Sound Quality Vs. Measurements

Status
Not open for further replies.
I had this problem at the pub . My friends were laughing their heads off about cables and also calling people who make them names usually reserved for criminals of the worst kind . I had to jump in and say I had recently designed a cable for a friend . I told the friend I do not beleive in such things however following the discussions already in circulation I gave him a recipe . I did it on one stipulation . The cost must be strictly Aldi supermarket . That is to sell it at the lowest possible price . Apparently it is a great success . The point is it cost $--- --- to tool it . Already 30% of the costs are repaid in 4 months . Even my friend is surprised . I haven't looked at the reviews and don't intend to . I will say this . It would almost certainly pass the most stringent safety tests including fire . As I said at the pub the customers insist on this with their cheque books . Any audio company who doesn't follow the customers wants is foolish . It is possible to do these things whilst still having some self respect . As John at the pub said he will use Van den Hul cable as it has customer acceptance and speeds the work due to nice handling qualities . That is became Mr Van den Hul is a proper engineer who without going too far meets the customer needs .

The worse thing about the people who make the suspect things is they also appoint themselves as pseudo doctors . They have voices trained in the art of gentle persuasion . Many people fall for it . Someone said at a hi fi show meal one of the professors of audio would not talk to him as he was of no importance ( for importance read money ) . I used a word I reserve for few and it doesn't begin with F yet has four letters .

I just had a dig at one of the doctors in a turntable thread . This professor of audio responded to my suggestion of a cheaper / better motor by saying " you will have to show me that " ( said with the usual gentle tones ) . In other words give him that gratis . The same people who rip the public off are quite happy to get it for nothing . This guy puts himself up as an expert . Yet he had no idea how to do some very basic electrical engineering . So I posted it free . It was no big deal . I did it in 2 hours beginning to end . DF 96 might have done it even quicker .

I have to be fair with the guy in question , most of what he does is fairly priced and I suspect he doesn't claim to know everything . The getting it for nothing them selling it on is my gripe . He probably thinks if I need paying I should say so ? Not is my world you don't . You play with a straight bat or not at all .
 
Last edited:
Highly regarded by who? Journalists, who routinely praise equipment with basic design flaws? Punters, who buy the stuff praised by journalists? Shops, who sell the stuff praised by journalists? Or competent audio designers, who recognise competent and creative design when they see it?
I'll spend two more cents :)

anyone heard the theory that some reputable and established company uses some 75% PR/25% EE guy as a marketing gimmick and that he has no idea what's inside the products his employer makes?

ginetto61, based on your last posts I think that you're going through a phase. when I found DIYaudio in 2003 it looked like a wealth of info. lots of knowledgeable people, so much stuff to learn. nowadays I tend to disbelieve most of the stuff I read here until I find some sort of proof. there are some few very vocal users who spread... let's call it stuff. most troubling thing is that they believe it themselves. some of them even started companies. some even get promoted to the rank of reputable audio designers. and when you start asking questions you eventually become ostracized. someone will eventually tell you that they don't take kindly of opamp lovers or something along the lines. you quote Douglas Self and you're persona non grata.

I believe that the audio high end field follows a Gaussian curve, just like everything else in the material world. some guys are terrible at it, some are great, most make great looking cases which reproduce a decent copy of the input signal.

I'm also interested in bikes as a hobby. I used to Google for info until I realized the amount of BS and hearsay that floats around. then I bought a book and learned in a matter of days more than I would've on forums in years. you are asking some fundamental (but nevertheless good questions) that are very hard to answer. not because the answers are complex but because a lot of factors that involve the most fundamental human behavioral patterns are at play.

yes, I'm making new fans with each day that passes :)
 
Highly regarded by who? Journalists, who routinely praise equipment with basic design flaws? Punters, who buy the stuff praised by journalists? Shops, who sell the stuff praised by journalists? Or competent audio designers, who recognise competent and creative design when they see it?

Good point Dave, except perhaps for the last part: speaking strictly personally, I would say that I don't really care what other designers think of a design I happen to like, nor the rest mentioned above, to me, MY ears are the Lord God Of Judgement.

