Room correction affecting loudspeaker choice?

[general_audiologist]

Hello fellow audioholics!

Every now and then I read loudspeaker reviews and it seems like a lot of what is being reviewed, with audioholics being an exception, is really the frequency response of a loudspeaker. Besides being up to preference, choosing loudspeakers based on their frequency response seems valid only if you aren't using EQ in the system. With advanced room correction software it would seem to me like these small differences often emphasized in reviews (sparkly, muddy, forwards etc comes to mind) are totally irrelevant? If so, how important is loudspeaker choice when you take room correction software into account?

The reason I started pondering this is that using audyssey XT and later XT32 I have noticed a huge impact on frequency response and as importantly, coherency between different loudspeakers (integration between MTM center speaker and large left/right floorstanders and even using loudspeakers of different series with obvious mismatch when not corrected by audyssey).

What I'm hoping for is for Audioholics to do a review/test of a couple of different loudspeakers, all with advanced room correction enabled, so we can see what remains of the differences (obviously dispersion pattern, resonance, break-ups will remain).

I welcome your thoughts and find out if others also would like to Audioholics do a test on this

Edit: added some air to the text

 

[TheWarrior]

Good idea, minus one small detail. The mics take measurements based on the effects of room boundaries, related to speaker placement, as well as listening location. The data would not provide a good basis for comparison due to those inherent inconsistencies from room to room.

 

[general_audiologist]

Important point, but that is why it has to be tested in the same room, just like you would comparing loudspeakers to each other. It would require more work to set up, but I think it's doable.

 

[AcuDefTechGuy]

...choosing loudspeakers based on their frequency response seems valid only if you aren't using EQ in the system...

...how important is loudspeaker choice when you take room correction software into account...

EQ can only do so much, not miracles.

If a speaker is all messed up like +/-10dB FR with rough off-axis response, no EQ is going to fix that.

But I think if you are saying "+/- 3dB to 4dB" FR, then perhaps Audyssey can do a great job fixing.

I believe it is best to choose neutral speakers first. Then allow Audyssey to possibly improve on that. I think the less Audyssey has to correct/fix, the better the results.

I know I've used Audyssey DEQ on the B&W 802 Diamond, which is more like +/-3dB, and they sounded pretty great.

 

[mtrycrafts]

Also, from reading and remembering some of Dr. Floyd Toole's comments, one would need a very good EQ, better than the ones you mentioned probably to do a good job; perhaps some that can do better than 1/6 octave, etc.
Don't know how many correction points those auto EQ's can do.

 

[general_audiologist]

While room correction might not fix the narrow spikes/dips in frequency response mentioned, I think the overall shape of the frequency response curve would be corrected to the point of being largely similar. So it would be interesting to see how much the remaining differences between loudspeakers would matter.

Through your responses I see that such a test would have to take into account different seating positions, seeing how dispersion patterns would remain unaltered by room eq. I therefore suspect you would find the least difference between loudspeakers at the targeted sweet spot (after eq has been applied).

 

[KEW]

I can offer a couple of anecdotes to address your question.
1) I found that using Audyssey XT with a pair of Pioneer BS-22 speakers made definite improvements in the sound quality of the speakers.
2) I recently compared with AudysseyXT 32 vs without AudysseyXT 32 using two pairs of similar speakers (Focal Solo6 vs Twin6) in a treated room. The differences in the resulting SQ were very small and it was unclear which was preferable.

From this "sample of one" no conclusions can be made, but I might offer the following theories to be tested:
1) Quality speakers in a treated room will result in Audyssey not providing much change in the sound quality.
2) As the quality of the speakers and room acoustics decreases, the likelihood of significant benefits from Audyssey increases.

The problem with doing actual useful research on this is you have three factors - Audyssey, room, and speakers. Audyssey is a simple on/off factor (assuming we only test one flavor of Audyssey); however for both room and speakers, we get into the non-quantifiable and subjective evaluations of the quality of the speakers and effectiveness of the room treatments.

An interesting possibility:
Would quality speakers in a room with poor acoustics result in Audyssey making corrections that decreased SQ?

 

[PENG]

EQ can only do so much, not miracles.

If a speaker is all messed up like +/-10dB FR with rough off-axis response, no EQ is going to fix that.

But I think if you are saying "+/- 3dB to 4dB" FR, then perhaps Audyssey can do a great job fixing.

