Recommendations For Surround Sound Production


Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


300 Club - QQ All-Star
Jul 12, 2007
Here is an interesting document from NARAS, the Recording Academy.

It gives a history of surround sound from their point of view.

Then it gives set-up and production recommendations.

Recommendations For Surround Sound Production

There are parts I disagree with.

From Mr. Dorren's post's, I expect so does he.
(Although different parts.)
Thanks for posting this.
Now I know why my DVD-A player flashed the "No Downmix" when I play a 5.1 DVD-A.
Of course, lots of other good info in the document.

Tom in So Cal
Excellent. Sticky for sure.

Thanks for posting that. A must read for all members - takes some time, though.
It's been up at the Grammy site for a while - a great document.
Almost holy writ.
The whole DVD-A/Stereo/Downmix or not issue is easily avoided as it says here:
PGC Blocks are the way to go. This has the added benefits of only needing a single group instead of 2 groups, and sharing menus as a direct result of that - but even better, it means you can set the disc up to either
A - Autoplay, or
B - load up to a point where pressing the PLAY button starts playback - and you will automatically get the stream your player is set up for.
(Oddly enough, Sonic DAC throws a spec violation error "this stream is unable to be presented in stereo" because of the lack of a downmix.
However, I have always just ignored this & never had a DLT set returned by the factory yet so it has to be a false flag)

Saves ages at authoring time.
Old Quad Guy said:

I'm most interested in trying to mix music into multi-channel. I've remixed 2 albums by NIN (Nine inch nails) from the multi-tracks and created DVD-Audio discs 5.0 mixes. I would be more than happy to send you copies of these discs if you like. I would really like to give a try at mixing what you have and send you back the results in either DTS-CD or DVD-A form.

Old Quad Guy said:
Well, unlike a lot of guys, I don't listen to rock-n-roll so I couldn't really tell you if your rock-n-roll mixes are any good or not beyond a few simple rules I follow:

1. No leads in the surrounds (i e no lead vocal or lead instrument)

2. Backup singers or audio sweetening in surrounds only, or sometimes with slight echoes in the mains (front) but crossed left for right and delayed maybe 8 milliseconds or so. That gives the SQ (or whatever Dolby Pro Logic they call it nowadays) a chance to use its logic to steer towards that dominant channel without swimming in phase errors.

3. Reverb mostly in surrounds. If you are doing 7.1 you can do a somewhat dry Mains mix (everything but suppemental players and backup singers) a Main Surrounds mix of mostly band/orchestra with more reverb than in the mains along with #2 above and mix that plus 90-degrees out left and minus-90 degrees out right and then do a reverb-only track for the 7.1 part and do 45-degrees out left and 45-degrees out right.

You can play around with 135 out left and 135 out right as well for effects and see what you get.


Here is an interesting document from NARAS, the Recording Academy.

It gives a history of surround sound from their point of view.

Then it gives set-up and production recommendations.

Recommendations For Surround Sound Production

There are parts I disagree with.

From Mr. Dorren's post's, I expect so does he.
(Although different parts.)

Out of interest, what parts would those be?
Incidentally, the same document (Recording Academy Producers & Engineers Wing) is linked from the Grammy site too.
I don't agree with all of it either. Have to look through it again but I disagree with "2.2 Professional Mixing Environment". Recommending Mid Field monitors and room treatment when there is no clear consensus on if you can actually tune a room let alone there being some sort of standard to treating rooms. Nearfield monitoring I personally think is not contingent upon room treatment and is therefore more predictable room to room.
Nearfield monitoring can be superb - no doubt about it - but nearly all nearfields are compromise designs with their own pitfalls.
Midfield is indeed the best solution, and there are many first-class facilities out there who would state categorically that you can tune a room.
Please remember we are talking ideal situation here - hence the word "recommendations" - so all monitors should be capable of 20Hz reproduction without being strapped to a Sub for the extension, and definitely not at -3dB 20Hz either. You need to be able to hear it to mix/master it. Sure we can get around this & use seriously well designed nearfields with Subs to extend response but that is a workaround.
Again with nearfields, you would need more constants.
Ported or unported? Minimum of 2' (some would say 3') from a wall. 1" or 2" compression driver or ribbon tweeter for top end? (the sound is dramatically different)
How are we acoustically isolating nearfields? Neoprene or acoustic foam? Stands? Spiked or not, weighted or not - and what with?
I don't know if I should really get into a point by point discussion on why I disagree with some of what you are saying. But I think making a standard that requires you to reach down to 20Hz is getting too greedy with the FR when it's not even needed and in most situations will cause more harm than it will help - exciting modes in rooms smaller than 25 feet while at the same time not giving much of an audible benefit. I find -3dB at 20Hz vs 0dB at 20Hz to be a negligible difference. I find most decoupling and acoustic isolation to not have much of an effect as well - Can any bass trap really trap 20Hz? I guess a big part of a standard for me is that it should be accessible as well as functional.

