Perceptual Effects of Room Reflections

[amirm]

Does perceived "dead-sounding" relate to sub-bass and bass frequencies though?

No. But classic absorbers don't have a brick wall response that stops at 100 Hz. They go on absorbing to quite high frequencies. Indeed, they are better at absorbing high frequencies than low due to wavelengths being shorter there. Here is a quick simulation of a 4 inch absorber:

This is how your room becomes dead. And look at how ineffective the response is below 100 Hz. This forces you to use a lot more of them resulting in that dead room.

You can use specialized, tuned absorbers that don't have this attribute but they are tricky to design/deploy and require skills well above a hobbyist.

 

[amirm]

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

If they are broad then you can EQ them.

 

[thewas]

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.

Isn't that what I wrote above, that this region is mainly minimum phase?
By the way that was based on observation of my limited measurements which show relatively flat excess phase in that region, that's why I asked you for some examples that show the opposite, here for example of my working place:

 

[thewas]

If they are broad then you can EQ them.

Although in my limited personal experience most of the times such usually doesn't sound really good, probably because the dips aren't minimum phase.

 

[Duke]

Dips are not nearly as annoying as peaks.

Agreed.

And you can use placement optimization (sub and listening position) to mitigate them.

Not always.

Using multiple subs requires ton more work to optimize unless you just get lucky...

That has not been my experience, and I have literally hundreds of customers who use a distributed multi-sub system. The placement of no one sub is critical, and setting the controls has very seldom been challenging.

 

[ernestcarl]

Maybe it’s just that I have gotten used to an over-absorbed room (mid-hi frequencies due to use of 4-inch wedge foam panels — though those aren’t the only ones used), but I don’t find the “too dead” thing a huge problem. It probably helps that the bass decay here is quite substantially controlled in my listening room, and that I’ve left untreated patches/surfaces some through lots of trial and error (objective measurements and subjective listening) so one does not get the feeling of sitting at a “dead spot” across most of the primary listening areas.

Also, some of that feeling of excessive dryness (if unwanted) from plain stereo mixes can be significantly reduced by upmixing to MCH and/or with DSP convolution. Such methods don’t always give the best results, for sure, but I find it better than having no room treatment at all.

 

[thewas]

It goes against conventional wisdom and measuerements we see conducted in small rooms.

By the way Toole also writes similar in his book:

To complete the story, one needs to examine what happens at frequencies below 200 Hz. It should be no surprise that at very low frequencies, the duration of ringing can be such that it becomes an audible extension of notes—bass “boom”—most easily detected in damped, impulsive sounds like kick drums, plucked bass, and so forth. It is reasonable to think, therefore, that in this frequency range, depending on the nature of the program, listeners at certain times may be sensitive to spectral characteristics and, at other times, to temporal characteristics. For now, it is sufficient to say that low-frequency resonances in rooms behave as minimum-phase phenomena, meaning that if there is a prominent “bump” in the frequency response, it is probable that this will be heard as excessive loudness at that frequency and that for transient sounds, there will be bass “boom” at that frequency. Using equalization to reduce the bump also attenuates the ringing so both problems are solved simultaneously
(page 160)
...
At subwoofer frequencies the behavior of room resonances is essentially minimum phase (e.g., Craven and Gerzon, 1992; Genereux, 1992; Rubak and Johansen, 2000), especially for those with amplitude rising above the average spectrum level. This suggests that what we hear can substantially be predicted by steady-state frequency-response measurements if the measurements have adequate frequency resolution to reveal the true nature of the resonances. In minimum-phase systems, the magnitude versus frequency response (henceforth simply “frequency response”) contains enough information to enable the phase response to be computed, and from those two data sets, the transient response can be computed.
(page 200)

 

[srrxr71]

I have a question if anyone knows. For my RT60 calibration i’m using the entire space dimensions. I’m my case 18’ x 35’ x 9’. However my listening space is only half of this.

Which dimensions should I enter?

