Perceptual Effects of Room Reflections

[fredoamigo]

The Cleveland orchestra has been doing that for years in the summer.

Not really outdoors for the Cleveland Orchestra

 

[tuga]

The Hollywood Bowl is an interesting concept, but it is still missing the reverb from a large symphony hall:

 

[Bjorn]

"The lows are minimum phase behaviour"?
That's not true at all. And if it was, equalization alone could have been a sufficient tool. And what does "lows" actually mean? Just throwing that out without defining it, is unfortunate. Part of the very lowest frequencies, typically below 60-50 Hz, can be minimum phase behaviour. But in most cases, most of the response in deep bass isn't. And the higher you go in frequency, the less chance it is of being so.

Bass is generally the frequencies between 20 Hz and 250 Hz. The most audible range is actually above 80 Hz, where multiple subwoofer seldom operate. If one desires the best, this range absolutely needs to be treated acoustically. Multiple subwoofers alone can never achieve a great time domain behaviour in the wide bass range and in the most audible one. Besides, multiple subwoofer with units placed behind the listeners tend to lack the slam and impact vs having all the bass coming in front of the listener. Bass always has the best quality when it's coming in front of you IMO.

Bur sure, sometimes compromises needs to be made and here multiple subwoofers like EQ can be an option. As mentioned, treating deep bass (below 60 Hz) with treatment requires covering large surfaces for a really great effect. But we shoudn't falling into believing that something with a major compromise can compare to the best. It can't and anyone who has truly heard a high end treated room knows this well. It's two different worlds.

 

[thewas]

Aren't room mode related peaks in the region below 100 Hz, which are (except wide dips) the most audible bass problems, mostly minimal phase?

 

[dasdoing]

Low frequency room modes are minimum phase behaviors. Correct the FR (with minimum phase filters), the time domain errors are corrected too.
References:

https://www.harman.com/documents/LoudspeakersandRoomsPt2_0.pdf

https://www.harman.com/documents/LoudspeakersandRoomsPt3_0.pdf

View attachment 241842View attachment 241843View attachment 241844

Yes. And since the in-room frequency response predicts perception at low frequencies, he is also talking about "how it sounds".



Speakers + room = a "minimum phase system" at low frequencies. So when you correct the frequency response, you have simultaneously corrected the time-domain response, and vice-versa.

Bass trapping directly affects the time-domain behavior and therefore improves the in-room frequency response.

Multiple subs intelligently distributed directly affects the in-room frequency response and therefore improves the time-domain behavior.

To put it another way, it is the in-room frequency response peaks which take longer to decay into inaudibility. Smooth those peaks (whether by multiple subs or EQ or bass trapping or whatever), and those frequency regions no longer take more time to decay than the rest of the spectrum.

(Toole reports in his book that we perceive the frequency response peak, and not the ringing in and of itself, which implies that fixing the frequency response is fixing the problem. See post #493 above by @NTK.)



Absorbing all the energy in the modes = anechoic conditions = not practical.

Matthew Poes is not the first person to conclude that a distributed multi-sub system is more effective than bass trapping at improving the in-room bass, but his extensive hands-on experience with both techniques qualifies him to comment from experience rather than from theory. I first learned the theoretical side from Earl Geddes in January of 2006.

Of course you can do both if space and resources allow, but most of the improvement will probably come from the distributed multisub system.



Multiple subs make the decay much more uniform across the spectrum and throughout the room, the "throughout the room" aspect being something that EQ of a single sub cannot accomplish for modal effects.

(Disclaimer: I have been manufacturing a distributed multi-sub system since 2006, so you are welcome to discount or dismiss my opinion since I am commercially involved.)

so you guys are saying basicly that bass treatment is "snake oil"?
wouldn't that also make multi-subs useless since EQing just solves anything anyway?

 

[Duke]

Aren't room mode related peaks in the region below 100 Hz, which are (except wide dips) the most audible bass problems, mostly minimal phase?

Yes.

so you guys are saying basicly that bass treatment is "snake oil"?

No; rather, imo bass treatment is less effective than a distributed multi-sub system in the subwoofer region. But there is a limit to how high up in frequency you can use a distributed multi-sub system, and fortunately north of the subwoofer region bass treatment becomes increasingly effective as the wavelengths become shorter.

And if resources permit it's not necessarily either/or. You can use an equalized multisub system with bass trapping.

wouldn't that also make multi-subs useless since EQing just solves anything anyway?

If EQ "solves everything", then it typically does so only within a very limited listening area. Since the room-interaction room peaks and dips occur at different frequencies in different locations throughout the room, EQ which optimizes the response of a single subwoofer for a single listening position is simultaneously making the response worse elsewhere in the room.

A distributed multisub system greatly reduces the frequency response variance from one location to another in the room, so any remaining significant frequency response issues are likely to be global (throughout-the-room) rather than local. In this scenario, EQing such issues improves the frequency response throughout the room, instead of improving it in one location at the expense of the rest of the room.

So if the goal is good bass over a wide range of listening positions, multiple subwoofers + EQ is a very effective combination.

That being said, ime most of the improvement is still attributable to the distributed multisub system. The four small subwoofers I sell as a package come with an amplifier which has a single band of parametric EQ, and to the best of my knowledge most of my customers do not even use the EQ.

 

[srrxr71]

I’ve been hearing the term “overdamped” a lot lately.

As someone who invested a lot in room treatments it’s easy to feel this is not possible.