I believe it is best to choose neutral speakers first. Then allow Audyssey to possibly improve on that. I think the less Audyssey has to correct/fix, the better the results.

I know I've used Audyssey DEQ on the B&W 802 Diamond, which is more like +/-3dB, and they sounded pretty great.

Agree, Audyssey won't boost more than a few dB, though it can cut more. I suppose they don't want to boost too much to the point it may overload amps and speakers especially tweeters.

 

[general_audiologist]

From this "sample of one" no conclusions can be made, but I might offer the following theories to be tested:
1) Quality speakers in a treated room will result in Audyssey not providing much change in the sound quality.
2) As the quality of the speakers and room acoustics decreases, the likelihood of significant benefits from Audyssey increases.

Both your points seem plausible to me. What we're really after is the ability of advanced room correction to:

a) Eradicate differences in loudspeaker frequency response by design (most loudspeakers don't seem designed to have a perfectly flat FR).

b) Mask loudspeaker flaws.

When it comes to point b), we can without testing say what properties can and cannot be corrected by altering input signal to the loudspeaker. We (or atleast I) don't know how well the workable properties are corrected and secondly, how much the in-correctable properties of a loudspeaker will affect SQ perception.

 

[PENG]

Both your points seem plausible to me. What we're really after is the ability of advanced room correction to:

a) Eradicate differences in loudspeaker frequency response by design (most loudspeakers don't seem designed to have a perfectly flat FR).

b) Mask loudspeaker flaws.

When it comes to point b), we can without testing say what properties can and cannot be corrected by altering input signal to the loudspeaker. We (or atleast I) don't know how well the workable properties are corrected and secondly, how much the in-correctable properties of a loudspeaker will affect SQ perception.

I have to agree with you on the point that Room EQ systems do have the potential to do more than just "room correction" but also "speakers in room correction". As far as I know Audyssey (probably others too) is wise enough to limit the boost to any frequencies for obvious reasons. Even if they don't further advance their current capabilities, if they would just offer an user enabled feature to open up the limits they could then do more immediately for those who are not limited by their electronics and speakers.

 

[herbu]

choosing loudspeakers based on their frequency response seems valid only if you aren't using EQ in the system.

Said another way, it IS valid to chose speakers w/ some known freq defect, and "fix" it artificially with your electronics?

I'm with ADTG that it is better to electronically fix speaker defects as little as possible. So your point that Audyssey could possibly diminish the defects highlighted by freq graphs is a good one. But wouldn't you still make your speaker choice based on the best unaltered performance possible?

 

[PENG]

Said another way, it IS valid to chose speakers w/ some known freq defect, and "fix" it artificially with your electronics?

I'm with ADTG that it is better to electronically fix speaker defects as little as possible. So your point that Audyssey could possibly diminish the defects highlighted by freq graphs is a good one. But wouldn't you still make your speaker choice based on the best unaltered performance possible?

I also agree with you and ADTG. My comments were based on correcting, albeit only to a point the in room frequency response but not distortions, bad crossovers etc related problems.

 

[general_audiologist]

But wouldn't you still make your speaker choice based on the best unaltered performance possible?

This is what we are forced to do in the current situation! It would of course present little problem if your budget is unlimited. However, reality for most people is that we have to compromise. How much do we spend on AV-entertainment? After we have established a budget, we still have to decide how to distribute it. Buying better loudspeakers could mean less room treatment, smaller/poorer video screen and so on. If buying as great speakers as you can afford was really what you want to do, then I suppose most of us would be stuck with a stereo system

So what if room correction can change how we set up our A/V-budget? Not to mention that loudspeakers present many different characteristics, maybe the one with the flattest or most preferred frequency response performs more poorly than a competitor in other regards? For example:

Loudspeaker A has a less appealing frequency response than loudspeaker B, but loudspeaker A has a superior dispersion pattern. For the sake of the example, let's say loudspeaker B is the all in all preferred choice when listening to them unaltered. Now we introduce room correction and both loudspeaker A and B now presents a similar frequency response. However, dispersion pattern is unaltered, now leaving loudspeaker A with the overall best characteristics with DSP active.

Perhaps this in practice will prove to be untrue, but at least in theory it is possible. And that is why we need to test it!

