|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
70.168.149.115
In Reply to: RE: surrogates posted by fstein on November 10, 2015 at 20:40:27
..take a flight in a plane that was not well engineered but someone recommended to you because they enjoyed the ride?
I can only assume the criticism of Stereophile measurements is because either people don't understand them and feel inadequate or the conclusion says their beloved kit is 'not well engineered' and are insulted.
There can't be many industries where a product does not have to meet certain performance levels. What those performance levels should be for audio is certainly open to debate given that poorly measuring equipment gets glowing subjective reviews. But where's the harm in running basic tests?
Criticizing measurements is shooting the messenger - hold the designers to account, do they know what they are doing or are they just glorified tweakers or even charlatans?
Regards
13DoW
Follow Ups:
I'm going to use your text here. When you say "not well engineered" when talking about an aircraft, the plane might have handling problems, structural problems and so on. In other words it is not able to meet the requirements of dealing with flight.
The problem you are dealing with in audio has to do with the human ear/brain system, which is in effect the equivalent of the air that a plane has to deal with (music being the other part of that- the analogy isn't perfect but it does work).
The fact of the matter is that amps that 'measure well' don't do the best when dealing with the human ear/brain perceptual system. What they do well is look good on paper. The problem here of course is we don't experience music through paper, we experience it through our ears and brain.
When the audio industry ignores the human hearing perceptual rules, it is ignoring some fundamentals! Its like not getting the airfoil right on the aircraft- pretty basic stuff.
But as an industry we do it all the time and think nothing of it. For example, its been known since the 1930s that the 7th harmonic causes a metallic hardness in audio reproduction and also that it does not have to be at a particularly high level to do that. In fact during the 1960s General Electric showed that the 7th harmonic plays a role in our ability to hear how loud a sound is- IOW the ear/brain system uses the 7th and other higher ordered harmonics to gauge sound pressure. But we routinely ignore that information too.
So when we measure the wrong things, all that happens is that amps continue to look good on paper but don't always sound good in practice. Not only that but amps that don't measure all that well might continue to sound good. We've been having this discussion now for over 35 years- how long do you think it will continue before we come around to the idea that we have to measure what is important to the human ear rather than the things that don't matter?
The bottom line here is when an amplifier is built to look good on paper, its doing much the same thing that the Volkswagon cars were doing- cheating on the exam. What is really important is what it does in real life- how well it makes music to the human ear. If its not designed to do the latter, its 'not well engineered'.
So, back to the original post,
Present measurement techniques are a surrogate endpoint, which may or may not relate to the "clinical endpoint" of good sound.
In other words, we still have to go listen.
nt
try it! you know you want to!
And that's even measurable, although it can be at really low levels such that there are measurement difficulties: it is hard to see in all the noise and yet can still cause a problem.
But if you wanted to point at a problem area, that is a classic that's been around for many decades and not really been solved by solid state designers, although tube amp designers seem to fare better. You might see it as the holy grail; build a solid state amp that lacks odd ordered harmonic generation in general and you are on to something.
Honestly, that would represent transformation and it would likely end the market for tubes.
nt
try it! you know you want to!
-I have often found the use of that term to be misleading, wherein the designer felt it was 'negligible'.
The fact is no solid state amp is lacking that harmonic in such a way that it is inaudible (many tube amps have it too, but apparently in lessor amounts). It is true though that some have less than others- and are often described as having 'tube like' sound. But it is still present and audible as our ears are more sensitive to higher ordered harmonics like this one than the best test equipment.
Therein lies part of the subjectivst/objectivist debate which has been going on for decades and will continue until this fact is realized. I'm not holding my breath.
to recognize sound....Get a hearing test and your doctor can confirm this for you...Then compare that level to that of distortion spectra as measured for any number of ss or class d amps. You will find the amps are not outputing 7th harmonic at levels you can hear.
try it! you know you want to!
And FWIW I was not suggesting 'infinite sensitivity'!!
What I did say is that our ears can detect the 7th harmonic better than test instruments can. This is why two amps can measure the same bandwidth on the bench but one can sound bright while the other does not. Its not due to frequency response variation, its due to the fact that the ear/brain system converts distortion into tonality. This and the ear/brain system uses higher ordered harmonics in order to calculate how loud a sound is. In that regard we seem to have about 140db or so range. That's wider than a lot of test equipment can do...
I certainly have never read anything that gave a figure of human auditory thresholds anywhere near that, especially when the signal contains mixed frequencies...Best to keep in mind as well the limits of the source media as well: vinyl is good for about 50-70 dB; a standard cd/16-bit is good for a 96 dB range, and a 24-bit DVD has a theoretical 144 dB range.
I suppose it comes down to the fact that while our senses are rather limited, our ability to fool ourselves is almost endless...
try it! you know you want to!
Edits: 11/13/15
Our ability to hear a 7th harmonic does not require that the amplifier be playing quietly. It seems that you don't understand how distortion is perceived by the human ear.
We can hear that 7th harmonic quite easily regardless of how loud the amp is playing.
and that some amps don't produce audible amounts of harmonic distortion....
try it! you know you want to!
The ear is so finely tuned that it can pick them out easily. It does not matter that they might be at a low level- the ones of concern occur in the frequency bands to which the ear is the most sensitive, and in addition it uses those harmonics to calculate volume. Its more sensitive to them than it is the fundamental tones that made them!
