Paul Hales on room equalization on Soundstage! Network

[singular]

Presumably so, but I'm basing this solely on the thread title and have not watched or intend to watch the video. As has been said and not only here - I do not use equalization as room correction.

View attachment 554975
and so on..

This curve looks so good.
But by my experience this says nothing.
Measured FR is not enough proof.

And do not ask me what is, because I cannot read IR by eye in subbas range.

I have always more flat curve the more is measured multiple source phase mismatched.

Also if you can measure subbas exactly this way, you probably have very well treated listening room, dedicated to this function, with very low enviroment noise.
Or you perhaps measure 100dB at 20cm distance.

 

[Flaesh]

Also if you can measure subbas exactly this way

Link below. In signature.

 

[Titurel]

”Massive”, meaning very large: ”that spans a significant physical space”.

If you meant large, as in "having large dimensions," say that or say that something is big. Massive is not a synonym for big, particularly not in a technical forum where mass has a very specific meaning. If you want to talk tech, (any expert field) learn the lingo.

 

[Geert]

If you meant large, as in "having large dimensions," say that or say that something is big. Massive is not a synonym for big, particularly not in a technical forum where mass has a very specific meaning. If you want to talk tech, (any expert field) learn the lingo.

I’m an electronic electronic engineer and worked 30 years as a pro sound engineer, but I’ll cancel my membership as you your demand since I’m not worthy. Sorry if my response contains errors, English is not my native language.

 

[Titurel]

I’m an electronic electronic engineer and worked 30 years as a pro sound engineer, but I’ll cancel my membership as you your demand since I’m not worthy. Sorry if my response contains errors, English is not my native language.

mass = massa = messa = Masse = m. M is kind of a biggy, as in e=mc2...

That's a pretty tough cognate to miss or confuse, regardless of one's mother tongue (Moedertaal?).

I'm not an engineer (by training), but I did work for a time in an engineering firm with native speakers of no less than seven languages. None of them would have made that gaffe.

 

[Cbdb2]

Frontiers | Playing to the room: how solo pianists perceive, understand, and adapt to acoustic conditions

Musicians perform in vastly different acoustic environments. Therefore, the abilities to perceive and adapt to acoustic conditions are crucial to expert perf...

www.frontiersin.org

In most halls the sound varies greatly depending on the seat your in, how many people in the hall, and whos mixing the reinforcement. So which seats are they adjusting for? I also think the piano player hears the stage more than the hall.
And its a percussion instrument so the only thing the player can change is how loud hes playing different notes.

 

[Cbdb2]

If you meant large, as in "having large dimensions," say that or say that something is big. Massive is not a synonym for big, particularly not in a technical forum where mass has a very specific meaning. If you want to talk tech, (any expert field) learn the lingo.

Actually it is. https://www.wordhippo.com/what-is/another-word-for/massive.html "large and heavy or solid"

Massive is an adjective, not a tech term, mass is the tech term.

And this is an international site for audio science not an English literature website.

 

[ace_xp2]

mass = massa = messa = Masse = m. M is kind of a biggy, as in e=mc2...

That's a pretty tough cognate to miss or confuse, regardless of one's mother tongue (Moedertaal?).

I'm not an engineer (by training), but I did work for a time in an engineering firm with native speakers of no less than seven languages. None of them would have made that gaffe.

I doubt you've ever done any work with anyone, because if you'd worked with any people at all you'd realize the informality of conversational language. If you have, know this, you are universally considered "that guy" by everyone. And it is never uttered as a compliment.

And it's silly that you lack the comprehension to recognize the nature of the verbiage being used here.

 

[Titurel]

And it's silly that you lack the comprehension to recognize the nature of the verbiage being used here.

No. I merely recognize the need for minimal rigor in conversation, even in casual ones.

Applying a little of that rigor, you might note that I did not object to Geert's opening statement -

"No, because an acoustic piano does not suffer from room mode issues to the exact same extent as a loudspeaker."

I'm probably wouldn't have written "extent" but I clearly got what he was saying and it's not wrong. My objection was to the bullet points offered to support the statement, which I thought were pretty thin gruel.

Yeh, I'm that guy. I'm one of the guys who gets called on the phone at 1 a.m. to settle bar bets.

 

[Suono]

If the listener places a notch in the response at 85 Hz because the reflection contributes a peak to the response, the first cycle will be notched out even though it could not have been influenced by the room resonance. Hales argues that the attempt to address the room resonance distorts that first cycle and audibly detracts from the speakers ability to deliver natural impact at low frequencies.

And what about all the spatial information that is captured but not reproduced?
Only a capable loudspeaker manufacturer could raise such discerning points.

 

[Sokel]

Yeh, I'm that guy. I'm one of the guys who gets called on the phone at 1 a.m. to settle bar bets.

...and I'm the (intoxicated) guy calling you at the other end of the phone

I get your point, really, specially as a principle.

And it would have some use at description of a maybe obscure item or something rare.
But in this case it's a piano, we can all imagine (or know) the dimensions and put it in perspective.

To me, the whole text is just shouting "I'm addressing people that have some knowledge, not some random audience" .
And it's nice to feel part of an informed audience.

 

[singular]

And what about all the spatial information that is captured but not reproduced?
Only a capable loudspeaker manufacturer could raise such discerning points.

The notch also make a dip in Group Delay needed for right perception time, thus spatial feeling of bass scene could be disturbed.

Thatswhy better result could be made throughout system capable of higher F6 - better match to room fundament - without notch.
The disadvantage could be absence of lowest octave.
It all depence on many variables.

Up to my experience, there is no such right natural behavior of any passive loudspeaker system for that case.
And DSP can help here up to some point.

