Sound Quality Vs. Measurements

Status
Not open for further replies.
I've had the chance to work a bit with analysers from R&S(UPV), AP (2722) and PrismSound (DScopeIII). Each one has it's pros and cons but the DScope became my favourite for everday work, price/performance/size ratio is hard to beat and the user interface is excellent and way ahead of them all (IMHO). As for the 2722, to reach its full potential we had to outboard the toroid... The DScope went away recently with the engineer who's private property it was, so now I have to live with the AP -- well, things could be worse :)

If you look at standalone analysers like the UPV, Stanford's SR1 could be underdog competitor whose specs look good on paper (never seen one in RL).
 
Last edited:
I think the tone of Thorsten's remarks ought to warn you that there is a tiny chance that he might be be somewhat exaggerating and misrepresenting the views and attitudes of those he disagrees with. For example, I don't know why Thorsten keeps banging on about THD - he mentions it far more often than the people he alleges are bound by it.

There might be some coincidence with the recent review of an AMR DAC in Stereophile, which was excellent, but showed THD not to be state of the art.
 
Hi,



Yes, de gustibus et coloribus non est disputandum.

Where I have problems is when people do not express personal preference (I like Apples better than Oranges) but conclude that the only fruit that can be eaten is either apple or orange.

I like Pears incidentally and a nice Perry, but also am partial to "Apfelkorn" [apple vodka] or a littel Cointreu [orange brandy]... So in fact I like all fruits once they have been made into beverages... ;-)

Ciao T

I don't drink at all, not for moral reasons, but because it does nothing for me, though I do like beer and an occasional glass of wine.

As ever, there are two exceptions to this: Martini and Cointreau. Cointreau is the only drink my wife will take a sip of with pleasure, just to unwind.

I've had the opportunity to try many whiskeies, and sadly they all taste the same to me, agin, except two: Jack Daniels and Grant's. These two I actually differentiate by taste, on the one or two occoasions every year I take them.

I am a junkie for fruit juices, especially orange and apple. Not to even mention milk.
 
Fully agree. However, if person W claims that X sounds better than Y, person Z may justifiably ask that this assertion is tested by scientific means, i.e. controlled ABX testing. Person W may refuse this, at which point person Z may conclude that it is all between the ears.

vac

I didn't mean that in absolute terms, but quite the opposite, in very personal terms.

Like when somebody tells you that your favorite cartridge, the one you have chosen after trying out many more, is in fact all wrong, and that you should get model X by Y, which will sound better. And you tried that one and dumped it. So that person knows better what you should have than you yourself.

Down to the utmost personal level.
 
This is an example of artificial sound enhancement that fooled me at the time.
Back in 1973, I was in NYC at the behest of the GD, and my present girlfriend then was with me by chance. She worked as the west coast representative for APHEX, and that is how I had met her. Well, we went to the APHEX office in NYC, since I was there in Manhattan.
I was given a demonstration of this device, and even allowed to use it with a Grateful Dead recording that they had. I 'dialed' it in so perfectly, that the owners of APHEX wanted to hire me on the spot as a recording engineer. I told them that I knew the band's sound because I had worked with them for years, and was on tour with them at the moment, not because I was a recording engineer, but this gave me an idea. I would get the Grateful Dead themselves to come over and listen to this 'amazing device', and they came.
They listened to the demonstration, and Jerry Garcia told me straight out! "It's an effects generator, man. We have stuff like that in the recording studio." I had thought it to be a 'miracle box' that returned 'life' to otherwise over-processed music, which it did at times. Oh well, at least they set me straight.
I still remained friends Kurt Kenopel (sp) the inventor of the APHEX, because I found him an interesting guy. And so it goes.
 
Hi,

Thorsten likes to paint things in black and white terms because it saves him time.

Not so much because it saves time, but because it creates clarity.

In reality I'm much more a shades of grey man, but in a place where one side persistently tries to paint the other black and itself white, I guess I have to add a paintjob of my own.

As for correlation of the measurements currently espoused as measure of quality, once someone demonstrates positive correlation that if I "improve" the Quantity Y of Quality A in these measurements I get better performance, I will accept that there is a correlation.

What you have stated (literally all of it) has been shown to be inadequate or irrelevant in one way or other. So sorry, correlation is ZERO from a scientific and statistical viewpoint.

Note, this does not mean that all measurements are useless, just that those commonly used as measure of quality are and should have been replaced decades ago (and would have been, if new data would cause correction).

So the "Black/White" is needed, if actually not that agreeable to me.

Ciao T
 
Sy,

Not so much from the look of the measurements. I would like to have seen a 10k load, since that seems to be a common value in real equipment these days, but I'd be surprised if the HD rose to the level of audible. More IM data would be nice as well, but that's me being greedy, I suppose.

Personally I only test with 10K (100K loading makes things too easy), it does not look a whole different, maybe a little more distortion.

As for anyone shouting "Effect Machine", may I suggest:

Goest thou an measure thy own loudspeakers at 0dBfs* and shew me just ONE Speaker that has less HD 20Hz to 20KHz and I shall relent...

Ciao T

*that is 108dB peak at the listening position based on THX, 111dB peak based on EBU
 
Hi,

There might be some coincidence with the recent review of an AMR DAC in Stereophile, which was excellent, but showed THD not to be state of the art.

It is coincidence. I have been banging on about how meaningless THD is since the days before DIYA, AMR and the offending DAC even existed.

Again, if you want "state of the art THD" to be taken serious as an indication of actual quality, please demonstrate that lower THD reliably produces lower audible distortion, based on the best knowledge we have at this point regarding distortion audibility.

Secondly, please illustrate that speakers exist in sufficient wide availability that allow this "state of the art THD" to be delivered to a listener at realistic levels (see THX or EBU recommendations).

Otherwise all this "state of the art THD" is good for is nice advertising copy and deceptively marketing crud to the unsuspecting public.

Evidence please, not polemic.

Ciao T
 
Hi,

There s indeed total correlation between measurements and sound quality.

There is?

So if I measure lower THD I will reliably have better sound quality?

Do you have any evidence to support this?

If ever listening tests do not agree with measured datas , then the logical conclusions is that measurements were not done adequatly in respect of the listening conditions.

Well, the listening tests are on record for many decades and they have been repeatedly confirmed.

So perhaps the measurements instead are simply meaningless?

Ciao T
 
A low THD is not the be all and end all in regards to what sounds "good".

You don't have to look far - Guitar amps. An overdriven tubed based guitar amp sounds much better then the comparable Solid State version with lower THD.

It not only sounds better live, it makes better sounding recordings too.

You would be hard pressed to find a person who prefers SS over tubes in that regard. It's almost universally accepted.

Just something to consider.

I understand the origin of the figure to imply that it does not impose much of it's own distortions.
 
Last edited:
dvv said:
I don't drink at all . . I do like beer and an occasional glass of wine. . . . two exceptions to this: Martini and Cointreau. . . . try many whiskeies . . Jack Daniels and Grant's.
Sorry, I just found the juxtaposition of the initial claim with all the minor exceptions after it rather amusing! I guess "not at all" doesn't always mean the same thing to all people.
 
Hi,

This is an example of artificial sound enhancement that fooled me at the time.

....

APHEX

I happen to know Aphex (to be prounced 'efex') intimately. The big Band sound system had one and I cloned it (incidently, Uli Behringer also started with Cloning APHEX).

What it does plain and simple is to add a lot of distortion in a specific band (where the human hearing is most sensitive). As a result everything is made to sound "louder".

Correctly set it could make a PA system sound subjectively as much as 6dB louder. Now 6dB louder is a LOT in terms of additional Speakers and Amp's.

So I absolutely loved the APHEX, even it made the quality of the sound rather poor (actually, it sounded like hammered proverbial, aggressive at the edge of unpleasant).

Later when I got to work with better (read louder and cleaner) PA Gear and no longer needed to lend the roadies a hand during set-up the APHEX quickly fell out of my favour, replaced by an almost fanatical devotion to the holy trinity named Community (Full Range & Mid Bass), Servodrive (Basstech 7 Subs) and Yamaha (Amplifiers, digital EQ, digital XO, mixing desk). Such a system could sound almost like a very good large format studio monitor, but at 135dB/1m...

Ciao T
 
Hi,

In any case, I no longer lust after an AP2.

I only need an AP2 to keep a check on what reviewers/mags that use it measure and if they may have made a mistake somewhere (as happens to the best of us).

In day to day use I find it often too cumbersome.

My day to day audio analyse is an EMU 1616M with a custom breakout box. It does > 90% of the routine stuff the AP does and often easier/more intuitive.

There are things I can only measure using an AP2 or something even more esotheric.

I would say for DIY Audio a EMU 1616M, various freeware software and breakout box is incredible value for money. You can measure literally any item in the chain, with the addition of an accurate iRIAA Network a Measuremenmt Mike and a cheap Chip Amp...

Add a pair of small speakers and the whole set-up can play music in your Lab while soldering...

Ciao T
 
Sorry, I just found the juxtaposition of the initial claim with all the minor exceptions after it rather amusing! I guess "not at all" doesn't always mean the same thing to all people.

Point taken - what I was trying to say so clumsily was that I drink only Martini and Cointreau for the pleasure of it, and that very rarely, say at most once a month, ona small glass, lots of ice. While I can make out on taste alone Jack Daniels and Grant's whiskey (i.e. be handed a glass and know what's in it without being told), I do not actually use them except on rare and special occasions, such as weddings, family day (local tradition), childbirths, that sort of thing. That would be like twice a year, if that.

On the other hand, I do enjoy my pint of beer in good company and/or with some foods, or a glass of wine, or a pint of strictly British cider. I don't think I average more than a pint of beer or cider a week, or thereabouts.

So, I think it's fair to say I am not a drinking man, even if I do have a pint here and there. "Not at all" was an ill chosen term, partly because beer and wine are not considered to be drinks locally, and are not taxed as drinks either, more like something in between hard stuff and fruit juices.

You're not going to make trouble for me about the milk now, are you? :D
 
dvv:
I guess there are sections of British society where you would be regarded as a non-drinker. Having grown up with totally 'dry' parents I look at things from the opposite end. By their standards I am a drinker because I might occasionally have a glass of wine with a meal. Don't like beer, never tried spirits. Can't quite decide whether I like cider, as most are too fizzy; maybe I need to try a few more.
 
Status
Not open for further replies.