You don't seem to understand how the ear works in this regard. Its not like an amp is on the bench where the less volume the less distortion.
try this:
https://phineasgage.wordpress.com/2007/10/13/audiophiles-and-the-limitations-of-human-hearing/
try it! you know you want to!
So now do you get that what you are talking about and what I've been talking about are not the same thing at all?
Once you do understand this, it will be like a light going on- you will instantly understand why the SS/tubes debate has been going on for so many decades.
if you can't hear it, it doesn't matter, 2nd, 3rd, 7th, ascending, descending, it is all MOOT if below the threshold of audibility. Of course, there are no tube amps I am aware of capable of that type of performance so it is entirely possible that your argument about the negative consequences of 7th harmonic is entire real for tube amps and why there are those who actually like the audible 2nd harmonic distortion that some tube amps produce.
try it! you know you want to!
Edits: 11/16/15
I really have no idea why you made this comment:
if you can't hear it, it doesn't matter, 2nd, 3rd, 7th, ascending, descending, it is all MOOT if below the threshold of audibility. Of course, there are no tube amps I am aware of capable of that type of performance so it is entirely possible that your argument about the negative consequences of 7th harmonic is entire real for tube amps and why there are those who actually like the audible 2nd harmonic distortion that some tube amps produce.
Could you explain the text in bold, as it seems to be at the heart of your comment.
I certainly agree that if you can't hear it then it does not matter. What you want to keep in mind here is that the ear converts all distortion into some type of tonality. The 2nd being 'warmth' for example, the 7th being 'metallic' and 'harsh'. The difference between the two examples is that the 2nd is relatively innocuous in "large" amounts while the 7th is annoying even though it might be buried in the noise of the test equipment and/or the amp itself.
It's all over ........ :)
Go Rossi ......
I mean with the amplifiers ....... :)
Go Rossi ......
(nt)
nt
try it! you know you want to!
nt
try it! you know you want to!
-
nt
try it! you know you want to!
- Cease responding to one anothers' posts.
- If you disagree with each others' response to a post, whether a matter of fact or opinion, direct your alternative perspective to the original poster (so long as it was neither of you). When doing so, do not refer to, obliquely or otherwise, or in any way comment in a manner that impugns, the other ; The Bored will be the sole judge of whether or not this has happened and will act accordingly.
- Failure on either of your part to hold up this directive will result in a ban.
- Don't use the Asylum Emailer services to communicate with each other.
Thank you for your immediate cooperation.
Fax mentis incendium gloria cultum, et cetera, et cetera...
Memo bis punitor delicatum! It's all there, black and white,
clear as crystal! Blah, blah, and so on and so forth ...
-
So What , you NCore CLass-D guys get together and compare graphs ...
:) ROFL
.
Go Rossi ......
nt
try it! you know you want to!
...and the mouse in your pocket.
nt
try it! you know you want to!
and some amps don't produce harmonic distortion that is audible. Thus, 7th, 2nd, whatever, if you can't hear it, it doesn't matter! Your comments on the conversion of harmonics to tonality only applies to those that are audible! Since there are no tube amps I am aware of capable of amplifying the input signal without producing any audible harmonic distortion, your comments are certainly relevant in regards to tube amps. There are, however, ss and class d amps that are indeed capable of amplifying the input signal without producing audible harmonic distortion and thus your comments about tonality are pointless in regards to amps which do not produce audible harmonic distortion.
try it! you know you want to!
Its like herding cats.
I don't know of a transistor amp wherein the higher ordered harmonics are inaudible even when they measure in trace amounts. Such a thing does not exist. The harmonic structure of all amplifiers is audible.
Since your education about the human ear is obviously limited, this conversation will simply go in circles. Instead of trying to make me wrong, you might instead take some time to study how the ear/brain system works. Obviously you are a proponent of class D; like all other amps class D has particular artifacts related to distortion that can be heard by the ear. If you at least knew that the ear can do such a thing it would open the up the possibility to do something about it. You can't do that if you are operating with a blind spot!
"...you don't want ANY 7th order harmonic distortion."(emphasis mine)
He doesn't mention anything about thresholds of audibility.
That's your strawman.
that anything below the threshold of audibility is moot. There is no straw man, sleepy...
try it! you know you want to!
...you have no understanding of what Ralph is telling you and Curl is saying.
Hint: they aren't talking about steady state sine wave measurements.
nt
try it! you know you want to!
-
then let me know how you or anyone else can hear its harmonic distortion when it's -100dB down and below the threshold of audibility. As you have said, "doesn't matter if it isn't audible...."
If you think it is audible with a complex signal, show some data to support you position.
try it! you know you want to!
Edits: 11/16/15
Think about the distortion riding on top of the energy of the waveform (a little calculus is helpful) and you will see what the problem is.
I couldn't disagree more strongly with one sentiment expressed in the linked article:
"In other words, it's not really worth trusting an audio reviewer who is older than you are, because there is a range of higher frequencies that you can hear while they cannot. "
While I could detect active burglar alarms at age 18, I couldn't begin to hear the kinds of differences among audio gear that older, trained ears I knew at the time could. There's more to music than the last octave.
In fact, I find getting the midrange right far more important than either extreme.
Music does not consist of isolated sine waves. Therefore, a test of how an amplifier responds to isolated sine waves can not completely characterize the response of the amplifier to music.
"Harmonic distortion" is not a fundamental characteristic of a system. It is one method of measuring the non-linearity of a system.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
nt
try it! you know you want to!