I think a lot of the differences in sounds and the perceived need for acoustic treatment from room to room is because of the tweeters and how they integrate with the polar pattern of the mid and bass drivers. Now that tweeters are starting to reach into the high frequencies with a wider and more uniform polar pattern the need for room treatment is actually going away. I know this seems counterintuitive and that in theory the more omnidirectional you make a speaker the more modes you excite. But I think you will find that the more area you can cover with a behaved polar pattern (omni, dipole, cardiod) the less the room interferes with the direct sound and the more the direct sound will swamp out first and late reflections. If you keep the polar pattern of the speakers smooth, eliminate early reflections where possible, and keep the first reflections of the room to more than around 6-20ms you can effectively blank out the rooms interference with the sound.

Now if what I say holds true - which I think most all of it should - then I think that nearfield monitoring with a FR of at least 30Hz-20kHz might be a hell of a lot more consistent from room to room than midfield monitoring with a mishmash of decoupling and acoustic absorbers which wont translate to the home in anyway.
Interesting points there, and well needing some serious consideration.
as far as 20Hz goes, I confess I routinely use HPF with a minimum of 48dB/Octave slopes unless there is genuinely something down there I need to allow through
as it tightens up the bass response no end by filtering out crap like rumble & unwanted thumps.
Here is an interesting document from NARAS, the Recording Academy.

It gives a history of surround sound from their point of view.

Then it gives set-up and production recommendations.

Recommendations For Surround Sound Production

There are parts I disagree with. Bot

From Mr. Dorren's post's, I expect so does he.
(Although different parts.)
You can get it here.
Neither of the above links are active after all this time.
I found the document here:

Now this is twenty years old, but I'm reviving the thread in light of the discussions around sound quality perceptions of the Deep Purple Atmos mix.

When I was setting up my first 5.1 listening space, I found great value in the guidelines, especially regarding the reference loudness level when evaluating music at the level that it was mixed at in the control room, not necessarily the volume I or my housemates are comfortable with in a given listening situation.

If someone is voting in a poll when listening to heavy rock content at 72 dB, he obviously is not hearing the same mix as someone listening at 85 or 90.

Rule of thumb for loudness: 3 dB is quite apparent, 10 dB doubles or halves perceived volume for human ears.



3.5 Reference Listening Level

The recommended reference listening level for surround sound production is in the range of 79 to 85dB C-weighted.

However, it is important to check mixes at varying levels, from very soft (as low as 40dB) to quite loud (not to exceed 92dB, however, and only for short periods of time).

I've attached a 13-page pdf of Chapter 3: Surround Sound Monitoring In The Professional Mixing Environment


  • Surround mix recommendations.pdf
    242.7 KB · Views: 0
Last edited:
The surround sound mixer should avoid mixing for one "sweet spot"
I'm not sure how I feel about this. I feel that a mix wouldn't really have much rear activity if it wasn't targeting SOME sort of sweet spot, because on some systems elements get completely lost if you DON'T sit in the sweet spot. Also, I've not released something in a while so this may not be evident but I'm a huge fan of non-standard placements provided that it works for the song. The problem with non-standard panning and placements is that they tend to ONLY work with a well-defined sweet spot.
Work should be evaluated on two or more distinctly different playback systems
I think the more distinct, the better. This is something I absolutely agree with. But I want to emphasize that one playback system that you should always evaluate on is a low-end surround system. Not a trash-can surround system, like one of those HTIB bundles, but a low-end surround system. I personally would not modify my mix to sound better on satellite speakers, but others may disagree.
To that end, the monitoring system must deliver the full range of audible frequencies, and it must be positioned and calibrated correctly.
...I don't think I'm qualified to comment on this.
I find the "neighborhood-shaking subwoofer" bit quite amusing.
Clubs are becoming an increasingly visible area of multichannel reproduction.
Where??? Forget about the setup, I don't think much club music makes ample use of stereo capabilities...
However, too much reliance on the center channel alone can be problematic due to the fact that the center speaker in many home theater systems is smaller than the main left and right speakers.
Normally I would say no, this can be solved with bass management, but as I have found out through experimentation, some AVRs only bass-manage the fronts, and not the rears or center. So, do not be afraid of putting challenging elements in the center, but be wary of how much is in the center relative to other speakers.
Multichannel bus compression and equalization tools have appeared in recent years which can help the front and rear channels be better integrated into a coherent soundfield.
Multichannel EQ? Yum! Multichannel bus compression? It depends. Sometimes it ends up really mushing up the sound field. Especially limiters, they seem to have more adverse effects in multichannel than stereo.

I don't know why I even rambled about this.