I've been let down, sometimes sverely, by too many highly touted designs for it to be any other way.
 
If you can measure it, how can you avoid having them sound just that little bit differently?

I can measure lots of things I can't hear. So what? In a competent design (and the vast majority of midmarket is competent), any variations in the final product due to component spread fall well below any known thresholds of audibility.

I'm strictly speaking about boxes of gain here- transducers are a different matter.
 
dvv said:
Good point Dave, except perhaps for the last part: speaking strictly personally, I would say that I don't really care what other designers think of a design I happen to like, nor the rest mentioned above, to me, MY ears are the Lord God Of Judgement.

I've been let down, sometimes sverely, by too many highly touted designs for it to be any other way.
Back in the days when I was a professional computer programmer we had a system for reviewing each other's code. This was mainly to catch the bugs which didn't show up in testing, but it also gave us a chance to learn from each other. Peer review is a good, although imperfect, system.

Designing electronics and writing software have a lot in common; maybe that is why I have done both (one for money, the other as a hobby). A good design ought to be recognised as such by the designer's peers. To be highly thought of by your peers ought to be more valuable, although less lucrative, than praise from your manager or customers.

'Highly touted' can mean a number of things:
- highly regarded by peers, after careful consideration
- highly marketed
- self-publicised

I continue to be astonished (although by now I ought to know better) by the number of highly praised, frequently bought commercial items whose circuit diagram (when it appears on here, often from a purchaser wanting to 'upgrade' it) shows serious design flaws. There are people out there making their living from audio who seem to know less than a competent DIYer, and who seem never to have had their designs subjected to proper in-house peer review. Either their boss doesn't understand electronics, or he doesn't care.

What may happen is that a 'designer' is hired in on the basis of his 'experience', given a rough spec and he comes up with a design. Nobody checks whether the design is good or not, or whether he has a track record of good design. I saw this happen in software, but there we did try to check. The fee charged by a freelancer bears little relationship with his competence. So the company buying in expertise takes pot luck who they get, unless they are careful. The sad result of this is that a really good designer probably can't charge much more than a useless one, if they both have similar years of 'experience'.

I remember once working on a project which hired in a 'technical author' to write some manuals for the end-users. Part of my job was to check the results for technical accuracy. I was shocked to find that I had to correct his English. He was shocked to get his drafts coming back with so much red ink on them! Maybe I was naive; I just assumed that a 'technical author' would write better English than a computer programmer.

As I have said before, I don't trust my ears - except that they may alert me to a design flaw which I have overlooked but then need to investigate.
 
What may happen is that a 'designer' is hired in on the basis of his 'experience'
I'll try to search for the article where this was discussed. to give a preview, it was about a guy that worked in the Israel army and realized that experience and competence are less related than it would seem.

The fee charged by a freelancer bears little relationship with his competence.
guy once told me "never fix stuff for free, even for friends. even if you mess it up, it'll be because it was unfixable to begin with, not because of your incompetence".

I just assumed that a 'technical author' would write better English than a computer programmer.
LOL
I know what you mean :D
 
Last edited:
diyAudio Member RIP
Joined 2005
The sad result of this is that a really good designer probably can't charge much more than a useless one, if they both have similar years of 'experience'.
I'd generalize this to almost any human activity, certainly any human "profession". The correlation between competence and compensation is weak at best. And "experience" is generally a factor, true, but the actual track records within that experience are frequently dismal --- but hey, it's experience!

My favorites include executives, particularly highly-placed ones. They often have the good sense to get out "while the getting is good", and their resumes glow with spiffy achievements, the most prevalent is cutting jobs and "streamlining" or "restructuring", which show an immediate reduction in costs and apparent improvement in profits, but result often in greatly diminished performance in the long run. But they are usually out before the long-term effects take hold.
 