It has been a long time since I read of these things (I think they were in latin) and I'm sure/I hope there's been more research since but my understanding is that, yes, something like a 140dB dynamic range might be possible from a tone played very loudly to one played very softly (even below background noise) but certainly not simultaneously!
IIRC, the ear has a working dynamic range much, much less than that, which depends on the loudest sound and tone spacing. In those terms hearing a harmonic 140dB below the fundamental is just not plausible.
This masking effect is interesting because I wonder if the relatively high levels of 2HD & 3HD in valve amps swamps-out the subjectively deleterious effects of higher harmonics, rather than the claim that these amps are free of high order products.
13DoW
And that is in the range of max sensitivity of the human ear (audibility thresholds depend on frequency!). When you add in masking effects of multi tone sounds, it is MUCH MUCH less, no matter what the frequency.Keep in mind the playback media range as well- if you are listening to vinyl, the range is only 70dB or so, redbook, better, and dsd closing in 140 dB. So with vinyl, if your amp's harmonic distortion is -100dB, my understanding is the media is in capable of resolving it regardless! Lots of tube lovers also love vinyl. Maybe because its limited dynamic range masks some of the distortion of their amps which would otherwise be audible with cd....
All this talk about higher order harmonics, harmonic patterns, etc., is meaningless if your equipment is capable of reproducing the signal with distortion levels BELOW the threshold of audibility. Few amps are capable of doing this, certainly not tube amps, but there are some that can. Of course we'll then hear about how terrible they must be because they use feedback. And feedback is bad because it creates higher order distortion products. A circular argument that fails to understand that unlike SET amps and other tube amps, not all amps produce audible harmonic distortion.
Of course there is nothing wrong with enjoying that tube distortion. Many do...
try it! you know you want to!
Edits: 11/14/15
You seem to talk out of both sides of your mouth. In one place claiming that any distortion products below -80 dB are inaudible and in other places touting the use of very high amounts of NFB to drive distortion well below that. Seems to me that you're just cherry picking your arguments.
Based on relatively new work it seems the pattern of the distortion is quite important. However, it is quite well known that even several percent 2nd order is inaudible while even very low amount of high order is audible.
For many musical instruments, the second harmonic is actually louder than the fundamental, so it should come as no surprise that adding a couple percent of 2nd harmonic isn't audible as "distortion".
"Secondly", (ha!, I made a funny), odd order harmonics give instruments more "bite" or "edge". As a trumpet player, this is quite evident to me in the sound of various trumpet brands and models. This is why we describe various trumpets as having a comparatively more "symphonic" sound or a more "jazz" sound. Take two trumpet models from the same manufacturer - Yamaha. The Yamaha "Chicago" horn has a fuller and more sonorous sound than the Yamaha "EM" (Eric Miyashiro) horn, which has more "sizzle". The same concept holds true for audio system components.
Thirdly, with regard to masking and psychoacoustics, it has been studied and shown since at least the 1950s that these general conditions apply:
1. Lower tones are better at masking higher tones rather than vice versa.
2. Tones which are harmonically related and fairly close together are more easily masked than tones which are not.
3. The "success" of the masking is related to the relative levels of the tones.
Folks who wish to explore the topic in more depth should go to the Audio Engineering Society website (aes.org) and the Acoustical Society of America (acousticalsociety.org). Even if you're not a member, you can search and peruse abstracts and purchase the papers for a fee. The topic area is also covered in Eric Heller's excellent "Why You Hear What You Hear" and other academic texts.
:)
It all comes crashing down at clipping .... Big un over little un .....
Go Rossi ......
Well, when an amp clips then it is of course not making the same distortion pattern it was before it was clipping...that is why one should avoid clipping. That being said it has been demonstrated that an amp without negative feedback will recover from clipping much fast than one with a lot of feedback. Clipping recovery can go a long way towards masking clipping because it is usually at high volume levels where our hearing is less sensitive...if it is short enough.
I agree. The pattern of clipping is important.
I was given some clipped recordings to restore and I started doing some experiments by taking clean digital recordings (24/96) and deliberately clipping them and then doing listening tests. In most cases clipping one cycle of the waveform will be inaudible, while the same clipping applied to multiple cycles will be audible. If one note is clipped it may be hard to notice the harshness and it may appear as just a bit more dynamics by the musicians. However, if the damage repeats for several notes then it can be perceived as a defect, either with the recording or the musicianship. In some cases repeated distortion won't be perceived as a deficiency in the sound quality or musicianship, but the listener will get a headache from listening. (I didn't try to do any scientific tests to "prove" these conclusions, because they appeared to be completely obvious.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
It doesn't get worse than digital clipping....
Go Rossi ......
Some of the tapes I worked on had analog clipping and they sounded different than digital clipping. In one case, the analog clipping wasn't really on the tape, it was a very, very hot metal cassette tape and with the gain knob on my cassette player cranked up the analog output stage of the Nak cassette deck was clipping. After I discovered this situation though careful listening I changed my procedures and ran the gain up on my ADC and ran the gain down on my tape deck. (There wasn't any S/N problem due to loss of bit depth when recording 24 bit audio from cassette tape.)
