Originally Posted by noah katz
Seems on point to me; please share!
OK, well there are a few details you would need to PM for but I'll post the bulk of it if you're interested.
Firstly, here are some comments I got from kevinzoe privately about the effects of diffusion in his music room, as while there are some things about diffuser use that could be done by measurement or calculation (such as using diffusers at early reflection points or to densify the late arriving field - see the ETC thread for those), much of their use will have to be guided by your goals and subjective assessments.That's almost embarrassing to say as I tend toward a fairly objectivist-heavy view. *shrug*
Originally Posted by kevinzoe
* bare reflections on the side walls provided the best/most mid/high freq energy and aliveness quality but reduced some musical details
* absorption on the side wall 1st reflection points offered the best solution for hearing the most musical details but at the expense of a narrower sound stage and uneasy fealing that the volume had to be turned way up to make up for the lost mid/high freq energy.
* latteral diffusion in the horizontal dimension seemed to offer the best of reflective and absorption by preserving the mid/high freq energy levels and ability to hear the musical details, although just slight behind what the absorption offered for musical detail.
One area I will mention that is pretty easy to calculate is use of diffusers to control flutter echo. I've read that you have to treat at least 20% of a surface for that, and I don't think there would be a significant difference in that value regardless of what type of treatment was used. Also you want to keep the treatment on non-opposite surfaces (so 20% on one wall and 20% of the opposite wall in areas not directly opposed to the first 20%). This is actually one of the reasons behind the diffusion I am putting into my HT, though it is probably unnecessary (for that purpose) as the flutter echo isn't really audible at the listening position. I just want to make the room have a nice, comfortable, tuneful ambiance everywhere - though the same diffusers will also help to densify the late arriving field as well as providing my diffuse lateral late arrivals for envelopment.
So, to determine how much diffusion to deploy, it seems logical to me to start with what we want the diffusion to do. This is why I provided kevinzoe's comments, as they relate directly to what you want the diffusion for.
The main things that diffusion is supposed to do (in a listening environment) is give a sensation that the room is larger than it is (questionably necessary in HT since the surrounds are supposed to reproduce the spatial cues anyway) and to provide a sense of spacious envelopment (which I see as very desirable in an HT). The envelopment is said to mostly happen with lateral, late-arriving diffuse reflections -- so that is what we need to create.
Put more technically, you want early (20ms after direct) reflection control (-10dB in multichannel industry standards, -20dB or more for typical studio control room) verified (or created) using the impulse/ETC (check the thread on that for more information) for LCR. After that you want the reflections seen during the decay to be "dense" rather than "sparse". Also, you need enough energy to be available later on to cause the envelopment feeling. Additionally, if you run RPG's room optimizer software, it recommends absorption at first reflection points of LCR, and diffusion at all first reflection points of surrounds. That last part seems like a good start which is one of things leading me to the setup I'm working on (though front wall diffusers are not happening). Quick note, whenever I say reflection I mean from any surface including ceiling, front wall, floor if not carpeted, coffee table, whatever.
OK so now how do you translate that into a useful design for your theatre room? Well, you start with translating that time into distance and make a few estimates, plot lines in CAD, whatever. The trick is that sound moves about 1.1 ft per millisecond. So you can determine the time taken by the direct sound, and look at the path lengths from the mirror points and determine the difference to know whether or not they need to be controlled in some specific way.
If your room is wide enough to provide the 20ms gap between direct and reflection, then definitely throw diffusers at the lateral reflection points! And if you have front speakers with great dispersion characteristics you probably want to use diffusers on the sides regardless of delay.
Now at non-first-reflection points? This is where you get to play. Try diffusers to the sides of the listening positions above your absorber panels for sure. Remember you have 2D units as well as 1D units in two different orientations available as options. In RFZ studio control rooms, the object is to return a specific amount of diffuse energy laterally to the listeners right at 20ms, so they actually would use flat deflection panels to create paths that are direct+18ft to the listener including the lateral diffusers as the end of the path. I actually toyed with the idea of doing this in my theatre but decided it wasn't worth doing until I already had all my other acoustical issues worked out. But even if you're not trying to return a lot of energy right AT 20ms, you should be making sure there are some of those direct+18ft paths without too many bounces to make sure there is still enough energy in them to create an audible effect. The real trick here is energy management, and to do it you have to have energy to manage - which is why localhost refers to it as "precious". And remember that diffusers have absorption too!
So there are a lot of factors in determining exactly "how much" diffusion to put into a HT. It's not quite like needing a certain amount of absorption to reach a certain decay time. It encompasses factors like orientation in the room, speaker and listener positions, speaker directivity, current acoustic treatments. And even more PITA, personal preferences. What is your 2-channel/multichannel balance, or at least where would you like to bias your setup? Do you prefer "more accurate" or "more aliveness and expansion"? Are there other reasons you want the room to feel "more dead" or "more alive"?
Hopefully that explanation is a bit useful. It gets complicated quickly because you can't really talk about diffusion without talking about the overall acoustical environment, and you get into subjectives fairly quickly. I don't claim expert status nor to have a final answer, because everyone could have their own idea of how best to deploy diffusers - and many of them would be right (for them).