Target Room Response/Equalization Article

amirm · Mar 20, 2015

Hi guys. As is usually the case, I get my inspiration for articles from online discussions. This one on room equalization is no different. People insist that for movies at least, there is a magical curve that once you match it, you are hearing what was heard during the production of the movie. This is wrong and I explain that in my article that came out in Widescreen Review Magazine a couple of months ago: there is no there in audio: http://www.madronadigital.com/Library/Room Equalization/Room Equalization.html

This is slightly revised and edited version. So please let me know if you see typos and such.

Also, I have written a much deeper dive into the "X curve" cinema target curve as a follow up to this article which should come out in the next issue of Widescreen Review magazine.

BTW, the article is just as applicable to music sound so I hope you read it if movie sound is of no interest to you

.

dallasjustice · Mar 20, 2015

Very interesting. Thanks. I wonder if B&W intentionally engineered their speakers with this -5db dip centered at 2khz. Is this what is known as the "BBC Dip?"

amirm · Mar 20, 2015

They did it intentionally because it looks really nice

. I doubt that it was done because they tried it both ways and that was the preferred sound.

marty · Mar 20, 2015

I was really stunned at the amount of low frequency roll-off show for a typical AV set-up in the article. Anybody else surprised?

dallasjustice · Mar 20, 2015

It would be intersting to see more blind testing between more current correction software.

Target curves are somewhat idiosyncratic to the room/setup and there's no ideal target. However, there's no question a smooth response is preferable to uncorrected as long as the smooth response isn't unnaturally tipped up in the HF. Pretty much all dynamic speakers will show a loss of HF at the seated position because almost all tweeters beam in the really HF. Trying to correct for beaming tweeter is a mistake.

I also think its a mistake to correct for natural bass rolloff. If the speaker can't do it, then it's foolish to force it out of the woofers using DSP.

My favorite target curves tend to follow the natural response of the speakers and subs. I like a gradual -4db slope starting at 1khz until my tweeters naturally rollof. I follow the HF rolloff with my target. I've also found a slight bump in very low frequency is helpful, especially if one likes to use a more aggressive rumble filter. I personally like rumble filters if they can preserve good phase response and still deliver good bass down to 20hz. I've found there's occasionally non-musical LF content, especially in live classical music.

edorr · Mar 20, 2015

dallasjustice said:
It would be intersting to see more blind testing between more current correction software.

Target curves are somewhat idiosyncratic to the room/setup and there's no ideal target. However, there's no question a smooth response is preferable to uncorrected as long as the smooth response isn't unnaturally tipped up in the HF. Pretty much all dynamic speakers will show a loss of HF at the seated position because almost all tweeters beam in the really HF. Trying to correct for beaming tweeter is a mistake.

I also think its a mistake to correct for natural bass rolloff. If the speaker can't do it, then it's foolish to force it out of the woofers using DSP.

My favorite target curves tend to follow the natural response of the speakers and subs. I like a gradual -4db slope starting at 1khz until my tweeters naturally rollof. I follow the HF rolloff with my target. I've also found a slight bump in very low frequency is helpful, especially if one likes to use a more aggressive rumble filter. I personally like rumble filters if they can preserve good phase response and still deliver good bass down to 20hz. I've found there's occasionally non-musical LF content, especially in live classical music.

I found that smoothing out the natural response curve yields the best result. If you have say a 10dB dip somewhere, lift it up say 5dB and create a smooth transition, rather than trying to enforce a flat curve. Don't try to do too much. I also bump up the lows by 5 - dB. I think this compensates for the fact that most recordings are bass lean by nature.

Bruce B · Mar 20, 2015

edorr said:
I found that smoothing out the natural response curve yields the best result. If you have say a 10dB dip somewhere, lift it up say 5dB and create a smooth transition, rather than trying to enforce a flat curve. Don't try to do too much. I also bump up the lows by 5 - dB. I think this compensates for the fact that most recordings are bass lean by nature.

I found in mastering, more than a 3dB correction in any direction yields less than optimal results. If I need more than 3dB correction, I use 2 Parametric EQ's with maybe a Bax shelf and then a peak adjustment.

dallasjustice · Mar 20, 2015

Bruce B said:
I found in mastering, more than a 3dB correction in any direction yields less than optimal results. If I need more than 3dB correction, I use 2 Parametric EQ's with maybe a Bax shelf and then a peak adjustment.

