What is Alan Shaw on about? (is "coloration" unmeasurable?)

[Waxx]

He says he has made dsp systems that are flat and neutral and fit his expectations, but is not ready to release them yet as he is doubting that access to DSP will be beneficial to his clients, and is trying to find out what he should give access to and what not.

If you want to follow the discussion, make a free Harbeth account and follow here: https://harbeth.co.uk/usergroup/threads/digital-signal-processing-and-loudspeakers.80536/page-5

 

[ahofer]

The OP is not correct in terms of Toole that Steady State in room responses can be enough information to understand what is being heard. It can't be more clear from his book that you first need a complete spin. That is the whole premise of the book.

If you look carefully, you'll see that the OP suggested a CEA-2034

 

[krabapple]

Toole touches on this well enough in his book and it is clear that at the time it was published he is not a big fan of micromanaging in room response nor has doing so been shown to correlate with improved subjective sound in a blind test.
Sure using it as a broad band tone control should be fine above the transition zone - expecting to fix design and tonal flaws with something that can only see the in room steady state is rather polar to what Floyd Toole's whole premise is in his book - that speakers that measure well anechoically sound the best in a real room and the mind of the listener adapts to the room.

Yes, Dr. Toole is opposed to DSP 'room correction' above the transition band if the loudspeaker is well designed i.e., 'measures well anaechoically' in parameters that correlate with listener preference.

If it doesn't...

 

[ROOSKIE]

Yes, Dr. Toole is opposed to DSP 'room correction' above the transition band if the loudspeaker is well designed i.e., 'measures well anaechoically' in parameters that correlate with listener preference.

If it doesn't...

Making changes via DSP to the anechoic response is fine and of course one must either verify the changes anechoically or deeply understand what the limitations are so as to avoid making adjustments that create a different problem.
Making changes, beyond broadband tonal adjustments, based on the in room response of any loudspeaker, regardless of design performance, above the transition zone is a crap shoot.
In terms of benefit, the odds are equally as great that whatever was done to make the design fit an in room idealized curve will actually degrade the sound even if it looks good on paper, again this is regardless of speaker quality. That room curve has almost no granularity all.

I do sometimes myself make such adjustments based off of in room data, but they are broad (basically tone controls) and typically minimal unless they are DSP adjustments based on the anechoic data from the Klipple here or if there is a chamber they were measured in like at Soundstage- then I do what is possible.

In any case this all needs to be blind tested someday and made public. I know Harmn did some testing that was made in part public and the results for 'room correction/adjustment' software at that time were very mixed despite the curves looking clean on paper.

Who knows, I just follow what Toole has prescribed for now until someone finally does a lot more testing in the next round of acoustic study. (which may not happen soon as I think many companies are fixed on the way things are $ or selling headphones and Bluetooth mini speakers)

 

[krabapple]

I don't disagree. I didn't say it would be easy or never-fail. In the end there's nothing wrong with trying. The hard part, for e.g. Audyssey users, is quick switching between say, full bandwidth correction vs below-Schroeder correction, to compare. I'd love to do it. I'm not aware of any easy way to do it from an AVR; even switching between Audyssey 'Flat' and "Reference' and 'off' in mine takes at least several seconds, and it doesn't let me load more than one calibration profile at a time. Then there's the issue of sighted listening...

And on top of that, at least in my Denons, you can't use broadband tone controls -- good old bass/treble controls -- when Audyssey is on. It's one or the other, even if Audyssey processing is limited to bass using the app version.

 

[ahofer]

I just follow what Toole has prescribed for now until someone finally does a lot more testing in the next round of acoustic study. (which may not happen soon as I think many companies are fixed on the way things are $ or selling headphones and Bluetooth mini speakers

It's a shame there hasn't been follow on. I wonder if it could be crowd-sourced/crowd-funded? The hard part is the speaker positioning.

