Speaker placement vs room EQ

[fbirch]

As a newbie to room correction, I believe the preferred approach is to experiment with speaker placement to minimize modes (typically, low frequency) at the listening position and then use EQ to clean up whatever modes remain.

Let’s say you have limited freedom of speaker repositioning, and you’ve got a large positive mode at a specific frequency for a given speaker placement. Is there anything suboptimal about just going straight to EQ to get rid of the mode, skipping the step of experimenting with placement? Will you end up with a frequency response that is less good in some way compared to what you'd get by first moving the speakers around to minimize modes before applying EQ?

 

[Curvature]

Optimal setups are hard to come by. Don't worry about it and use EQ, especially if you don't have the freedom to move speakers around as much as you'd like.

 

[staticV3]

I believe the preferred approach is to experiment with speaker placement to minimize modes (typically, low frequency) at the listening position and then use EQ to clean up whatever modes remain.

Main speakers should mainly be placed for good performance above Schröder (e.g. equilateral triangle, reference axis at ear height, toe in/out to taste), with one exception being to follow the manufacturer's advice regarding SBIR.

Only subwoofers should be placed primarily with room modes in mind.

Let’s say you have limited freedom of speaker repositioning, and you’ve got a large positive mode at a specific frequency for a given speaker placement. Is there anything suboptimal about just going straight to EQ to get rid of the mode

Two potential issues:

1. If the listening position varies significantly, then a constructive mode in one place may be destructive in another.
Cut the response peak in one and you may introduce a deep trough in the other.

2. Response below Schröder should be measured with both speakers playing, since most audio content is Mono there and the bass response is likely quite different between one speaker playing vs both.

An example:

 

[fbirch]

Thanks for the reply. I've ordered some full range floor standers, and will hopefully be able to use them without subs. I can pull them out from the front wall quite a bit for listening sessions, but I can't leave them too far out as a permanent position. So if I end up with the same mode I measured with my sub close to the wall (68 Hz), I'll have two choices: (1) Find a good speaker location for listening and push them back when not listening, or (2) Pick one spot and use EQ

 

[holbob]

Main speakers should mainly be placed for good performance above Schröder (e.g. equilateral triangle, reference axis at ear height, toe in/out to taste), with one exception being to follow the manufacturer's advice regarding SBIR.

Only subwoofers should be placed primarily with room modes in mind.


Two potential issues:

1. If the listening position varies significantly, then a constructive mode in one place may be destructive in another.
Cut the response peak in one and you may introduce a deep trough in the other.

2. Response below Schröder should be measured with both speakers playing, since most audio content is Mono there and the bass response is likely quite different between one speaker playing vs both.

An example:
View attachment 454621

Do you mean below Schroeder we should use eq for both speakers together rather than eq them individually. Bit of an issue for Dirac and other room correction software.

 

[sigbergaudio]

Dragging them out from the wall may bring other problems.

You can eq down the peak at 68hz, but you may find that you shouldn't remove it altogether. How high is the peak (dB)?

 

[staticV3]

Do you mean below Schroeder we should use eq for both speakers together rather than eq them individually.

Yes. At low frequencies, the sum of all contributing sound sources matters.

Measure and correct each individually and the complex sum of all will likely be poor.

This holds true with basic PEQ, not so much with solutions like MSO, Dirac Bass Control, Dirac ART etc.

 

[sigbergaudio]

Both might work, but if you EQ individually you need to measure the sum afterwards to see that the result was actually as expected.

 

[fbirch]

Dragging them out from the wall may bring other problems.

You can eq down the peak at 68hz, but you may find that you shouldn't remove it altogether. How high is the peak (dB)?

I measured it with a pair of small bookshelf speakers and a single modest sub close to the front wall. The 68 Hz peak was around 5 or 6 dB if I'm remembering it correctly. I tried pulling the sub away from the wall by a couple feet and the peak was reduced quite a bit but still present. I will soon be replacing that speaker arrangement with two full-range floorstanders that I believe will generate substantially more bass than my one modest sub. So I'm expecting a bigger mode, but I realize it could be totally different with two bass-generating sources vs one.

 

[sigbergaudio]

I measured it with a pair of small bookshelf speakers and a single modest sub close to the front wall. The 68 Hz peak was around 5 or 6 dB if I'm remembering it correctly. I tried pulling the sub away from the wall by a couple feet and the peak was reduced quite a bit but still present. I will soon be replacing that speaker arrangement with two full-range floorstanders that I believe will generate substantially more bass than my one modest sub. So I'm expecting a bigger mode, but I realize it could be totally different with two bass-generating sources vs one.

5-6dB isn't that much. If you get something similar, try reducing with around 3dB.

 

[dualazmak]

As a newbie to room correction, I believe the preferred approach is to experiment with speaker placement to minimize modes (typically, low frequency) at the listening position and then use EQ to clean up whatever modes remain.

Let’s say you have limited freedom of speaker repositioning, and you’ve got a large positive mode at a specific frequency for a given speaker placement. Is there anything suboptimal about just going straight to EQ to get rid of the mode, skipping the step of experimenting with placement? Will you end up with a frequency response that is less good in some way compared to what you'd get by first moving the speakers around to minimize modes before applying EQ?

At least I myself believe that you would "first" test room acoustic treatments, then you may proceed, only if needed, to some EQ(s).
Just for your reference and interest, please visit my below posts on my project thread.

- Identification of sound reflecting plane/wall by strong excitation of SP unit and room acoustics: #498

- Not only the precision (0.1 msec level) time alignment over all the SP drivers but also SP facing directions and sound-deadening space behind the SPs plus behind our listening position would be critically important for effective (perfect?) disappearance of speakers: #687

- Perfect (0.1 msec precision) time alignment of all the SP drivers greatly contributes to amazing disappearance of SPs, tightness and cleanliness of the sound, and superior 3D sound stage: #520

In case if you have difficulties in having sound deadening large spaces (or more preferably adjacent open rooms) in the rear of your SPs as well as behind your listening position, I would like to suggest trying some relatively large sound deadening (sound absorbance) materials, like sponge mattresses; if you would find they are effective, then you can proceed to more expensive and esthetically acceptable such treatments.

In some cases (like mine), the acoustic treatment(s) on the floor and side wall(s) (including thick curtains) as well as the ceiling material would be also important.

In any way, there should be no perfect "way-to-go", and we need to settle ourselves on acceptable "compromise" of room treatments and EQs if needed; after getting that, you would please enjoy listening to music, not listening analytically to the sound.

Our ordinary home listening room environments would never be quasi-anechoic (or should not be anechoic), and our furniture alignments would never be strictly symmetrical...

 

[Arindal]

you’ve got a large positive mode at a specific frequency for a given speaker placement. Is there anything suboptimal about just going straight to EQ to get rid of the mode, skipping the step of experimenting with placement?

According to my experience with a number of rooms, this might indeed bring problems, at least under certain conditions:

- if the mode shows signs of audible sustain, boominess, bloated notes (try picked double bass!), muddy, overly voluminous bass even after it has been corrected by EQ
- if the mode shows any signs of being unevenly excited by events of different transient behavior. That is particularly the case when long, standing notes of that frequency band will cause boominess and bloated bass, while transient, ´tight´ impulsive event will not audibly excite the mode.
- mode behavior dependent on dynamics, non-linear excitement of the mode depending on level. In most of cases this leads to an impression of bass-rich material sounding ´thin´, lacking bass after EQ has been applied, while at higher SPLs the bass shows a tendency of dominance, boominess, bloated frequency bands

If any of these three might apply, there is high risk the bass cannot be corrected solely by PEQ to perfection.