Perceptual Effects of Room Reflections

[Curvature]

Nothing wrong with something from the last century when it's correct. We don't have newer studies that disprove this.
Fazenda is completely off the track when he is considering the reverberation time of small spaces. That leads to a lot of wrong conclusions.

Here's more recent research by Fazenda here, which should be read in light of a few of his other papers IMO: https://www.audiosciencereview.com/...atment-goldensound.45104/page-74#post-1629663

What's he off track about?

 

[Bjorn]

Reverberation time doesn't exist in small spaces.

Lyd & Akustikk | Small rooms and reverberation time

www.lydogakustikk.no

 

[Curvature]

Reverberation time doesn't exist in small spaces.

Lyd & Akustikk | Small rooms and reverberation time

www.lydogakustikk.no

Did you even read Fazenda's work? He's not a simpleton.

 

[dasdoing]

It's a typical response from the last century, since the beginning of our century there has been a lot of research on the subject and although the advances are obvious nothing has yet been written in stone...

the paper seams to suport the last century response though lol

 

[kemmler3D]

I just don't like listening to music in rooms with substantial reflections. Maybe I hear the comb filtering as suggested in Amir's OP. In any event the deader I make my room, the happier I am.

Same. People say "comb filtering isn't audible" sometimes and I can't get my head around it. If you have standing wave reflections at high frequencies, you can easily hear comb filtering and also pick out the nulls by moving your head a bit. Trivial to hear if you play tones in a s***ty reflective small room.

I think when people say "comb filtering isn't audible" they might mean you can't pick out individual nulls while listening to music. I guess that's fair. But if you apply a comb filter based on a room response via DSP, the A/B is extremely obvious. So what does it mean to say comb filtering isn't audible? Maybe I just haven't read enough.

 

[Bjorn]

Comb filtering isn't audible if it delayed with 30 ms, because so late in time it will be perceived as a separate sound. But anything within that time span is audible. How much depends on angle, arrival of time and instrument type/frequency. And we are generally more sensitive to peaks.

I think you are correct about that we don't hear discrete reflections individually well when there are multiples of them.

 

[josh358]

It's a typical response from the last century, since the beginning of our century there has been a lot of research on the subject and although the advances are obvious nothing has yet been written in stone...

the paper seams to suport the last century response though lol

Yeah, guess I'm missing something, because excepting his discussion of distributed mode loudspeakers it seems to be very conventional stuff.

 

[fredoamigo]

the paper seams to suport the last century response though lo

Yeah, guess I'm missing something, because excepting his discussion of distributed mode loudspeakers it seems to be very conventional stuff.

the paper doesn't assert anything ... it simply asks the right questions about trends, evolutions, acoustic problems in control rooms more than 20 years ago ...

 

[amirm]

An experienced mixer will also know how the sound in his room translates into the sound in a home listening environment.

Since the percentage of people who a) listen to music and b) have acoustic products in their room is negligible, I trust they mix for a normal reflective room. Yes?

 

[kemmler3D]

Since the percentage of people who a) listen to music and b) have acoustic products in their room is negligible, I trust they mix for a normal reflective room. Yes?

I'm not a pro but I used to read a lot about mixing, and have at least attempted properly mixing two albums. I don't recall seeing guidance about mixing for listener acoustics either way. A variety of speakers or headphones, yes - acoustics no. But I was working on a more or less amateur level 10 years ago, so don't take this comment too seriously.

 

[Eckerslad]

I hear about mixers checking their final mixes on known poor speakers like Auratones or NS10s, or even in their cars to check their product sounds good on the average Joe’s boombox or whatever. I don’t know how true this is, however

 

[dasdoing]

Translation is impossible if you mix for the sound of a specific room. It will only work in that exact room.

Real engineers built those million-dollar studios for the producers/mixers, and yet started to bring small speakers and place them nearby because it still wasn't truly working in the far field.

Translation isn't guaranteed when mixed "without the room" because, of course, there will be the influence of that room in the real world. However, mixing for one room and expecting it to translate somehow to another room is impossible.

 

[josh358]

Since the percentage of people who a) listen to music and b) have acoustic products in their room is negligible, I trust they mix for a normal reflective room. Yes?