I overcame it by active system with some extreme solutions not applicable for real product.

Anyway, I think today with FIR it is far more easier and effective to get there.
I do not have experience with FIR EQ of passive speakers, and if I go active now I may prefer FIR for each channel ( bass unit / midrange unit / tweeter ).

 

[Titurel]

And it would have some use at description of a maybe obscure item or something rare.
But in this case it's a piano, we can all imagine (or know) the dimensions and put it in perspective.

The problem, from my point of view, is that most people (including many on this forum) don't really understand pianos or most non-electronic instruments nearly as well as they think they do. Just sayin'

Addendum: In retrospect, what probably set off my terminology buzzer was the fact mass does play an important role in piano construction AND that the masses do have acoustical implications, but not in ways germane to this discussion. Thus, the red alert. (Soundboards are expansive, not massive. Cases and plates are massive and expansive.) It reminded me of the difficulty this forum experienced some months ago discussing timbre.

 

[Lars Risbo]

piano: it goes deep in the bass where the wavelength spans much longer than the piano. hence, there is little spatial averaging of the mode shapes of the room for the bass modes.

Another point raised here is that a piano belongs to the percussion group in the orchestra. It has fast transient attack and variable sustain. this makes it very susceptible to distortion of the signal envelope (aka instantaneous amplitude) caused by the long ring out time of room bass modes. A steady state sine wave can only change amplitude and phase due to the room. it’s really the time domain blurr (aka boom) of more dynamic signals that is annoying.

 

[Lars Risbo]

generally on EQ: a system is modelled by its poles (aka room modes) and its zeros (dips). you EW it flat by applying the inverse filter, ie flip the poles and zeros. the room modes (poles) get exactly cancelled by the zeros of the EQ. The problem is that the system may have deep nulls (zeros) and if you invert this we get a huge gain at the dip frequency, ie we introduce a new ringing pole. Therefore, these deep nulls from boundary interference or mode to mode interference should not be inverted, ie we must accept these dips. But this still can remove all blurring from modes ringing because we place the zeros exactly to cancel the room modes ringing poles. This fixes much of the time domain blurr which may have an RT60 of seconds whilst the boundary reflection delay is a few msec. The limitation is of course that if the source or listener move a little then our EQ needs to change accordingly. the listener may move away from a node of a mode and now hear it.

 

[kemmler3D]

As an impact lover, I read the thread twice and still don't get it:

Is it solely about DSP room correction or about room correction (proper, passive) in general?

Cause apart from the obvious (high SPL capable speakers, sane, decent conditions) best and proper, lifelike impact I have felt (yes, that's not BS audible differences, it's physical and far and away from subs rumble) was at a control room, treated, flush-mounted mains, etc, the works.

It gets theoretical here, with only maybe a handful of chest punch-hitting speakers tested (so limited data) .
The rest of them, DSP corrected (specially filling dips) will be even less capable of that.

So what's the rant? Is that already limited SPL speakers get more limited with (certain) corrections and they don't kick?
That's a rant for the shake of the rant, alone.

Edit:I would maybe try to catch a real hit with REW's SPL chart just to see the time and SPL (usually an near instantaneous spike north of 110dB (Z) ) but it would be extremely case-specific, so...

The main objection is that EQ to fix room modes is really treating direct and indirect sound at once. (Or rather, treating the total room response by treating the direct sound.) But if you EQ out the dips, this robs energy from the direct sound and will make transients less impactful, even if the overall frequency response and steady tones look perfect.

The idea is that the sound will be incorrect until it has a chance to interact with the room.

I think there is probably something to this objection, but I think in practice I prefer to deal with peaks and not worry that there is a notch in the direct sound which lasts for all of 6ms before the reflections get there. Also not expert enough in acoustics and psychoacoustics to really get into the validity of the objection and how close the intuitive interpretation is to reality.

 

[Flaesh]

EQ to fix room modes is really treating direct and indirect sound at once

First, an equalized signal\direct sound is emitted, then RTF makes sound power similar to the original signal. We can discuss and study what we hear\not.

 

[NTK]

Much of the "transients" of the sound are in the higher frequency components. Those aren't affected by low frequency room EQ.
Spectrograms from: https://www.researchgate.net/public...rum_Transcription_and_Drum_Pattern_Generation

 

[Flaesh]

Much of the "transients" of the sound are in the higher frequency components

https://www.audiocheck.net/audiotests_subwooferkicktest.php - I would remove the unnecessary word "subwoofer", but otherwise it's a useful test . One weirdo claimed that his laptop speakers made his cabinets and windows shake equally with any of this test signals .

 

[SigbergAudio]

The main objection is that EQ to fix room modes is really treating direct and indirect sound at once. But if you EQ out the dips, this robs energy from the direct sound and will make transients less impactful, even if the overall frequency response and steady tones look perfect.

The idea is that the sound will be incorrect until it has a chance to interact with the room.

I think there is probably something to this objection, but I think in practice I prefer to deal with peaks and not worry that there is a notch in the direct sound which lasts for all of 6ms before the reflections get there. Also not expert enough in acoustics and psychoacoustics to really get into the validity of the objection and how close the intuitive interpretation is to reality.

You should only use EQ to fix room modes below 100hz or so, where A) there's not really direct sound in the same sense, and B) there's not really any transients (as others have mentioned they are made up of higher frequency components).

So while this objection sounds like it makes sense, my experience is that it has little merit if you apply EQ in the right way. That means to not necessarily dampen peaks 100%, and restore overall level in the bass if significant peaks existed before EQ, so that post-EQ bass level isn't too soft.