The problem with digital clipping is that it voids the preconditions for the Nyquist theorem to be applicable. As a gross example of that, consider a 1000 Hz sine wave. With purely analog clipping one will get a series of odd harmonics, 3000, 5000, 7000, etc... But if you take a digital square wave (+1 and -1 samples at 1000 Hz) one will get all kinds of alias products that will be at odd frequencies, including beat tones below the 1000 Hz fundamental. These will show up on a spectrum plot (FFT) and will also be clearly audible, especially if you use a sweep tone. A sweep tone with analog distortion will move in one direction. A sweep tone with digital distortion due to aliasing will have "birdies" moving in the opposite distortion.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
...good sound and good technical performance aren't mutually exclusive and there are real-life examples out there.
I don't know of one. But we have to be clear on what is meant by 'good'...
This thread initially was aimed, more or less, at Stereophile and their measurement regime. Mostly I think we can look at THD- if less than 0.005% at full power, that is good. Is 1.0% at full power bad?
I've yet to experience an amplifier with super low THD that sounds good. I would not mind hearing one that **did** sound good with distortion that low, but right now I don't think such a thing exists.
So as a result we have to deal with amps that are low in the distortions that the ear cares about, and are otherwise probably fairly high in the sorts of distortions that the ear does not care about. The latter however are likely to measure "poorly".
t
If what is meant by "good engineering" expects a super low distortion figure ala the claims made by JA elsewhere on this thread.
At the end of a review of at least 1 iteration of the 275, he wrote something like "Good engineering is timeless."
Jeremy
But I suspect it does not measure the way that JA would want to call it "good engineering".
However I am not saying it is not well engineered. Although I don't have direct experience with the new ones I have plenty of experience with the old ones and IMO they were well built and accomplished what the designer had in mind. Reliable too- definitely part of "good engineering"...
Well, but he did write the remark at the end of his measurements section, so I infer that it represented his reaction to what he measured.
Jeremy
he thinks its not "well engineered"?
Croft amp
"To me, it seems, at best, inadequately engineered, and at its worst-that nonflat RIAA response, the high levels of harmonic and intermodulation distortion-just plain inadequate."
Do read ST's response. :)
-
...the Benchmark AHB2 as reviewed in Nov 2015 issue of Stereophile. You can read the review for yourself for details but THD+noise was almost beyond capabilities of test equipment. It also got a very good subjective review by Kalman Rubinson. Other reviews have resulted in similar assessments. All this with a tracking switchmode power supply no less. I haven't heard it personally but have no reason to doubt reviews at this point.
for more that 10 minutes without begging for a gun with which to shoot myself to end it all.
OK, it's an amplifier, not a DAC.
Never mind.
A friend of mine borrowed one some years ago...we had a similar negative impression. My inexpensive (but very good) Monarchy M24 DAC wiped the floor with it.
it's an amplifier with THX technology inside
Others, like Henry Wolcott have used a feed forward approach before and gotten similarly *wonderful* metrics. As have those using multi-rail power supplies.
It also got a very good subjective review by Kalman Rubinson.
"... but through a pair of Monitor Audio Silver 8 speakers, the sound was somewhat hard and thin. "
Yeah, give me more of that!
From the same review:
"Just because I found Benchmark Media Systems AHB2 not to be absolutely perfect under all conditions - after all, what is? - Doesn't mean that I want to represent it as anything less than a marvelous sounding amplifier. The AHB2 was capable of drawing more music from my B&W 800 Diamonds than I'd anticipated and mostly sounded better than the other amplifiers I've used to drive those speakers." .
Yeah, give me more of that!
I think "we" have a tendency to cherry pick data points out of context to support a personal POV.
I think "we" have a tendency to cherry pick data points out of context to support a personal POV.
Perhaps. I would never buy an amplifier, however, that sounds thin and bright - at any time.
Don't know quite where to go with that one so I'll pass. ;-)
Ah, so you haven't heard it yet and you put it forward as a good measuring and good sounding amp?? So what if KR liked it?
Are you saying KR's opinion is irrelevant or is it that the opinions of audio reviewers in general are irrelevant? Thin ice ahead!
BTW, he's not alone in his opinion as you must already know.
I am saying that you are putting forward this amp as a good sound/measuring amp and you haven't personally heard it. On top of that, KR found that the amp wasn't good in all situations (the Monitor Audio RS8 is not that hard of a speaker to drive). And on top of that, what KR uses for reference gear and tends to like is, IMO, rather unmusical and rather "hifi" ish gear.
I would like to know what someone like Martin Colloms or Art Dudley, or even Michael Fremer think of it before i would give it a second thought.
...You pretty much state that I err in putting forward an amp I haven't heard. However....by this reasoning isn't it also erroneous to dismiss the amp as you've done assuming you haven't heard it either? Seems I also have to conclude that you consider the opinions of reviewers irrelevant in general.
Curiously, you previously dismissed KR's opinion ("So what if KR liked it"), then you cite a portion of his review out of context in support of your personal point of view regarding the amp you haven't heard. You finish by stabbing KR thru his audiophile heart by describing his tastes in gear as "rather unmusical and rather "hifi" ish". The horror! This is all kinda inconsistent dontcha think? It also strikes me as cherry picking for best effect.
Even more curiously you state that you'd want Art Dudley's (or even Michael Fremer's - EVEN? yoiks!) opinion of the amp before giving it a second thought. I'll admit that I really like AD's style and outlook -followed him since the "Listener" days. However.......you must surely know that one of his reference speakers is a pair of Altec Valencias - possibly modified in undocumented ways. Have you ever heard these things? I have: I too am the proud owner of a pair of them. While dynamic and sometimes engaging, esp w/tubes, they're also colored with a "prominent" (shouty) midrange and rapid roll off above ~10-12KHz. IOW, "fun but fake". I don't begrudge AD for liking them but I also don't consider them reference quality, esp as a tool of the professional reviewer. If you can criticize KR for his "hifi" ish tastes in equipment..........