 

[AcuDefTechGuy]

If Room Correction can turn Bose's FR to +/- 0.5dB, that would great.

 

[herbu]

Buying better loudspeakers could mean less room treatment, smaller/poorer video screen and so on.

Sorry, but I don't see how any of that conflicts w/ my original statement to, "make your speaker choice based on the best unaltered performance possible".

"Possible" includes budget. I get your point that perhaps you can spend less on speakers, and put the savings into other parts of the HT. It sounds like you're chasing some magic algorithm on how to split an overall budget into components to achieve "The Best" system for the money. Good luck with that.

My experience tells me The Best system for the money includes The Best speakers you can squeeze into your budget.

 

[general_audiologist]

Sorry, but I don't see how any of that conflicts w/ my original statement to, "make your speaker choice based on the best unaltered performance possible".

I'm sorry that I didn't catch that properly!

It sounds like you're chasing some magic algorithm on how to split an overall budget into components to achieve "The Best" system for the money. Good luck with that.

Not at all, I'm just trying to explain the possible implications of taking room correction into the consideration when deciding on what loudspeakers to utilize. There are so many different properties of a loudspeaker that it is very likely that each will have different strong and weak points. If some of these properties can be fixed with room correction, would it not make sense to put much greater weight on the remaining/other properties of the loudspeaker when deciding?

For the record I am fully aware that signal processing is not some magic fix. That said, it is a powerful tool and I have personally tried a surround setup using loudspeakers with very obvious differences in high frequency response. Without room correction active there was a very obvious mismatch between the center and left/right speaker. Activating room correction really evened out their frequency response and there was no (obvious) mismatch anymore. If we could test and quantify the effect it might actually affect how we approach loudspeaker choice and I think especially with surround sound setups where transparent sound panning is key.

Edit: I wanted to add a little to my previous post with a theory based on the use of room correction in a surround sound setup:

If we assume that most of the (audio) information is going to be presented through the L/C/R speakers (this is almost always the case?) it would make sense to spend the most money on there three speakers, right? We would still want to assure we have the same tonal balance between all our loudspeakers, otherwise it is going to ruin the experience when sound are transitioning from the surround speakers to the front ones. Therefore it would make sense to ensure all our loudspeakers are of the same series? This will most certainly limit our choice of L/C/R speakers if we are aiming for a 7 or even 11 channel setup. I think this is one example where taking room correction into account might make a difference in loudspeaker choice. If the signal processing allows us to match the tonal balance between different series loudspeakers, then it may very well make sense to buy expensive L/C/R and instead save some on the remaining loudspeakers.

 

[TLS Guy]

I'm sorry that I didn't catch that properly!


Not at all, I'm just trying to explain the possible implications of taking room correction into the consideration when deciding on what loudspeakers to utilize. There are so many different properties of a loudspeaker that it is very likely that each will have different strong and weak points. If some of these properties can be fixed with room correction, would it not make sense to put much greater weight on the remaining/other properties of the loudspeaker when deciding?

For the record I am fully aware that signal processing is not some magic fix. That said, it is a powerful tool and I have personally tried a surround setup using loudspeakers with very obvious differences in high frequency response. Without room correction active there was a very obvious mismatch between the center and left/right speaker. Activating room correction really evened out their frequency response and there was no (obvious) mismatch anymore. If we could test and quantify the effect it might actually affect how we approach loudspeaker choice and I think especially with surround sound setups where transparent sound panning is key.

Edit: I wanted to add a little to my previous post with a theory based on the use of room correction in a surround sound setup:

If we assume that most of the (audio) information is going to be presented through the L/C/R speakers (this is almost always the case?) it would make sense to spend the most money on there three speakers, right? We would still want to assure we have the same tonal balance between all our loudspeakers, otherwise it is going to ruin the experience when sound are transitioning from the surround speakers to the front ones. Therefore it would make sense to ensure all our loudspeakers are of the same series? This will most certainly limit our choice of L/C/R speakers if we are aiming for a 7 or even 11 channel setup. I think this is one example where taking room correction into account might make a difference in loudspeaker choice. If the signal processing allows us to match the tonal balance between different series loudspeakers, then it may very well make sense to buy expensive L/C/R and instead save some on the remaining loudspeakers.

I have been at this a very long time now.

Good speakers get the best out of poor rooms

Bad speakers magnify room problems.