 

[Bjorn]

By the way Toole also writes similar in his book:

To complete the story, one needs to examine what happens at frequencies below 200 Hz. It should be no surprise that at very low frequencies, the duration of ringing can be such that it becomes an audible extension of notes—bass “boom”—most easily detected in damped, impulsive sounds like kick drums, plucked bass, and so forth. It is reasonable to think, therefore, that in this frequency range, depending on the nature of the program, listeners at certain times may be sensitive to spectral characteristics and, at other times, to temporal characteristics. For now, it is sufficient to say that low-frequency resonances in rooms behave as minimum-phase phenomena, meaning that if there is a prominent “bump” in the frequency response, it is probable that this will be heard as excessive loudness at that frequency and that for transient sounds, there will be bass “boom” at that frequency. Using equalization to reduce the bump also attenuates the ringing so both problems are solved simultaneously
(page 160)
...
At subwoofer frequencies the behavior of room resonances is essentially minimum phase (e.g., Craven and Gerzon, 1992; Genereux, 1992; Rubak and Johansen, 2000), especially for those with amplitude rising above the average spectrum level. This suggests that what we hear can substantially be predicted by steady-state frequency-response measurements if the measurements have adequate frequency resolution to reveal the true nature of the resonances. In minimum-phase systems, the magnitude versus frequency response (henceforth simply “frequency response”) contains enough information to enable the phase response to be computed, and from those two data sets, the transient response can be computed.
(page 200)

You need to look at actual measurements from many rooms.

 

[thewas]

You need to look at actual measurements from many rooms.

I showed you actually a measurement and even some well regarded literature, now its your turn.

 

[srrxr71]

On the topic of overdamping. Is it even possible to overdamp the front wall? All my treatments are focused behind the monitors.

How can corner traps on the front wall every hurt? I see studios hitting all edges behind the monitors. Like ceiling, floors, sides.

The major difference is that they use a nice diffusion panel between the monitors where most of us have a display. Given that I treated behind the TV to avoid any reflections from behind it. The front is smooth and there’s not much I could about that.

The debate is on the “anechoic” path or what they call “space 2”. That could be absorptive or absorptive/diffusive. I guess that would be to taste.

The back is going to be absorptive/diffusive in all cases. Perhaps only diffusive. I guess possibly one could overdamp those back corners if they don’t place diffusive screens on those back corner traps.

 

[Inner Space]

By the way Toole also writes similar in his book ... if there is a prominent “bump” in the frequency response, it is probable that this will be heard as excessive loudness at that frequency and that for transient sounds, there will be bass “boom” at that frequency. Using equalization to reduce the bump also attenuates the ringing so both problems are solved simultaneously ...

Or are both problems merely replaced by a third problem? Suppose you have a four-note bass turnaround that by sheer bad luck hits two room modes on the way down an octave. You hear:

note BOOM note BOOM

So you apply EQ, which drops the timbre-rich bass signal into inaudibility, so that its energy at its fundamental excites the room mode much, much less. Now you hear:

note boom note boom

The overall amplitude is now consistent, but effectively you have fired the bass player and replaced him with a guy blowing quietly across the top of an empty beer bottle for half the notes. Musically, that's a very poor result. You could coin a slogan: "EQ: making your graphs look good and your music sound bad."

The only answer is passive, physical room treatment. Hugely expensive and inconvenient, yes, but you get to ditch the beer bottle and actually hear the notes.

 

[NTK]

So you apply EQ, which drops the timbre-rich bass signal into inaudibility ...

If your EQ strategy to address a bump in the FR is to completely eliminate that frequency band, no wonder your EQ sounds bad

 

[srrxr71]

Or are both problems merely replaced by a third problem? Suppose you have a four-note bass turnaround that by sheer bad luck hits two room modes on the way down an octave. You hear:

note BOOM note BOOM

So you apply EQ, which drops the timbre-rich bass signal into inaudibility, so that its energy at its fundamental excites the room mode much, much less. Now you hear:

note boom note boom

The overall amplitude is now consistent, but effectively you have fired the bass player and replaced him with a guy blowing quietly across the top of an empty beer bottle for half the notes. Musically, that's a very poor result. You could coin a slogan: "EQ: making your graphs look good and your music sound bad."

The only answer is passive, physical room treatment. Hugely expensive and inconvenient, yes, but you get to ditch the beer bottle and actually hear the notes.

If we were to correct for those notes more than half the room would be occupied with treatments.

The math on trying to absorb 50Hz means treatments extending feet out into the room.

Worst case you get:

Note blank note blank

which imho is better than:

Note boom note boom

By far. That is the worst case. You can tune your room with the help of the right material or invite a buddy who plays bass to create test track for you and you can tune by ear.


