.  It also depends on the volume of the space.

Again, I'm not in any way saying don't use EQ when it's necessary or when you want to tailor the sound more to your liking or to address a shortcoming in a product.  I'm just saying it's not a fix all solution and should be used judiciously, as low as possible in frequency, and only after proper setup and treatment is done.

Bryan

[Doctor Fine]

I get best results from running great sounding mains full bandwidth and unprocessed in a treated room and only touching up the deep bass below them with a digitally assisted pair of subs if it is stereo.  I wouldn't say it works when running mini monitors for mains but anything flat to 50hz would be a candidate for this approach.  And it pays to be very precise about finding where to place your mains and subs.  Even 1/4 inch off will lose quality.  Read some books about placement.

Subs generate huge sound waves where the modes and nodes of most rooms  are so bad that the very idea of "quality" is laughable...  We are not talking about quality here if the response is already off by plus/minus 20dB!  So with subs digital notch filtering, and anything else you have at your disposal is necessary to fight how horrible the room is.  If you can treat the corners with huge bass traps then your wife is unusually understanding.  Most hate "even more junk" in THEIR living rooms.  So digital can save the day for subwoofer aficianados.

Anyway I like to leave all processing OFF for the mains.  My reasoning is that I have yet to hear a truly wide bandwidth reference speaker that benefitted from digi-itis.  Processing, no matter how sophisticated seems to take the natural clarity away from a great speaker and replace it with grit.  You may think it improves things on a MidFi speaker but a truly HiFi speaker will say "No."

When I run multi-channel surround I favor 7 or 9 full range speakers and NO crossover point and no Audessey or whatever.  I only fiddle with volume and distance calculations for each speaker.  I turn it on Direct to increase the ability of the receiver to handle its job decoding the original signal.  Every time I have asked Audessey to beat me at my game by working on speaker delay and EQ it turns out to sound much grittier and lose spatial definition.  Running Direct seems to require minimum extra calculations and really opens up the separation.

Now this is just MY experience.  I could be wrong.  I was pretty impressed with the last Lexicon Pre-Pro I played with as it seemed to have a lot more firepower digitally than any receiver but I was not able to take it all the way in an A/B comparison because it did not belong to me.  Same with the Meridian stuff I have played with.

But with all my earlier recievers like Sony ES, Marantz THX certified and now my new Onkyo all of which ran $2000 for the receiver the results were pretty consistent.   And my advice may not be useful on speakers that demand a lot of correction due to frequency response problems...  My advice is to stay away from spending money on speakers you won't keep long term.  Spend more, keep longer.

Use good to better speakers that are full bandwidth, fix the room and save digital just for subs.  And don't ask a home theater receiver to do ANYTHING but decode the Direct signal and set distance and speaker balance.

Your results may totally disagree with this.  I am just saying what I have noticed FWIW.

[JohnR]

Will see what I can do.

Any luck?

Again, I'm not in any way saying don't use EQ when it's necessary or when you want to tailor the sound more to your liking or to address a shortcoming in a product.  I'm just saying it's not a fix all solution and should be used judiciously, as low as possible in frequency, and only after proper setup and treatment is done.

Why "after"? Don't they go hand in hand? Couldn't one just as well say, see what you can fix with electronic EQ, and then pay the $$ for acoustic treatment when you know the limitations of that (the EQ) approach?

[bpape]

I'll get some of our results from Glenn. We were not specifically guided by this one goal - but rather to achieve a much smoother overall response within reasonable decay times and without the use of any EQ - purely through movement of seating and speakers and adding treatment.  We were also looking specifically at some more targeted, tuned, absorbers which I had designed to do a comparative assessment of the cost vs benefit.

IMO, EQ is definitely an after thing.  Step 1 is to get you and the speakers in the proper position for response and imaging.  Step 2 is to address the frequency response, reflections, and decay time issues.  Then you do whatever EQ is needed, if any.

If you EQ before treatment and you're say working on a peak or null that's related to a boundary interaction and is a phase related cancellation, then you treat that area and eliminate the problem, your EQ is now set to correct a problem that no longer exists.  Even if it does somewhat, likely the Q and intensity will be vastly different - even if the center frequency is the same.  When using EQ, if the center frequency, Q, and amplitude are not all exactly the same, ringing can actually be made worse rather than better.