Sure, and normal crappy equipment. I think it still sounds better on a good home system and decent acoustics, or at least did until Auto-Tune and the loudness wars, anyway. But I also think the type of good system may be different, perhaps sacrificing some accuracy for midbass slam, etc. And acoustical music is usually mastered on good monitors, or at least was back in the day (I've been retired for some years now).

 

[josh358]

the paper doesn't assert anything ... it simply asks the right questions about trends, evolutions, acoustic problems in control rooms more than 20 years ago ...

I'm not sure I follow. The control rooms of the day were designed with all of these issues in mind, using the principles in the paper. What has happened since, other than more small rooms and abominations like the loudness wars and Auto Tune, and the increased importance of earbuds? Modern pop doesn't usually sound good on a good system.

 

[josh358]

I hear about mixers checking their final mixes on known poor speakers like Auratones or NS10s, or even in their cars to check their product sounds good on the average Joe’s boombox or whatever. I don’t know how true this is, however

Yes, that's true.

 

[josh358]

Translation is impossible if you mix for the sound of a specific room. It will only work in that exact room.

Real engineers built those million-dollar studios for the producers/mixers, and yet started to bring small speakers and place them nearby because it still wasn't truly working in the far field.

Translation isn't guaranteed when mixed "without the room" because, of course, there will be the influence of that room in the real world. However, mixing for one room and expecting it to translate somehow to another room is impossible.

True, but what they do is listen to their mixes on a variety of home systems, which do tend to have certain characteristics, e.g., limited bandwidth. These days, I gather that they focus more on earbuds.

 

[dasdoing]

True, but what they do is listen to their mixes on a variety of home systems, which do tend to have certain characteristics, e.g., limited bandwidth. These days, I gather that they focus more on earbuds.

The final balance is actually more or less determined by existing references. They are not (re)creating real recording spaces anyway... unless we are talking about classical stereo recordings. But even those often have their real space/ambiance not fully preserved.

The check on other systems is more like a sanity check. "It should work here." It won't guarantee universal translation.

 

[tomeh]

What is largest wildcard from location to location and playback system to playback system?

What is hardest thing to tame in a room?

If you haven't designed, measured, refined and verified that you have the <300hz region in control, you'll never know if you need correction, have made proper corrections and can hear the results.

Standards have existed for a long time that ensure the frequencies that are directional as a result of the sources physical size and wavelengths can make those above 300hz easier to duplicate from location to location. https://www.itu.int/dms_pubrec/itu-r/rec/bs/R-REC-BS.775-4-202212-I!!PDF-E.pdf

If the playback system measures reasonably flat >300hz, the off axis response is fairly flat and symmetrical then applying the standard allows us fair control of this range. (assuming low distortion levels from system to system)

Headphones and earbuds? No room in the equation so it's strictly the response of the device while near or in the ear.

So the highest nail to hit is; <300 hz, hit the bullseye of design, measurement, refine and verify.

This is a mixing and master bullseye.

Now there is an infinite number of consumer's playback systems, rooms, cars ...... where the engagement of the environment will add reverberation (on top of any from the environment where the recording microphones where plus added by the mixing engineer), mixing of direct and reflected sound and create another unique layer.

You can either choose to endlessly create these unique layers or you can choose to hit the high nails and aim to hear it as close as possible to what the creators did?

Nothing wrong with either that you enjoy.

 

[dasdoing]

If the playback system measures reasonably flat >300hz, the off axis response is fairly flat and symmetrical then applying the standard allows us fair control of this range.

that might be true for an empty room, but as soon as you put objects in the path LP response becomes caotic

 

[tomeh]

that might be true for an empty room, but as soon as you put objects in the path LP response becomes caotic

I'm talking about hitting the highest nail here and having a target. "Now there is an infinite number of consumer's playback systems, rooms, cars ......"

Yes, an endless amount of large objects in the room has a negative effect. Large cavities and recesses have a negative effect.
Four mid-wall subs have a positive effect.

"You can either choose to endlessly create these unique layers or you can choose to hit the high nails and aim to hear it as close as possible to what the creators did?"


Right?


Cheers