One famous manufacturer had not used any bias on his first amplifier . My brother when repairing one pointed this out . A simple fix was devised . The strange thing is the amplifier sounded OK and became a cult product . Years later it's designer admitted he was learning at the time . When I said how come it worked he said perhaps he had enough hiss to do something ( like dither I guess ? ) . The fixed version did sound better ( warmer and more detailed ) . Some high spec Japanese amps never had it's charm . What should be said is his dad owned one of the largest makers of stereo systems . These were sold through catalogs along with clothes etc . His dad would have said what to do if he had seen the circuit . That's how these things happen . Daft and typical . He is very successful now and his products have some of the better specs . His customer service was fantastic I should say and priced below what should have been possible . The parts were in the vital areas a cut above average considering it's price . Most of all he is a decent person and honest . He never laughs much . I think he has found it hard to be an audio man .

Talking to this man recently I asked what he knew of Alan Sugar ( Amstrad and a TV show ) . He said Alan had a van and was delivering something to one of the largest electronic companies ( aged 17 ) . He had a look around and asked to see the boss as he had a few ideas . The man turned him down . Incensed he said , OK I will do it and show you .

Before you say it I will never earn a living as a writer . My English is dreadful and getting worse ( I invert phrases Yoda style ) . Alan Blumlein also it is said !
 
Last edited:
"... higly specified ..." - now, THERE'S a can of worms.

I can't even remember all the times I was let down by higly to extremely well specified products. I realized in the late 70ies that specification have, practically speaking, nothing to do with sound quality.

An ultra low THD and IM spec would suggest that the amp sounds good, yet far too many such highly specified amps don't sound good and involving at all. It's not just getting the specs, it's much more HOW you get them.

Experience has taught me that if a device uses wild amplification factors and lots of global NFB, chances are it'll sound shut down, dark and/or bland.

The other extreme, no global NFB, sounds much better, but also not right - I have yet to hear any such product which does not sound sort of unfinished, short of the mark, a little loose.

It seems that in audio, as in life, it's the golden middle that gets the prize.
 
As I sometimes say, the right amount of feedback is the right amount of feedback. People who believe that 'none' must be best (however poor the amp), or that 'lots' must be best (to hide a poor amp), are both missing the plot. Sometimes 'none' is the right answer. Sometimes 'lots' is the right answer.

However, I am sometimes suspicious of claims that high feedback amps are bland or clinical. I wonder whether such folk are merely expressing a preference for a little distortion with their music. Too much feedback would actually lead to either instability (or at least ringing) or sharp clipping; neither of these would sound bland.
 
As I sometimes say, the right amount of feedback is the right amount of feedback. People who believe that 'none' must be best (however poor the amp), or that 'lots' must be best (to hide a poor amp), are both missing the plot. Sometimes 'none' is the right answer. Sometimes 'lots' is the right answer.

However, I am sometimes suspicious of claims that high feedback amps are bland or clinical. I wonder whether such folk are merely expressing a preference for a little distortion with their music. Too much feedback would actually lead to either instability (or at least ringing) or sharp clipping; neither of these would sound bland.

Dave, there are always exceptions to just about any rule I can think of.

Think back a few months to the example I named, and supplied its schematic diagram, the German made LAS power amp. It is truly an outstandinding example of a strangely designed amp, which had no right to sound as good as it did, but it stuck its finger in my eye and managed to sound really good. Seriously good, even if I have heard better even then.

1.5 Ohm/17W emitter resistors? Using power devices rated at 100V with +/-50V rails? Isn't that shaving it too close? Using a wildcard of sort, since BFT28A transistor was made by a very lonely California company only? Open loop bandwidth iof just 4 kHz? Lots of global feedback? Zero input stage degeneration? No resistors even for the current mirror?

Yet, just 5 minutes with it will convince you that those people knew EXACTLY what they were doing, and the some.

Obviously, using just that one example (and there are more), it's perfectly clear that it's quite possible to use a lot of global NFB and still end up with an excellent product. Even, in some cases, a very advanced product most manufacturers have not made to this day, and that one was from 1978 or so.

I agree with you that it's not just the feedback scheme. If you design from the start for high global NFB, you'll very likely go about it differently than otherwise, and may well end up with a good sounding amp. It's the undecided, the half way houses, who suffer the most, in my view.
 
Dave, here's a cute test you will end up loving (I think).

Step 1. Go to any e-bay (German, British, etc) and find a Harman/Kardon integrated amp model 6550. It's rated at 50/70W into 8/4 Ohms. It uses just 17 dB of global feedback.