Would you agree there needs to be a standard all mastering engineers accept? Without a baseline, aren't mastering engineers just EQing their room's modulated response to the recording as much or more than the real recording?

Michael.

Bruce B · Mar 20, 2015

dallasjustice said:
Would you agree there needs to be a standard all mastering engineers accept? Without a baseline, aren't mastering engineers just EQing their room's modulated response to the recording as much or more than the real recording?

Michael.

No, because that's why "Professional" mastering engineers spend so much in the room. I want to make damn sure that if I adjust something .5dB, that I'm doing it for the music and not because of some deficiency in the equipment or room!

andromedaaudio · Mar 21, 2015

Nice article

i d love to see some more in room FR graphs of different speakers/brands.

dallasjustice · Mar 21, 2015

Bruce B said:
No, because that's why "Professional" mastering engineers spend so much in the room. I want to make damn sure that if I adjust something .5dB, that I'm doing it for the music and not because of some deficiency in the equipment or room!

I doubt all mastering engineers take care of their speakers/room to the extent others like you do. Erik's point about bass is the most obvious example. If there's a bass peak in the mastering studio, then the EQ applied to bass during mastering will result in lean bass. I bet this is a common problem.

edorr · Mar 21, 2015

Bruce B said:
I found in mastering, more than a 3dB correction in any direction yields less than optimal results. If I need more than 3dB correction, I use 2 Parametric EQ's with maybe a Bax shelf and then a peak adjustment.

I could be totally off base here (i don't know what I am talking about), but I have this theory that the maximum amount of correction that can be applied without making things worse is a function of the playback system, in particular the speakers. If you are correcting a two way system where each driver has to work hard across a broad frequency range, you can't just go in a jack up a specific narrow range in that frequency range by 5db or more without introducing distortion in that driver, especially at higher volumes. A large system 3 way or more system which has to work less hard to produce sound has more headroom to mess within the spectrum, and will not run into distortion problems as quickly.

Bruce B · Mar 21, 2015

dallasjustice said:
I doubt all mastering engineers take care of their speakers/room to the extent others like you do. Erik's point about bass is the most obvious example. If there's a bass peak in the mastering studio, then the EQ applied to bass during mastering will result in lean bass. I bet this is a common problem.

edorr said:
I could be totally off base here (i don't know what I am talking about), but I have this theory that the maximum amount of correction that can be applied without making things worse is a function of the playback system, in particular the speakers. If you are correcting a two way system where each driver has to work hard across a broad frequency range, you can't just go in a jack up a specific narrow range in that frequency range by 5db or more without introducing distortion in that driver, especially at higher volumes. A large system 3 way or more system which has to work less hard to produce sound has more headroom to mess within the spectrum, and will not run into distortion problems as quickly.

Actually I feel the problem is artists that don't go to a good mastering engineer or that can't afford one in the budget. They have the mix engineer "master" their music or go to a "mastering" engineer that has a set of cracked plug-ins working in their bedroom. Usually the mix engineer has a pair of 2-ways on the meter bridge and they don't have an acoustically correct room. It may sound great in their room in the exact position they are seated, but take the music any where else and you will see the deficiencies of the room/system that it was done on. That's why mastering engineers get paid the big bucks (LOL) to make sure it translates across all rooms and playback equipment. The first thing I do when I finish a mastering project is to tell the artist to play the music over as many different situations as you can over the next few days. I also take the music and play it in the car. A car has dreadful acoustics... usually if it sounds good there, it will sound good everywhere!

dallasjustice · Mar 22, 2015

Thanks Bruce.

Another common problem is the HF are boosted too much. I've noticed it from Japanese SHM remasters. I can imagine as I'm listening to these recordings how the mastering studio was setup. I bet these bright recordings are mostly caused by over treated rooms in the HF; so called "dead" rooms. Maybe elderly mastering engineers are to blame?

I wonder how speakers play into this equation as well. I wonder whether work done using speakers with the so called "BBC Dip" adds high mids to the recording since the engineer might need to fill in the energy lost at 2khz by EQing the recording instead of his speakers.

853guy · Mar 22, 2015

dallasjustice said:
Thanks Bruce.

Another common problem is the HF are boosted too much. I've noticed it from Japanese SHM remasters. I can imagine as I'm listening to these recordings how the mastering studio was setup. I bet these bright recordings are mostly caused by over treated rooms in the HF; so called "dead" rooms. Maybe elderly mastering engineers are to blame?