"by this reasoning isn't it also erroneous to dismiss the amp as you've done assuming you haven't heard it either?"
Except that it is a fundamentally wrong design from a psychoacoustic POV. Not to mention I have heard their other big product, the DAC, and it was frankly crap sounding. Given their design priorities and the methods they use to achieve them I can pretty safely conclude this amp breaks no new ground sonically.
"Seems I also have to conclude that you consider the opinions of reviewers irrelevant in general."
Where are you taking this from? Did I not state that there ARE reviewers whose opinion I value? Yes I did.
"This is all kinda inconsistent dontcha think? It also strikes me as cherry picking for best effect. "
How so? I said that I don't value his opinion and I explain that is largely two things A) He raves about gear that I have heard and find unmusical and not worth buying and B) His reference gear is largely made up of similar items, IMO. Nothing inconsistent with that and my dismissal of his Benchmark review.
"then you cite a portion of his review out of context in support of your personal point of view regarding the amp you haven't heard."
Where did I quote him at all? Perhaps you have me confused with someone else?
"I'll admit that I really like AD's style and outlook -followed him since the "Listener" days. However.......you must surely know that one of his reference speakers is a pair of Altec Valencias - possibly modified in undocumented ways. Have you ever heard these things? I have: I too am the proud owner of a pair of them. While dynamic and sometimes engaging, esp w/tubes, they're also colored with a "prominent" (shouty) midrange and rapid roll off above ~10-12KHz. IOW, "fun but fake". I don't begrudge AD for liking them but I also don't consider them reference quality, esp as a tool of the professional reviewer."
I said I value his opinion, not that I am a fan boy or want to duplicate his system. I have heard old JBLs and Altecs and some are amazing and some are as you describe. I haven't heard the Valencias. What you describe though doesn't surprise me. My Odeons sound completely modern in tonal balance and lack of coloration but have the fun and dynamics of one of these older horn systems.
You sure seem to reading a lot of your own interpretations into what I am posting rather than sticking with the facts about what I am writing...strange.
There have been tremendous improvements in SMPS units in the last 20-25 years so I don't discount a product that might use them.
However the specs look a little too good to be true and some of the claims on the website are pushing the envelope a bit; bottom line is we'll have to see on that one. I will look for it at upcoming shows.
So why no SET's Ralph or Hybrids, why only OTL's ......?
Go Rossi ......
-
the "good technical performance" criteria is largely irrelevant.
The fact that any given amp may sound good has very little to do with the conventional metrics used to determine such.
It's a crap shoot when you base the determination on metrics which have little correlation with actual sound quality.
My point in the original post
.
...I was commenting on statements like this one: "The fact of the matter is that amps that 'measure well' don't do the best when dealing with the human ear/brain perceptual system. What they do well is look good on paper. The problem here of course is we don't experience music through paper, we experience it through our ears and brain." . Actually, the fact of the matter is that there indeed exist amps that measure very well AND are generally considered good sounding. i.e. Good measurements do not preclude good sound.
While it's probably reasonably true today that the correlation between good meas performance and good sound is mildly positive, IME, the correlation betw poor meas performance and poor sound is considerably more positive. Preference also has a role to play in this. I'm fairly certain a few of my audio acquaintances (and a noted reviewer for SP) find relatively higher levels of 2nd hd and amplitude compression very pleasing....not that there's anything wrong with that...
"While it's probably reasonably true today that the correlation between good meas performance and good sound is mildly positive, IME"
Earl Geddes actually found out that it is a slightly negative correlation but of course the R^2 is rather poor...so probably no correlation is more correct.
The only reason there is no correlation is that THD is nearly meaningless and so is IMD. The specific harmonic pattern IS significantly more important (See Gedlee metric or Cheever metric). This is the point that Ralph is trying to make. The metrics ALL indicate that low order harmonics even in relatively high % are benign and the higher order harmonics need to fall off exponentially to essentially zero. This monotonic PATTERN is more important than having very low, but evenly distributed harmonics that nearly all push/pull SS amps with negative feedback produce. This pattern is wrong to the ear/brain...if it wasn't then you would not have the poor correlation with THD and IMD that clearly exists.
Now, you claimed that there are some very good measuring amps that also sound very good. Care to name them? I have yet to hear one and I have heard nearly all of the top praised über SS amps, most of which measure tops. Ironically, the best sounding SS amps are the ones that measure somewhat poorly by the usual industry benchmarks (Pass, darTZeel, Vitus etc.). Please don't tell me that Soulution sounds good...I am not sure which psychadelic coolaid the reviewers are taking when they praise this brand...it just isn't on I tell you. Sure the latest and greatest SS beasts don't sound hard or etched like a poor SS amp from the past but they also don't sound really natural and still electronic or "hifi".
Based on psychoacoustics, NO push/pull amp, whether Tube or SS, actually qualifies as correct from an inherent design flaw...the cancellation of even order harmonics. That is usually further exaccerbated by the addition of negative feedback and Class A/B that further reduces low order even and odd harmonics and gives rise to an infinite series of high order harmonics. The maths are clear as are the measurements. With all due respect to Ralph, his OTLs are wrong as well but perhaps less wrong than most.
Nelson Pass also realizes this and his white paper on distortion is a very interesting read.