Step 2. Refresh it, meaning exchange each and every capacitor inside, starting from the two main electrolytics. Do not install parts obviously way above what was originally used, although I would suggest you use silver mica for small value caps.

Pause after Step 2. Listen to it, learn how it sounds. Then go to Step 3.

Step 3. Exhange the shitto volume pot for an ALPS Blue.

Step 4. Meditate and prepare yourself for the surprise of your lifetime once you hear it. Space, air, stage width, depth and heigth like you would never expect from a commercial product, and rarely hear even with High End devices. Yet, sound full of energy and impetus.

It could be your prized component for a long, long time. Frankly, I have never heard such high quality sound from a mass produced component in my life, and believe me, I've heard a lot of them.
 
Hello !
I have a curiosity.
Let's take, as an extreme, a single ended zero feedback tube amp
then select a speaker that sounds very good with the SET above mentioned.
Then take a solid state high feedback amp and use it with this speaker
What would it be the sound ? bland and clinical ?
I do not know really, but maybe it could be a worthwhile test
The comparison must be made with same speakers.
I think it is almost never the case
Kind regards,
gino
 
Last edited:
As I sometimes say, the right amount of feedback is the right amount of feedback. People who believe that 'none' must be best (however poor the amp), or that 'lots' must be best (to hide a poor amp), are both missing the plot. Sometimes 'none' is the right answer. Sometimes 'lots' is the right answer.

However, I am sometimes suspicious of claims that high feedback amps are bland or clinical. I wonder whether such folk are merely expressing a preference for a little distortion with their music. Too much feedback would actually lead to either instability (or at least ringing) or sharp clipping; neither of these would sound bland.


I once asked Michael Gerzon about this , his answer was much the same . I started by saying what were his thoughts about negative feedback . His answer " well , I never have , had I I might have said this . Some amplifiers need a lot and some virtually none , after a pause with a giggle " you can be sure of one thing , nearly all will have the wrong amount " I didn't need him to say more , however he said this . Suppose you have an amplifier with either 3 or 4 stages , both cause problems . What you need is 3.7 stages . Obviously Michael although very good at maths and the most difficult problems obviously knew less about amplifiers . For once in my life I kept my mouth shut . I was tempted to say degenerate the VAS a little . Arguments rage about that and as Douglas Self wrote to me it is the worst idea of all as it doesn't bring the expected benefits . I do it because it helps match stages better ( lifts the base voltage for one ) . Sound wise 0R = punch 47R = soft and spacious 16 R = just right . This was 8 mA VAS and 2 mA LTP . Measurements were mostly identical either way . I never tested IMD for this , my hunch is it has an effect . This seems even more effective with 1970's amps . Raise input current , add a VAS emitter resistor . The Goodman's 80 has a diode in the emitter . Replacing that with a 56R resistor seems to do a lot of good . DC conditions are unchanged . When I see valves with LED's in the cathode I often wonder if it is a good idea due to the Goodman's . The Goodman's was terribly harsh before the mod . Transistor sound or IMD . The guy my brother did this for paid £100 for the work . His joy was taking it to hi fi shops to defeat all-comers . The other mods were polystyrene , polyester , high grade electrolytic caps . Also DC levels checked and set . We often worked together . Me with the ideas , him at the test station . £100 would be £300 now .

I have told these stories before so please forgive .

I don't think people who prefer low feedback amps are wrong . Equally I have heard plenty of high feedback amps that sound great . Take my favourite cheap op amp MC33079 ( 78 ) . I much prefer it with gain 1 over gain 2 . 2 sounds less focused . Up at gain 62 it also sounds fantastic ( MC stage ) . Between 2 and 62 I am not so sure . A friend I built a preamp for is both musician and PHD in electrical engineering . We worked through this for fun and did find that truly there was evidence to support this . Our conjecture is that unity gain was so much a priority that it fashions the performance . I prefer 33078 over NE5532 . The one thing I would like is a MC33077 single version with all the 5534 tweaks ( comp , offset ) .
 
Status
Not open for further replies.