I wonder how speakers play into this equation as well. I wonder whether work done using speakers with the so called "BBC Dip" adds high mids to the recording since the engineer might need to fill in the energy lost at 2khz by EQing the recording instead of his speakers.

Most of the mixing/mastering guys I know don't have perfectly flat rooms or monitors, nor do they claim to possess them. What they have is a set of developed preferences based on many years' worth of experience working with amplitude-related frequencies captured over time. Y'know... music. That is, they know the gear and room is only half the equation. As Bruce alludes to above, yes, there are many poor decisions going on in bedroom "studios". But there are just as many poor decisions being made in well-established pro studios and A&R boardrooms all over the world. Nothing new there. I don't know Bruce personally, but for me the definition of what makes a "good" mixing/mastering engineer is not based on a list of the gear they have, nor is it be based on a response curve of their room. "Let your ears be the guide", indeed.

audioguy · Mar 22, 2015

marty said:
I was really stunned at the amount of low frequency roll-off show for a typical AV set-up in the article. Anybody else surprised?

VERY surprised. Given there are a number of movies on Bluray that have a lot of (measurable) output below their cut off point. Including some into the single digits.

I would also suggest that different room correction systems work very differently. What I mean by that is that I use DiracLive on my music server for 2 channel listening and the default target is a very slow rolloff as you go up the frequency spectrum. I have tried many others including modifications of that curve but still come back to their default.

I then created a target curve in Audyssey Pro virtually identical to the Dirac target curve and it doesn't sound anywhere near as good. Not even close. So more is going on between those 2 products than different target curve choices.

dallasjustice · Mar 22, 2015

I guess if the "professional" system doesn't require accurate reproduction, then why have fancy mastering studios? I know Mastering Engineer Bob Katz feels that his system playback accuracy is key to what he does. That makes sense to me. It's impossible to know what something should sound like if one can't hear it in the first place.

853guy said:
Most of the mixing/mastering guys I know don't have perfectly flat rooms or monitors, nor do they claim to possess them. What they have is a set of developed preferences based on many years' worth of experience working with amplitude-related frequencies captured over time. Y'know... music. That is, they know the gear and room is only half the equation. As Bruce alludes to above, yes, there are many poor decisions going on in bedroom "studios". But there are just as many poor decisions being made in well-established pro studios and A&R boardrooms all over the world. Nothing new there. I don't know Bruce personally, but for me the definition of what makes a "good" mixing/mastering engineer is not based on a list of the gear they have, nor is it be based on a response curve of their room. "Let your ears be the guide", indeed.

853guy · Mar 22, 2015

dallasjustice said:
I guess if the "professional" system doesn't require accurate reproduction, then why have fancy mastering studios? I know Mastering Engineer Bob Katz feels that his system playback accuracy is key to what he does. That makes sense to me. It's impossible to know what something should sound like if one can't hear it in the first place.

I guess the question I would ask then is: If we were to stick someone else, lacking Bob Katz's experience in developing a very specific set of preferences for "what something should sound like", what are the chances they will achieve the same results as Bob, given the "accuracy" of his playback system?

Perhaps a better question might be: If Bob Katz's "accurate" system is key to what he does, then why is his particular "accurate" system not key to what Bruce does or Doug Sax does, given their respective levels of experience?

dallasjustice · Mar 22, 2015

I guess someone could devise a scientific study to figure out whether listeners had a preference for records mastered with an accurate system versus an inaccurate one.

OTOH, you don't need a weatherman to know which way the wind blows.

853guy said:
I guess the question I would ask then is: If we were to stick someone else, lacking Bob Katz's experience in developing a very specific set of preferences for "what something should sound like", what are the chances they will achieve the same results as Bob, given the "accuracy" of his playback system?

Perhaps a better question might be: If Bob Katz's "accurate" system is key to what he does, then why is his particular "accurate" system not key to what Bruce does or Doug Sax does, given their respective levels of experience?

Target Room Response/Equalization Article

Banned

Member Sponsor

Banned

Well-Known Member

Member Sponsor

WBF Founding Member

WBF Founding Member, Pro Audio Production Member

Member Sponsor

WBF Founding Member, Pro Audio Production Member

VIP/Donor

Member Sponsor

WBF Founding Member

WBF Founding Member, Pro Audio Production Member

Member Sponsor

Active Member

WBF Founding Member

Member Sponsor

Active Member

Member Sponsor

Similar threads