 

[ahofer]

And to return to the original subject-

I still don't have a clear picture of what might be "coloration" that would not be revealed in spinorama FR and distortion measurements (inclusive of dynamic distortion testing like Erin does). Whether it exists or not, do we even have a theory?

 

[ROOSKIE]

And to return to the original subject-

I still don't have a clear picture of what might be "coloration" that would not be revealed in spinorama FR and distortion measurements (inclusive of dynamic distortion testing like Erin does). Whether it exists or not, do we even have a theory?

I think the Spin does get one close, 80ish % and then the rest is very complex and objectively subjective if you like.
Some distinct issues have already been mentioned-
IMD distortion which can vary wildly from driver to driver and is very difficult measure and subjectively for one person may not be an issue while the same system might sound messy to another, Doppler distortion (from high excursions creating turbulence, this is different from IMD and a reason for large woofers with lower excursion-SPL requirements) Complex Harmonic Distortions that are not measured during a sweep, imagine a woofer with 1% HD at 800hrz on a sweep now at a high excursion due to big cone movements at 40hrz - HD at 800hrz may actually be different and this is not IMD.
Higher percentages of cabinet colorations (or lack of colorations if you like a lively cabinet). Drivers ringing when under stress. Some folks are sensitive to sounds inside the cabinet being re-transmitted through the cone (and others may find the sound lifeless without some of this). Bass tuning changing when temperatures increase in the V.C. and the TS parameters change - especially at higher-high SPL. People who are sensitive to extraneous port emissions. Geddes found cabinet round overs help a lot - in blind testing folks preferred a big rounded edge to sharp even if the rounded edge had higher HD from the system.
The list is large, I am certain more items can be added, some of which may be trivial unless combined with other issues and then perhaps very audible - especially for listeners really analyzing the sound vs just enjoying tunes.

 

[Waxx]

I think the Spin does get one close, 80ish % and then the rest is very complex and objectively subjective if you like.
Some distinct issues have already been mentioned-
IMD distortion which can vary wildly from driver to driver and is very difficult measure and subjectively for one person may not be an issue while the same system might sound messy to another, Doppler distortion (from high excursions creating turbulence, this is different from IMD and a reason for large woofers with lower excursion-SPL requirements) Complex Harmonic Distortions that are not measured during a sweep, imagine a woofer with 1% HD at 800hrz on a sweep now at a high excursion due to big cone movements at 40hrz - HD at 800hrz may actually be different and this is not IMD.
Higher percentages of cabinet colorations (or lack of colorations if you like a lively cabinet). Drivers ringing when under stress. Some folks are sensitive to sounds inside the cabinet being re-transmitted through the cone (and others may find the sound lifeless without some of this). Bass tuning changing when temperatures increase in the V.C. and the TS parameters change - especially at higher-high SPL. People who are sensitive to extraneous port emissions. Geddes found cabinet round overs help a lot - in blind testing folks preferred a big rounded edge to sharp even if the rounded edge had higher HD from the system.
The list is large, I am certain more items can be added, some of which may be trivial unless combined with other issues and then perhaps very audible - especially for listeners really analyzing the sound vs just enjoying tunes.

That is what i also thought. There is a lot more to measure than is done today. Not all is relevant, but it's still the case that certain speakers that measure very good (cfr older Genelec monitors) still are hard to listen to for me, after a few hours my ears start to ring, even when listening on low volume. And i'm not the only one who has that issue with those I don't know why, and i never saw measurements that can explain it, but i would love to find out on scientific base why that is. I first thought it was because they are too neutral for my ears, but when i eq'd them to my preference curve, it stayed the same. I actually discovered this site in that search (by looking for measurement of Genelec monitors).

And there are many other neutral speakers that don't have that. I can listen for hours to Neuman KH420 or Kii Three BXT speakers (who also measure neutral low distortion and even dispertion like the Genelecs) and get no ear fatigue. And those are also not fit for my "preference curve"...

 

[kongwee]

Well, coloration was used to describe tube amplification. Pretty sure you can measure that against solid state amplifier.