You can not correct a speakers ills with Eq. An aberrant frequency response is but a manifestation of root cause problems such as retained energy. Even Equalizing the frequency response does not make a bad speaker sound right. Far from it.

In a good system in a good room, (and that does not mean a dead room, they are among the worst), auto Eq only mucks it up.

There is no such thing as speakers that are too good.

The lions share of the budget needs to go to speakers.

Two good speakers are far better then 3, 4, 5, 7 or even 11 lesser ones.

 

[herbu]

most of the (audio) information is going to be presented through the L/C/R speakers (this is almost always the case?) it would make sense to spend the most money on there three speakers, right?

Agree.

We would still want to assure we have the same tonal balance between all our loudspeakers, otherwise it is going to ruin the experience when sound are transitioning from the surround speakers to the front ones.

Not so much. My side and rear surrounds do not match my L/C/R. I have never said, "Honey, did you hear how funny that plane/bullet/missile sounded when it went by?".

it may very well make sense to buy expensive L/C/R and instead save some on the remaining loudspeakers

Agree again. In fact, this is what most of us routinely recommend.

I'm still not sure what you propose. Can a speaker reviewer test and report, "This speaker is a bit weak/strong in the XYZ frequencies, but if your room has ABC geometric characteristics, and your AVR has QRS equalization software, this speaker will work fine for you".

Hmmm. Would I buy this speaker over one for the same price w/o the "weak/strong in the XYZ frequencies"? Nope. I think your premise might assume that you have to go to a more expensive speaker to get a better frequency response. Have you seen the AA Monitor for $195/pair?
http://philharmonicaudio.com/aa.html
I'm not convinced price and frequency response share a symbiosis.

 

[general_audiologist]

Not so much. My side and rear surrounds do not match my L/C/R. I have never said, "Honey, did you hear how funny that plane/bullet/missile sounded when it went by?".

Perhaps this is not as big a problem as I would have feared. Have you tried multi-channel music (as in both pre-coded and up-mixed with the likes of Pro Logic II)? Sometimes voices, but mainly instruments are played through the surround speakers, I'm concerned this would sound weird if the sound of these speakers was different.

I'm still not sure what you propose. Can a speaker reviewer test and report, "This speaker is a bit weak/strong in the XYZ frequencies, but if your room has ABC geometric characteristics, and your AVR has QRS equalization software, this speaker will work fine for you".

No, this would be far too complicated! All I'm interested in is a stand alone test where you test a limited number of loudspeakers in the same room, with and without room correction and analyze the results. I'm sure there would be some kind of interesting data to be collected, and if so, perhaps a follow up test is warranted.

Good speakers get the best out of poor rooms

Bad speakers magnify room problems.

Could you please explain how this is possible? As far as I know a good or a poor room acoustically speaking is down to geometry, size and materials? I don't see how choice of loudspeakers are going to alter this, if anything a loudspeaker with better low end extension risks exciting more room modes.

Two good speakers are far better then 3, 4, 5, 7 or even 11 lesser ones.

Well, two loudspeakers make for a very poor surround sound experience Let's face it, you will always be able to afford better loudspeakers going for two instead of seven, the question here is what is a "lesser one". At some point you have to accept that in order to have five, you will have to go for a lesser model than you could have if you only needed two.

Please remember that a the "lesser option" might be a pair of loudspeakers retailing for 800 dollars, and the better option is a 1400 dollar pair. The good old "bang-for-the-buck" ratio declines as we move up the price ladder, so if you take room correction into account, will that ratio remain unaltered, or does it grow or decline even more?

 

[herbu]

Have you tried multi-channel music

Yep. Useful for filling a room w/ even-volume background music, avoiding a loud spot in front of the room and a soft spot at the back. Multi-channel is not for quality listening. I know of no real place where there are guitars/trumpets/violins placed in front, to the left, and behind you and all playing the same thing.

test a limited number of loudspeakers in the same room, with and without room correction and analyze the results. I'm sure there would be some kind of interesting data

Interesting? Perhaps.
Useful? For what?

Sorry. I'm beginning to tire of this discussion. If you're in the market, I hope you can find someone to do all these tests and reports for you, or do them yourself. Please write up a report describing the test parameters, results, show the charts along with your conclusions, and share it with us. Then tell us if you still think it is something that could be doable and profitable for a reviewer.