However I agree 100% with pretty graphs being meaningless. I don’t care how ugly the graph is if it sounds right.

I wish we would report more on our tests and observations than talk theory. We are on the forefront of this on this forum.

I think an application of scientific method wherein we try to replicate other’s experiences and comment honestly is best.

We will all put ourselves “out there” perhaps even be subjected to ridicule (which must be avoided).

Just try a particular technique and report back.

I’m trying high Q filters at the mode frequency. I don’t know yet how much to dip and at what Q factor. I’d imagine if you drop a steep dip at the center frequency of a room mode you still preserve the harmonics. You don’t suddenly “fire the bass player” just because you created a high Q dip.

You don’t fire the bass player even if the speaker can’t reproduce 50Hz even. Your brain creates the fundamentals.

On this it’s better to take out a minimal defect rather than leave it and ruin everything.

 

[Inner Space]

If your EQ strategy to address a bump in the FR is to completely eliminate that frequency band, no wonder your EQ sounds bad

Some folks EQ down 10 - 15dB ... given ELC etc, do you imagine any of the original signal is still audible? All that's left is lower-amplitude room boom.

I don't use EQ to fix room problems - I fix them in the room, not in the box. Far better.

 

[amirm]

Some folks EQ down 10 - 15dB ... given ELC etc, do you imagine any of the original signal is still audible? All that's left is lower-amplitude room boom.

You eq down the peaks. Then listen and if bass is not to your liking, you raise the entire region. This is why proper target curves have a bass boost. The sound with it is rich in bass but not boomy. This is clearly shown in research.

 

[ernestcarl]

However I agree 100% with pretty graphs being meaningless. I don’t care how ugly the graph is if it sounds right.

That probably just means one is only looking at or showing one snapshot angle.


Pretty graphs is maintained at its best at a certain equalized "main listening position" (MLP).

Even with just the simple RTA graphs, if you measure enough mic positions across a small room -- even one that is "treated" -- you're still bound to find "ugliness"

3x RTA averages from different listener points across the couch (too lazy to perform MMM this time around):


Amplitude peaks above the 80 dB here do not sound awful due to the relatively "fast" reflection-decay profile of the room.

Ideally, past the second curve (decay should have slowed down) -- in my "over-absorbed" room, it doesn't. Nevertheless, the overall effect of room treatment, optimal positioning, and EQ in the end sounds quite good, IMO.

 

[tuga]

No. But classic absorbers don't have a brick wall response that stops at 100 Hz. They go on absorbing to quite high frequencies. Indeed, they are better at absorbing high frequencies than low due to wavelengths being shorter there. Here is a quick simulation of a 4 inch absorber:

View attachment 242163
This is how your room becomes dead. And look at how ineffective the response is below 100 Hz. This forces you to use a lot more of them resulting in that dead room.

You can use specialized, tuned absorbers that don't have this attribute but they are tricky to design/deploy and require skills well above a hobbyist.

But does the amount of mid and upper frequency not depend on the surface area?

Besides, mids and highs radiate forwards, if the bass absorbers are located behind the speakers then they should make little difference.
On side-wall early reflection zone absorption may have a impact but behind the speakers?

 

[amirm]

But does the amount of mid and upper frequency not depend on the surface area?

Besides, mids and highs radiate forwards, if the bass absorbers are located behind the speakers then they should make little difference.
On side-wall early reflection zone absorption may have a impact but behind the speakers?

A few absorbers on front wall is not going to put a dent in bass response. Secondary reflections also go to front wall. As does midrange. You can stand behind me as I talk and still hear me.

 

[Inner Space]

You eq down the peaks. Then listen and if bass is not to your liking, you raise the entire region. This is why proper target curves have a bass boost. The sound with it is rich in bass but not boomy. This is clearly shown in research.

Sorry ... not following. My fault, I'm sure, but let me check - first, you knock down a peak, to where it no longer excites a room mode. Then if necessary you raise that frequency back up, as part of an "entire region", perhaps to where it once again excites the mode. How does that help?

Further, does the research that shows smoother amplitude discriminate between musical notes and quietened room modes? Or does only raw amplitude show up? My preference is to hear correct timbre as well as correct amplitude, and in my experience full-fat physical room treatment is required for that.