Bryan

[JohnR]

If you EQ before treatment and you're say working on a peak or null that's related to a boundary interaction and is a phase related cancellation, then you treat that area and eliminate the problem, your EQ is now set to correct a problem that no longer exists.

Yes, of course. So you readjust it. It's not difficult to redo, once you've done it once.

Perhaps the missing element here is measurement. Does one measure first in order to determine how to address perceived problems?

[bpape]

I guess my point is that if you EQ it out, why would you try to address it with treatment?  This would say make the difference between a 2" or a 4-6" reflection panel decision.

Measurement and being able to find what is causing each problem absolutely let's you bring a much more targeted approach to treating a space.

Bryan

[Ethan Winer]

Perhaps the missing element here is measurement. Does one measure first in order to determine how to address perceived problems?

The real missing element is that EQ can reduce peaks (only) where you place the measuring microphone. Move the microphone 6 inches (distance from one ear to the other) and the benefit is gone, and the response is likely even worse.  This doesn't happen with bass traps. So yeah, you definitely need to measure so you can see these effects.

--Ethan

[Don_S]

The real missing element is that EQ can reduce peaks (only) where you place the measuring microphone. Move the microphone 6 inches (distance from one ear to the other) and the benefit is gone, and the response is likely even worse.  This doesn't happen with bass traps. So yeah, you definitely need to measure so you can see these effects.

--Ethan

Ethan,  I strongly disagree that the benefits of EQ only encompass the area less than 6 inches from where the microphone is placed. I have used both acoustic treatment and equalization together for years.  Each has an important role to play. Please don't minimize the benefits of equalization by understating the amount of area that can be influenced in a positive manner.

[Ethan Winer]

Please don't minimize the benefits of equalization by understating the amount of area that can be influenced in a positive manner.

I actually measured this and wrote it up as an article on my company's web site:

Audyssey Report

This one goes into even more detail, showing the enormous change in LF response over a distance of only four inches:

EQ Versus Bass Traps

I will concur there is one situation where EQ can help generally around most of the room. This applies only to the one or two worst, lowest frequency, modal peaks. If you reduce them by about half as much as measured, it can help to reduce bass boom in rooms where that's the main problem. I mention this in the article above. But this is useful only for ringing frequencies well below 100 Hz, which in many rooms is not the main problem. And the ringing is not improved, only reduced in level.

--Ethan

[Don_S]

How is reducing the level NOT an improvement?  Acoustic treatment is not 100% either.

[Ethan Winer]

How is reducing the level NOT an improvement?  Acoustic treatment is not 100% either.

It is an improvement, but it's not a solution as is often claimed. I don't say never use EQ. I only point out the limitations of EQ. If you read my Audyssey Report you'll see that I use the one-band cut-only EQ built into my subwoofer to reduce 40 Hz by a couple of dB. But I'd never use EQ above 60 to 80 Hz or so, because that's where the highly positional effects are worst. This (shorter) article shows that even as low as 42 Hz the response changes 3 dB over a span of four inches, and a peak at 71 Hz becomes a null the same distance away with a level difference of 14 dB:

A common-sense explanation of audiophile beliefs

How can EQ target the 71 Hz peak without making the null four inches away even worse? It can't, and that's my point. This doesn't happen with bass traps, and that's also my point. Likewise for the 20 dB nulls on both sides of the 92 Hz marker. EQ cannot improve those, and bass traps can. That's all I'm sayin'!

--Ethan

[RCduck7]

I presume room correction for stereo amps are often used through a progam on the computer if i am right. I guess becausse there is only a very small percentage of stereo integrated amps (or pre amps) that has some sort of room correction on board. Does someone here use an integrated amp that has room correction without the need from a computer program? So far i have read that many had good results with Lyngdorf digital class-D amps with integrated room correction.

[JohnR]

The real missing element is that EQ can reduce peaks (only) where you place the measuring microphone. Move the microphone 6 inches (distance from one ear to the other) and the benefit is gone, and the response is likely even worse. 

Yes of course, which is why you measure in a number of locations in the desired listening area and average. Is there something wrong with that approach?

Just looking at the articles you linked, there was a graph showing response with 17 "mondo" traps installed - I assume those are bass traps. It looks terrible, even with all the traps - why don't you use EQ to flatten it out? (Or did I miss something?)