X-Curve and Levels

[Gordon Hookailo]

Hey all you post mixers...

I've been mixing to picture for about 20 years.

I have some questions designed to get a better understanding of what is actually happening in different sized mixing and playback environments…

1) The reason for the X curve…

My understanding of the reason for this, is that in a large room, the high end builds up because of an additive effect from the reverb in the room, thereby necessitating a roll off of the top on the an RTA to keep the “apparent” top end sounding correct for the human listener. We did this for concerts back when I did music, and I would like to know if this is the same reason for the X Curve or is there is another reason.

2) Why my mid field mixing stage translated perfectly to tuned large theaters here in LA with no x-curve at 85db?…

I had a stage outfitted with M&K speakers. While we had to do lots of tuning for the low end, we didn’t touch the top end. The speakers were about 10 feet from the console and the room level was set to 85db spl.

I had Dolby confirm that my tuning was on, and test played back my mixes on large stages and theaters around town and everything sounded just right. They also sound great on home DVD.

Why did my mixes sound just right in both large and small rooms???


Assumtion: If indeed the reason for the X-curve is to compensate for the acoustic build up of high end in a big room, wouldn’t something mixed in either (big with X or small without) translate? I know people say it doesn’t, but why. The only reason I can think of would be that the X curve is there for other reasons…?

And if the X doesn’t translate to a small room, then why did my non X mixes sound just right in large rooms?

3) Apparent levels, how do they translate from large to small or the other way…?

I was set up to mix at 85db Spl and it again, it translated well (even with an SPL meter!) from my mid size stage to large theatrical playback.

However, I now have a smaller mix suite and I have noticed that monitoring at 85 sounds loud now to what I am used to, and I need to bring the mastering levels up if it is to sound as loud in a big room…

Question:

Does mixing in a smaller room at the same ref levels as larger rooms sound louder even though the SPL meters read the same?

Thanks to all for helping me and I’m sure many others get a good handle on these issues.

[Serge Perron]

1) The reason for the X curve… My understanding of the reason for this, is that in a large room, the high end builds up because of an additive effect from the reverb in the room, thereby necessitating a roll off of the top on the an RTA to keep the “apparent” top end sounding correct for the human listener. We did this for concerts back when I did music, and I would like to know if this is the same reason for the X Curve or is there is another reason.

The reason is psychoacoustical. Program known to be well-balanced played on speakers with essentially flat EQ in large rooms is perceived as being too bright. The x curve compensates for that phenom. There is no high end build up in large rooms. Reverb generally decays more quickly at high frequencies than low frequencies in large theaters.

Does mixing in a smaller room at the same ref levels as larger rooms sound louder even though the SPL meters read the same?.

Yes, but this is not as well quantified as the x curve. There has been some informal research mentioned in the past on DUC. I don't know if anything has since been published on this subject. Please see this thread http://duc.digidesign.com/showthread...ighlight=curve and the room cal sticky.

Serge

[Gordon Hookailo]

Originally Posted by Serge Perron

The reason is psychoacoustical. Program known to be well-balanced played on speakers with essentially flat EQ in large rooms is perceived as being too bright. The x curve compensates for that phenom. There is no high end build up in large rooms. Reverb generally decays more quickly at high frequencies than low frequencies in large theaters.
Serge

Then I must ask what is the cause of the high end sounding louder in large rooms. You have confirmed where I thought the X curve came from, same as when I tuned concert systems I used a similar curve or the top end would "seem" to loud. But in a small room flat seems to work. So if it isn't some kind of reverb build up what is it? It has to be something. Anyone know why this happens?

[Gordon Hookailo]

Maybe the reason why things seem softer in a large room at the same pink noise measured SPL, has to do with how the transients are integrated in the large rooms reverb field since we are hearing way less direct signal from the speakers. This seems reasonable to me. You can ever try this in an IR reverb system. The transients usually (depending upon what the room is made of) sound as sharp in a larger room where the direct sound from the speakers is usually less than in a smaller room.

I'll bet that the interference patterns from the large room (both early and late reflections) act as kind of non linear system, or a system that has more damping when it comes to transients.

Anyone?

[Branko]

Serge,

The reason is not psycho acoustical, it is pure physics. Speaker response in a reverberant field is different from the one at 1m on-axis.
There is a good thread on this:

Loudness and room size

You'll find many answers in this thread.
Also, take a good look at Marti's Room calibration thread on top of Post production forum page.
Hope this helps
Branko

[Serge Perron]

Serge,

The reason is not psycho acoustical, it is pure physics. Speaker response in a reverberant field is different from the one at 1m on-axis.

With all due respect, derivation of the x curve is absolutely tied to psycho-acoustics. Our ability to distinguish direct from reverberant sound is a factor, as is the precedence effect, localization, integration time and other hearing traits not shared by microphones. Of course, much of what we hear can be explained by the physical properties of the room and playback system.

There is a good thread on this:

Loudness and room size

You'll find many answers in this thread.
Also, take a good look at Marti's Room calibration thread on top of Post production forum page.
Hope this helps
Branko

Those are the same threads I pointed to.

Serge

[Branko]

Serge,

if this statement:

derivation of the x curve is absolutely tied to psycho-acoustics

is true,

what would be the measured response of an X-curve equilized speaker at 1m, on axis?
Will it be the same as 7-10m away from the screen?
If yes, why? If not, why not?

Sorry for not looking at the list of links you mentioned in your post.
Branko

[Serge Perron]

if this statement: derivation of the x curve is absolutely tied to psycho-acoustics

is true,

what would be the measured response of an X-curve equilized speaker at 1m, on axis?
Will it be the same as 7-10m away from the screen?
If yes, why? If not, why not?

Branko,

Good question. The answer depends on the measurement technique. If steady-state, frequency domain measurements are made using pink noise as the test signal, then the distant measurement will be a measurement of the speaker and room response combined, unlike the nearfield measurement. If a time domain measurement method is used, such as FFT using an impulse test signal, the room has negligible influence in the measurement and so the distant speaker response will be an almost linearly attenuated version of the nearfield response. There will be some minor high frequency roll-off due to air absorption, but not enough to explain the x curve. I think our ears respond more like FFT analyzers than RTA in that they can differentiate direct from reverberant sound.

Now, from past reading on the subject, I had the impression that psycho-acoustics played a role in the x curve. I'm hardly an authority, but I have found several articles online that touch upon this. Here are links to two:

http://www.hps4000.com/pages/general...al_x_curve.pdf

http://www.micasamm.com/publications/surround_0100b.htm

Serge