ITU surround specs flawed for film work

[ggegan]

I am a film rerecording mixer with more than a few of decades of experience (check IMDB for my bona fides) who in the last few years has found myself doing a lot of premixing in my small personal ICON studio. When I first began premixing at home, I tried to conform to ITU 5.1 standards for small rooms, but have come to the conclusion that they are fatally flawed for film and TV work, especially for mixing sound FX, due to the problem of phasing when panning sounds to both the front and surround speakers when the speakers are positioned equidistant from the mixing position.

If I am mixing background sounds like wind, traffic or ocean waves that have an element of random noise similar to white, pink or brown noise, panning the effect simultaneously to front and surrounds results in horrible phasing when the speakers are all positioned equidistant. This is almost never a problem on a large dub stage or movie theater, presumably because the speakers are not even close to being equidistant from the listener.

I have abandoned ITU and other specs that require equidistant speakers for small rooms and find it is better to introduce enough delay into the surrounds to eliminate phasing. There is still a degree of phasing that occurs across the front speakers, but I see no solution to this.

I feel very strongly that there should be a reassessment of speaker placement/delay standards for mixing TV and feature film projects in small rooms.

[Sebg]

Gary, do you mean comb filtering? Could you elaborate on what you mean by "phasing"?

[Dange]

There's more in ITU than just the loudspeakers just being equidistant from the listening position.

So just to check, Is the room well treated and the first reflections from every loudspeaker suppressed?

The ISD should be around 15ms and reflections 10dB down (basically no early reflections within 15ms of direct sound)

Also if you follow the ITU standard you need a room that allows you to have the loudspeakers 1m from reflecting boundaries, e.g. walls, ceiling.

[jhbrandt]

Gary,

You should have no phasing problems with that setup no matter how you pan it.
Can you do some REW impulse testing at mix position - one speaker at a time w/o moving the mic and post the response? - and post the mdat file as well.

Are you sure that your speakers are wired in phase? (sorry for the dumb question)
Cheers,
John

[avare]

Originally Posted by ggegan

I feel very strongly that there should be a reassessment of speaker placement/delay standards for mixing TV and feature film projects in small rooms.

Thank you for a well thought out post regarding mulitchannel sound. With regards to your point about standards and cinema sound, cinema sound has a different standard from small rooms. The NARAS recommendations fro multichannel sound describes several different unique enviroinments for multichannel sound, including one for cinema sound. The closest to a standard for cinema sound is the THX recommendations. The NARAS P&E Wing Surround Sound recommendations are here.

I see from your profile that you have worked at Skywalker sound. Lesley Ann Jones is one of the authors of the document.

Please email me for more info.

Andre

[Dange]

Originally Posted by avare

The NARAS P&E Wing Surround Sound recommendations are here.

Just skimming the summary, the recommendations for placement of loudspeakers is the same as ITU.

On the point of delaying signal to loudspeakers it recommends this:

It is absolutely critical that the signal coming from all five main speakers arrive at the mixing position (the sweet spot) at the same time. This is best accomplished by having all five speakers equidistant from the mixing position. If this cannot be achieved because of the physical layout of the room, disparity in arrival time can be corrected with the use of delay.

[ggegan]

On a large dub stage or movie theater it is impossible to have all the speakers equidistant from all audience members. Aside from the fact that the front speakers are placed in a plane, there is just too much variation between the listening positions. There is no way I know of to compensate for this in terms of delaying the front speakers, but there is for the surrounds. According to my understanding, the solution generally used is to introduce enough delay into the surrounds so that for all listening positions, the sound from the surrounds arrives later than from the front speakers, in order to avoid the Haas effect for sounds panned equally to both front and surround speakers localizing the source as coming from the surrounds rather than the front. This would mean that at the sweet spot the delay between front and surrounds would be fairly significant. The amount of delay typically used is somewhere between 20 and 30ms. This creates a very different listening environment from a small mix room that has all speakers placed equidistant from the listening position.

If I introduce a 10ms delay into my surrounds, which is similar to what there would be on a large dub stage at the sweet spot, the phasing goes away and the small mix room sounds the way I would expect it to.

[ggegan]

Originally Posted by jhbrandt

Gary,

You should have no phasing problems with that setup no matter how you pan it.
Can you do some REW impulse testing at mix position - one speaker at a time w/o moving the mic and post the response? - and post the mdat file as well.

Are you sure that your speakers are wired in phase? (sorry for the dumb question)
Cheers,
John

John, speakers are in phase, no offense taken. I'm right in the middle of predubbing a feature that we final mix a week from Monday, so I don't have time to do any testing until I'm done with the project the second week of August.

Everything sounds good and is translating really well, so I don't think there are any huge problems, I'm just really irritated by phasing of some of the background wind effects in my home studio, which aren't phasey on the big stage, but I have pretty much solved that by putting a 10ms delay in the surrounds, which is a solution that removes the phasing without creating any mix problems and also more closely matches the situation on a large dub stage.

I don't think there is anything wrong with my monitor system, I am very meticulous about how I put it together, and like I said, my tracks are translating really well. I just think that ITU specs for speaker placement are flawed for film work because of the unique nature of the sounds we work with. Sound FX are very different from music.

I will definitely do more investigation once I get through this project and send you some files.

[Dange]

Does the phasing occur when the panning is dynamic? I.e an effect is moving from front to rear?

I take it you mean it sounds like a flanger effect. That modulates a comb filter over time up and down in frequency. This is created by modulating the delay of a copy of the original signal and adding it to the original. So you get Output = Original sound + original sound(Δt). A phasor effect is similar

I can then see this being related to untamed early reflections, the reflections acting as a comb filter which shifts as it's panned between speakers.

If not and the sounds are static, then it's odd! The signal has got to be delayed (and the delay modulated) or have it's phase altered somewhere to give the effect you describe. (assuming your head is kept perfectly still!)

[avare]

Originally Posted by ggegan

Everything sounds good and is translating really well, so I don't think there are any huge problems, I'm just really irritated by phasing of some of the background wind effects in my home studio, which aren't phasey on the big stage, but I have pretty much solved that by putting a 10ms delay in the surrounds, which is a solution that removes the phasing without creating any mix problems and also more closely matches the situation on a large dub stage.

Great! So the problem was a (lack of) decorrelation in your space, compared to a dub stage.

Thanks for sharing with us the practical acoustic difference in mixing in a small compared to dub stage studio, studio and a dub stage.

Andre

[jhbrandt]

Gary,

Thanks for that information... I look forward to your tests. I find this very interesting and would like to get my head around it to understand what is going on mathematically. There may be several other things going on that we are not aware of, one of which could be what Dange suggested.

-and as Andre said, "Thanks for sharing with us the practical acoustic difference in mixing in a small compared to dub stage studio, studio and a dub stage."

Cheers,
John