Given we have no truly linear amplification devices at our disposal (if we did then these discussions wouldn't exist) the best we can do is to "hide" the distortion behind our perceptive blind spots. SET done right gets the cloeset to mimicing our ear/brain pattern and puts those distortions in our blind spot...mostly. The only real downside to SET compared to other designs is power and cost.
For sure a short cut in SET execution will result in the bloated soft sound that many accuse the design for but it is the execution. A good output transformer that won't saturate within the power bandwidth is neither easy nor cheap to make. Making sure the driver doesn't distort before the output stage takes a sharp designer as well. Also, load line setting and parts matter because there is no negative feedback to squash it all out.
I was commenting on statements like this one:
Perhaps next time you might quote the text to which you are responding.
Good measurements do not preclude good sound.
Certainly not. Nor do they have any correlation to good sound. Usually the converse is true.
While it's probably reasonably true today that the correlation between good meas performance and good sound is mildly positive
If you refer to THD measurements, I find the converse to be true. Those with "heroic" distortion metrics (via boatloads of "corrective feedback") usually sound unnaturally lean and flat dimensionally.
Setting aside good measured performance, does poorly-measuring equipment sound good despite that or because of it? Is there some quintessence of music that can only be conveyed by less-than-stellar objective performance or does that poor performance add something that makes it sound preferable?
Steve O thinks that there is a stronger correlation between good reviews and poor measured performance than between good reviews and good measured performance. And, in regard to THD, you agreed with him.
Not wishing to put words in anyone's mouth, but the point then is that if well-measuring equipment can sound good it is more likely that poorly measuring equipment is preferred for what is adds rather than what well-measuring equipment cannot do. So, in those terms, measurements do matter.
"it is more likely that poorly measuring equipment is preferred for what is adds rather than what well-measuring equipment cannot do. So, in those terms, measurements do matter."
This is not the case. What is defined as "poor" measuring is that it has a somewhat high THD or IMD compared to "state of the art" but if the harmonics are in the human perception "blind spot" then it is acutally the other way around in terms of which product has good measurements. Those that measure poorly are high in low order harmonics relative to high order harmonics...this is a pattern similar to one observed in nature. Those that measure well have disrupted this pattern because they have pursued the ENGINEERING goal of low absolute distortion rather than the SONIC goal of low to no higher order distortion.
.. it shouldn't become a hall-pass for designers to not do a good job.
You can't help getting high 2HD with something like a single-ended triode so how much care/knowledge does the designer really need? But, you are correct about it all depending what the goals are.
I do wonder whether high levels of consonant distortion are subjectively better subjectively because they mask-out dissonant distortions or because they just sound nicer on their own. But, in the great scheme of things, it probably doesn't matter too much.
Regards
13doW
". it shouldn't become a hall-pass for designers to not do a good job." If it measures good and sounds bad does the designer get a pass for producing excrement with good specs?
Define not doing a good job, please.
"ou can't help getting high 2HD with something like a single-ended triode so how much care/knowledge does the designer really need?"
Very much actually because the goal is to keep the distortion low order and prevent the rise in high order harmonics...this means a lot of care in the design of each stage of the amplifier and careful balancing of operating points, passive parts choice and most importantly using a great transformer (at least the best that is affordable for a given design price point). It is a lot harder than making a mediocre design and cleaning it up with copious amounts of negative feedback.
" I do wonder whether high levels of consonant distortion"
Let me be clear, there is no such thing as "consonant" distortion...all distortion degrades fidelity but some does much more damage at much lower levels than others. Read the articles in Stereophile by Keith Howard where he added different distortion patterns to different music selections. He found no added distortion was the best (no surprise) but he also found that some patterns were far worse sounding than others. The monotonic pattern that was championed by Hiraga was found to be the least offensive of the distortion patterns he tried. This is the same pattern that Cheever concludes should be the least audible and that Geddes data also corroborates. Now, ideally we have a linear amplification device that gives no additional harmonics or IM distortion but since all our devices are "bent" then that ain't going to happen.
JA often calls triodes without feedback "bent" but the truth is that the triode is the least "bent" of all amplification devices, with the BJT and the Pentode being the most "bent". That is the true reality.
I think the white paper from Nelson Pass regarding negative feedback is interesting and illustrates quite well what is wrong under the surface. You don't really get rid of distortion but you seem to just push it around...Crowhurst said that the result is a kind of signal modulated "noise" floor that will mask low level resolution and damage dynamics.
And yet , regardless of what you , me and others think, many more like class-d sound over SET sound, oops, so much for the theories.
Since it's all wrong, whats next ....?
Go Rossi ......
Edits: 11/13/15 11/13/15
What I find is this:
Tubes and SS...
"Vocals are run through passive Renkus Heinz mono PA with Cary Audio Hi-Fi tube amplifier...
The other end is a two channel hi-fi playback system with Revel Saloon II speakers coupled with Levinson solid state, bi- amps. Rogue Audio tube pre - amp, Benchmark DAC , Esoteric transport and Sony Media Server. "
"I've owned numerous equipments..."
Start about 9:55. And none of it class D. :)
There is a room with Wilson Alexandrias in it. So, what does Dave Wilson use for making his recordings?
Answer here .
How about when Wilson shows their product at shows? Find out here .
Dave definetly know how (i have a few ) to make recordings , I'm also a big fan of prof Johnson reference recordings ....
!
Go Rossi ......