However I think i’m skirting a line here.

Before I continue I would like to add that no amount of room treatments will get you to an anechoic chamber. Those cost $millions and have to be engineered that way with a lot of treatments.

Having said that when I look at my GRADE report I have peer feet results now except - I actually have RT60 below target between 100-300Hz. What a funny problem to have.

So I know I have over corrected somewhere. At least I know.

That’s the great thing about standards. Hit the standard and stop worrying.

I know now which direction to work. I can move some panels or change to diffusive/absorptive panels.

So there is some truth to this concept of “overdamped”. Trust me it’s not that easy to accomplish. I probably have 40+ panels in here and many just to control the asymmetry of the room. I have to get more creative now to solve a couple of final issues.

Also oh yes a 3rd sub is coming. Now I know how to place them. One is definitely going behind me - I know that much. 4 subs is better but apparently not that much better than 3.

 

[Bjorn]

Aren't room mode related peaks in the region below 100 Hz, which are (except wide dips) the most audible bass problems, mostly minimal phase?

No, they aren't.

 

[thewas]

No, they aren't.

Can you please show some examples or literature which show that?

 

[Duke]

In case anyone is interested, here is Earl Geddes talking about the modal region, and in particular what's going on psychoacoustically in this region. Briefly, he takes the position that the steady-state response tells us pretty much everything we need to know in the modal region because the ear's time-domain resolution is very poor in this region. The video should be cued up to about 4:57, I suggest watching until about 8:22 (at a minimum):

 

[amirm]

so you guys are saying basicly that bass treatment is "snake oil"?
wouldn't that also make multi-subs useless since EQing just solves anything anyway?

No they are not snake oil unless you go by the name "bass trap" as none of the devices sold for that use do anything below 200 Hz or so. You can indeed use acoustic products to fix bass response but it requires heroic effort and tons of expertise. Just throwing this and that absorber in the room does little to tame bass response but can serve to create a dead room at higher frequencies.

In contrast, EQ trivially solves the same problem. Even systems that have extensive amount of acoustic treatments, EQ is mandatory for what is left.

As to multiple subs, the primary reason for that is to produce smooth bass across multiple seats. For one seat, it provides little to no value.

 

[Bjorn]

Can you please show some examples or literature which show that?

It goes against conventional wisdom and measuerements we see conducted in small rooms. Your question therefore should have been the other way around. Can those who claim this, show actual measurements to back it up?

You can easily see this by for instance looking at the measurement's excess group delay. Generally you'll see that most of the deep bass response doesn't have a flat excess group delay, but some areas might be close to being flat. It varies greatly, but you'll almost never see that it's entirely flat.

One could also debate since bass is something we not only hear with our ears but also feel with our bodies, that measurements here would need to in a wider area.

Either way. While the lows in small rooms isn't in general minimum phase behaviour, that doesn't mean multiple subwoofers don't work to flatten the frequency response. Combined with EQ and working with placement, it will and thus it works ok. The EQ applied will also generally make the time domain behaviour look better. But the time domain is really more audible above this region.

 

[Duke]

As to multiple subs, the primary reason for that is to produce smooth bass across multiple seats. For one seat, it provides little to no value.

Ime EQ has limited capability to fill in local nulls in the bass region, whereas a distributed multisub system generally results in smaller and shallower dips which are more amenable to equalization.

 

[Bjorn]

In contrast, EQ trivially solves the same problem. Even systems that have extensive amount of acoustic treatments, EQ is mandatory for what is left.

Maybe shelving to tailor to taste and placement (gain from boundaries or not), but enough treatment will truly give an expectional even frequency response.

Here's the response after about 70% of a room was treated. Don't have the final measurements yet, but they should be better. Other than lifting the lows with shelving, I would say this is fine. And floor bounce isn't avoidable with traditional speakers.

Waterfall before and after.

 

[amirm]

Ime EQ has limited capability to fill in local nulls in the bass region, whereas a distributed multisub system generally results in smaller and shallower dips which are more amenable to equalization.

Dips are not nearly as annoying as peaks. And you can use placement optimization (sub and listening position) to mitigate them. Using multiple subs requires ton more work to optimize unless you just get lucky or have an automated system like JBL Arcos and Dirac multi-sub optimization.

 

[amirm]

Here's the response after about 70% of a room was treated.

That is huge expense and likely creating a dead room.

 

[tuga]

That is huge expense and likely creating a dead room.

Does perceived "dead-sounding" relate to sub-bass and bass frequencies though?
I though it would be from the mid-mids upwars.

 

[tuga]

Dips are not nearly as annoying as peaks.

That is a bit subjective, and not only when the dips are broad.

 

[Bjorn]

That is huge expense and likely creating a dead room.

Quite low expense and not dead at all. Not much mid and high frequency absorption here. Besides, the latter can easily be adjusted to personal taste.

 

[Bjorn]

Does perceived "dead-sounding" relate to sub-bass and bass frequencies though?
I though it would be from the mid-mids upwars.

Right. Low frequency absorption never leads to a dead sounding room. Too much absorption in especially the highs on the other side, can very much do so. Also because of uneven absorption or random misplaced absorption. Which is common with the wrongly belief that RTx measurements are valid in small rooms.

I generally prefer live sounding rooms with little absorption in the highs, a lot of absorption in the lows and late diffuse energy of mids and highs. However, the combination of lively and correct generally can't be achieved if the room is very small or the listener sits up against a wall.