 

[ThoFi]

about the coloration and dsp topic I hoped that there will be some new information on HUG.
but there are none.
also no new information about results of using dsp of competitor speakers.
nothing.
Disappointing because HUG claims to be a forum to discuss technical topics.
instead a thread is published „speeding up the energy transition“ ?????
it looks like there is nothing further to discuss about speaker related topics.

 

[Willem]

By now we know you like to disparage Harbeth and its forum because you were shown to be wrong. In fact, there has been some discussion of the forthcoming line of active Harbeth speakers and what kind of dsp functionality would be desirable. Beyond that, I think we just have to wait, as with most new products before their official introduction.

 

[ThoFi]

Please share the link of the discussions and maybe detailed information here.

 

[Waxx]

Product development takes time, so it's not abonormal you don't see the results yet. It can take years, especially with a picky guy like Allan Shaw. For the discussion, go to that forum, it's out there, but you need to be a member to see it. Links won't work (they block it).

 

[lashto]

I admire Alan, but I still don't understand how "coloration" isn't frequency response or distortion

Should be one of those two, yes.

But it's not necessarily the issue there. The problem might also come from the guy's ears: i.e. nothing changed in the sound of that room/system but he's getting older and has a tinnitus-like problem.

Pretty much impossible to diagnose such things, especially from afar. But AS's responses are indeed too 'marketing-heavy'.

 

[ahofer]

Having now had a pair of Revel F228be for a while, I’m not sure I detect anything ’colored’ vs. my Harbeth SHL5+(AE). They are both easy to listen to for long periods of time.

Harbeth’s naming convention, however, is due for a reset.

 

[ThoFi]

Should be one of those two, yes.

But it's not necessarily the issue there. The problem might also come from the guy's ears: i.e. nothing changed in the sound of that room/system but he's getting older and has a tinnitus-like problem.

Pretty much impossible to diagnose such things, especially from afar. But AS's responses are indeed too 'marketing-heavy'.

AS is marketing-heay and vague.

 

[theREALdotnet]

Harbeth’s naming convention, however, is due for a reset.

Yes, they should follow Sony’s lead and adopt catchier model names

 

[ThoFi]

Product development takes time, so it's not abonormal you don't see the results yet. It can take years, especially with a picky guy like Allan Shaw. For the discussion, go to that forum, it's out there, but you need to be a member to see it. Links won't work (they block it).

Of course, product development takes its time and is secret.
But in this case, its about topics that are discussed vague at HUG.
only creating content…

 

[DSJR]

about the coloration and dsp topic I hoped that there will be some new information on HUG.
but there are none.
also no new information about results of using dsp of competitor speakers.
nothing.
Disappointing because HUG claims to be a forum to discuss technical topics.
instead a thread is published „speeding up the energy transition“ ?????
it looks like there is nothing further to discuss about speaker related topics.

The HUG is *first and foremost* a MANUFACTURERS forum and heck, I've been pulled up by Alan on a good few occasions regarding this fact! It's there to promote and discuss audio matters from a Harbeth perspective and yes, technical aspects are discussed in a more general form so the non-tech users who read there can keep up.

As for measuring colourations, it depends where they are I reckon. Many port-driver resonances in the upper hundred Hertz region can be easy to see on the Klippel plots posted here and elsewhere and it seems most sub £/$1,000 pairs of active wonders are riddled with them! The 'waterfall' plots posted by HFN and now Amir can, if set up carefully, show resonances especially at upper mid and higher frequencies.

Having said that, I remember many years ago, that ALL speakers using the then famous KEF B200 bass driver, had a very slight cupped-hands 'awww' colouration to them, no matter how well the crossover was set up, even the much loved (by me) 104ab speaker. You only noticed it when compared instantly with other speakers, bextrene coned or not, but it was there. How this could be measured I really don't know as the main response of this driver was known and dealt with well by those who knew how to use it properly.