Edits: 11/14/15
if your point was simply the need for room treatments (to me an obvious requirement for any high performance system), then I heartily agree.
How many audiophiles spend big $$$ on the hardware and not the room?
I will bet Hypex has sold more ncore amps in the last 2 years than all the SET manufacturers combined....One would think that if they were all that, more people would be rushing to buy one...
No doubt SET offers something some people like, just as class d does. Trying to build a theory for their "superiority" is like trying to "prove" Thai food is superior to Italian....
It's also not necessary for enjoyment of one's hobby. I fail to understand the need to find some sort of rationale as a basis for the ownership of a certain type of equipment, or the need to belittle what others enjoy. I suppose there will always be those who need to feel superior, small and insignificant as it may be....
try it! you know you want to!
many more like class-d sound over SET sound
I have yet to meet someone who actually stuck with Class D. It definitely sounds different and for many that is good enough.
I know a lot of guys who tried it and tried to live with it though and all gave up on it.
I haven't given up yet, but if it takes 50K to hear what class-D sounds like,i will have to take Kurby's word for it..Have to sully myself with the usual bad SS stuff ... :)
Regards ..
Go Rossi ......
Edits: 11/13/15
For my own part, I think class D is the rising star in audio amplifier technology, but so far hasn't brought home the bacon. But that might be different tomorrow- the technology is still in a steep portion of the price/performance curve.
If you need $50K to understand what its about- well, traditional technologies need nothing of the sort to demonstrate what they do in spades. I think I'll wait too...
I think class D is the rising star in audio amplifier technology...
when implemented as a "Power DAC" or using other means (Devialet's current dumper/Stasis) to eliminate the prodigious amount of HF noise.
"Good measurements do not preclude good sound.
Certainly not. Nor do they have any correlation to good sound. Usually the converse is true."
If the converse is true, there is a correlation, a negative correlation, yes?
"If you refer to THD measurements, I find the converse to be true. Those with "heroic" distortion metrics (via boatloads of "corrective feedback") usually sound unnaturally lean and flat dimensionally. "
Not necessarily THD alone since there are other measures of linearity and maybe even good sound. Also not clear what you mean by "heroic". FWIW, an amp with low THD and other good traditional meas ( thru use of corrective feed forward ) is also getting tairly decent listening reviews too...the new Benchmark amp. I want a pair!
If the converse is true, there is a correlation, a negative correlation, yes?
Yes.
Not necessarily THD alone since there are other measures of linearity and maybe even good sound.
To wit?
To wit!: IMD, SID, smoothly decending levels of high order distortion products (there's name for this condition but it escapes me at the moment), basic stability, rapid overload recovery and so on. Optimizing any one or two parameters is probably insufficient so it's really an exercise of judicious balance. Much of this stuff was known and effectively dealt with in the middle of the previous century by a few designers.
Geddes found that IMD also has no correlation (actually slightly negative but statistically insignificant at the 95% confidence).
An interesting paper by Geddes. Since you've cited Geddes on a couple of occasions here, I assume you're familiar with his paper.
While I'm not qualified to directly butt heads with Mr. Geddes, what he describes in his paper raises a number of questions.
1. The testing methodology employed a 15 second excerpt from ALW's "Phantom of the Opera" ripped from the redbook CD and converted to a .wav file. This was considered "the reference". Is 16/44.1 sufficiently resolving considering the potential subtleties involved? Is the music selection sufficiently revealing? Is a 15 sec sample of sufficient duration?
2. Twenty one different stimuli (distortions) were mathematically simulated (calculated). The exact nature of the distortions is unclear as is the correlation of them to real life distortion generating devices. The calculated distortions were then used to multiply the input reference wav file to produce the test sample or signal. All stimuli were digitally adapted and presented to the subjects via computer...as 16/44.1 wav files. Is a mathematically generated distortion a valid means of simulating the performance of a physical device? Has this methodology been validated by Geddes or independently by others?
3. Subjects rate the test signals as better or worse than the reference. Doesn't this predispose the results towards "preference" instead of "accuracy"?
4. Geddes finds a weak negative correlation between THD and IMD and preference but essentially dismisses this finding based on statistics and pursues it no further. He also finds a moderate positive correlation between the proposed Gedlee measure (Gm) and preference. He then chooses to improve the statistics of Gm by throwing out three unsupportive stimuli. Is this appropriate?
5. The paper is 12yo. Has there been any independent validation of the methodology or verification of the results? Have there been any additional developments or refinements of the Gm concept since 2003? Is any independent mfgr using Gm in place of or as a supplement to more traditional measures? A quick web search turned up nothing but not being a member of AES may be a limiting factor.
Overall, I found Geddes' findings and proposal to be interesting. But really isn't this just one relatively old data point and another dead end in the search for a means of correlating objective device performance with subjective preference or "goodness" using one or two measures?
1) I would assume it was chosen specifically it's revealing nature and i think 16/44 is sufficient.
2) There is another paper that discusses the distortion in more detail (it is a theory paper).
3) No, preference is a better way IMO as it takes off some of the pressure of "do I hear" vs. "what do I hear".
4) I think the findings of THD and IMD mirror what a lot of audiophiles out there complain about. I do not think that what Geddes does with his own metric is all that valid (one data point maybe if it fails a Q test but not three). His metric suggests he is on the right track but the fact of a weak correlation means there is something missing from his model to make it more complete. Maybe Cheever's model would fair better? Apparently, Shorter's simpler model was not that effective either or maybe more would use it.
5) I found a more recent pub from a Swedish (or Danish) group that reached similar, but more difficult to understand, results
Based on the synthesis of all recent research I have read I reached the conclusions that I posted some time ago on the amp/preamp forum.
http://www.audioasylum.com/forums/amp/messages/20/208108.html
I stand by my analysis until such time that I hear something or see research that falsifies what I wrote.
I posted a link where you can take the test yourself ,
Geddes findings and research is valid, what is not valid is how much of it actually correlates to us in actual use , when one hears their favorite song vs non familiar music , our sensory system will and usually does change gears...Short of directly taking and using amplifiers with the distortion characteristics suggested by Geddes research with golden ear audiophiles , what are we really proving....
Regards
Go Rossi ......
Edits: 11/14/15
Geddes found that IMD also has no correlation (actually slightly negative but statistically insignificant at the 95% confidence)- Morricab
Go Rossi ......
nt
try it! you know you want to!
Yep, that's the one
That wasn't difficult, was it?
Much of this stuff was known and effectively dealt with in the middle of the previous century by a few designers.
Then such got lost when SS first arrived and the arms race for "better" metrics and worse sound was prevalent. :)
"Then such got lost when SS first arrived and the arms race for "better" metrics and worse sound was prevalent. :)"
Seems that way although I don't believe "worse sound" was an actual design goal back then. I'd guess that the degraded sound was the result of the pursuit of unbalanced objectives with advertisable power and distortion specs taking precedence at a time when SS technology was not well developed. I have to believe the better designers knew what was going on but weren't in a position to resist.
The problem is the outcome of the techniques used to lower distortion, namely push/pull and negative feedback. The outcome gives the desired ENGINEERING solution of low distortion but doesn't achieve the SONIC soluton of having a proper harmonic content for good sound.
I'd guess that the degraded sound was the result of the pursuit of unbalanced objectives
Absolutely and to the point of this thread - chasing metrics that fail to correlate with qualitative aspects of the listening experience.
Which continues today with some designers.
Ralph,
I don't really disagree, except I'm still really pleased with the plane analogy:). I think we all should step back occasionally and consider what humans can and do create using our accumulated knowledge and contrast that to the some of the undefined minutiae of high-end audio.
There are many, many different amplifier topologies and philosophies and pretty much all reviews I read are positive so is there one true approach to amplification - probably not? Though each proponent probably thinks so. Add into that cost, appearance and peer approval and, I'm sure, that is as significant as a bit of 7HD.
So, is there a secret to perfect audio and how do you test for it? Until we have that answer the only examination question is measure what you can. If the results look good on paper you might think 'at least this designer has tried to be faithful to the input', if the results look poor you really don't know if the designer has studiously eschewed measurements for subjectivity or just doesn't have a handle on what they are doing.
The Volkswagon analogy doesn't quite work. If an amplifier 'looks good on paper' but is found subjectively lacking, where is the cheat? What VW did was design for good subjective performance and then cheat with a test setting to make the measurements look good too.
Regards
13DoW
But then you when you listen it doesn't sound right. That's because the specs that look good on paper don't sound good to the human ear because the specs on paper aren't right.
In the case of VW, it pollutes when it shouldn't. kinda but not quite the same thing, unless like me you think that too much 7th order is pollution...
The Bottom Line is we don't measure the right things, and further, it does not like we will be measuring the right things anytime soon- see JA's response below in this thread.
VW had 3 options,Option 1: Bad ignition switch that kills ..
Option 2; Bad AIRBAGS that kills ..
Option3: Spew a little more diesel nox , harmful to ...?
VW went for option 3, fibbed about and had happy owners dancing at 800mls a tank full .... Yeah !!!!
Go Rossi ......
Edits: 11/12/15
I think the question might be "would you necessarily fly with a pilot who had passed the exam for a Masters in Fine Arts"
:-)
I was really pleased with my analogy so I don't agree. But, yours does remind me of audio reviewers who after regurgitating enough manufacturer copy think they they've become technical experts.
Further down the discussion, we DO have a quote from a technical expert:
"Richard Heyser, JPL scientist and engineer extraordinaire made the following observation in an article in Audio Magazine :
" I claim that it should be possible to measure audio systems and have those measurements correlate with what we hear out of those systems. We are not doing that now. Our measurements are more precise than ever, but our understanding of what those measurements mean to the way a system "sounds" is still hazy.I further assert that we are locked into that dilemma because we do not truly understand the meaning of those technical concepts which we now use. I don't think I can be more blunt about the matter."
To simplify the analogy: We can perform a whole series of measurements on 2 different glasses of wine to as many decimal points as we wish, complete with graphs and powerpoint display, giving us the illusion of being "scientific", but these cannot distinguish the plonk from the fine bordeaux.
Again, the question of surrogate endpoints.
link below
My point exactly
What a sad situation high end audio must be in that a nearly 30yo article describing difficulties of the design process is still cited as support for what's understood to be current practice.
yes, it is pathetic
"If it sounds good and measure bad, you are measuring the wrong thing"
I would more likely be inclined today to say, "If it sounds good and measures bad, then your interpretation of what is a good measurement is wrong" because most of the right data is possible to measure today.
Try Keith Howard's Distortion simulator for Kicks ...
Go Rossi ......
Tried it many years ago in fact with a solo violin piece that I recorded. I even added distortion patterns from real amps that I took off Soundstage.com! It was good fun, you should try it.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: