Head-Fi.org › Forums › Equipment Forums › Sound Science › Testing audiophile claims and myths
New Posts  All Forums:Forum Nav:

Testing audiophile claims and myths - Page 195

post #2911 of 3206
Quote:
Originally Posted by ab initio View Post
 

 

Wait a second.... are we accusing the leading manufactures of electronic test instruments of selling test equipment that doesn't meet the performance specifications because they didn't use audiophile components?

 

No scientist in their right mind conducts a measurement with an instrument incapable of resolving the desired signal. This is experimental methods 101.

 

Cheers

Precisely. But not for not meeting specs - only not trying to do best that can be done, they are clever enough to spec their equipment correctly.

 

I do not like quality parts being called audiophile - as it has become some sort of derrogative term. But it is indicative tha audiophiles demand such quality and measuring pros do not.

 

If you check the capacitors measurement links - the FIRST thing that had to be done in order to be able to measure well was making of an oscillator with low enough distortion. By a man who spent great chunk of his life understanding and perfecting capacitors - and presumably had everything commercially available for testing. When you reach the point the instrument is less perfect than device under test, no other way than replacing the instrument. Which usually means building your own.

 

It happened before - late 70/80s best tuner manufacturers were coming up with tuners that had better specs than the best officially in the US available FM signal generator by HP. It was Matsusiita generator from Japan used to establish those figures - and Tandberg of Norway also had their own , at least on the par with Matsusiita - or else they could newer have built (and align) tuners of the quality they did. As well as was done by the Sage Audio in England in 80s - their power amp

modules were at least one order of magnitude better spec'd for distortion than the best HP signal generator at the time.  Again courtesy of custom made generator.

 

So much for the leading manufacturers. They do not per default push the envelope. Mostly only when forced by the competition.

post #2912 of 3206
Quote:
Originally Posted by analogsurviver View Post

 

But it is indicative tha audiophiles demand such quality and measuring pros do not.

 

Yes - It is indicative - but not of what you think it is indicative of. 


Edited by liamstrain - 7/12/14 at 1:32pm
post #2913 of 3206
Quote:
Originally Posted by analogsurviver View Post

So much for the leading manufacturers. They do not per default push the envelope. Mostly only when forced by the competition.

-And this is where a well-renowned Köln (Cologne) - based company enters the picture.

There's a lot of excellent, semi-affordable lab equipment out there nowadays - but the one manufacturer I've encountered which still builds to a standard, not a price point is Rohde u. Schwarz.
post #2914 of 3206
Quote:
Originally Posted by analogsurviver View Post
 

 

The differences in amps will be for the most part in the quiet ranges - those Sansui pre/amps are exactly as powerful in either version, but the better amp will have much better clarity and resolution at low level, specially on complex passages. Like symphony orchestra with chorus at p and below level.

 

 

 

Are you saying the difference between amplifier is at low level? The distortion of a "quality" amplifier is lower at low volume than a non-quality amp. How in your opinion does switching regulator affect this low level amplification performance? Would a "quality" power cable help in this case?

 

If two amplifiers are perfectlt faithful to the original sound, how would they sounded different? What is the definition of "better " in this case.

 

I often see the term "resolution" being used. I never did understand what it means. This is first time I saw it being used with low level amplification. Does resolution means ability to produce signal at low level? So when people say the system does not have enough resolution, it means the system cannot reproduce certain sound? I have a recording of Sting playing in a cabin. In the background, there is a fire place with the logs crackling. I never noticed it until someone told me abut it. Now I can hear it on my system. Does this mean I have a high "resolving" system?

post #2915 of 3206
Quote:
Originally Posted by OddE View Post


-And this is where a well-renowned Köln (Cologne) - based company enters the picture.

There's a lot of excellent, semi-affordable lab equipment out there nowadays - but the one manufacturer I've encountered which still builds to a standard, not a price point is Rohde u. Schwarz.

Yes, agreed. Rohde & Schwartz to my knowledge never did produce to a price point.

 

An anecdote from my retail days. In store comes a "freshly baked" retired male customer. Interested in some plastic fantastic music system, typical CD/Tuner/cassette box + platic speakers, say around 200-300 EUR max. 

And starts asking about selectivity of the FM tuner. A most unusual request in this price range, to say the least. 

After switching to capture ratio, pilot tone suppression etc it was clear to me that this customer "knows something" - and I dully went to check the specs for this box. Most of what the customer wanted to know was not specified - and what was, was simply far below his expectations.

 

Then I asked where he used to work. National Radio & Television - FM radio department. I knew immediately where he grew accustomed to such stellar performance/specs as he obviously took for granted - 

Rohde & Schwartz :beerchug:

 

But he obviously did not have to pay out of his own pocket ...

post #2916 of 3206
Quote:
Originally Posted by dvw View Post
 

Are you saying the difference between amplifier is at low level? The distortion of a "quality" amplifier is lower at low volume than a non-quality amp. How in your opinion does switching regulator affect this low level amplification performance? Would a "quality" power cable help in this case?

 

If two amplifiers are perfectlt faithful to the original sound, how would they sounded different? What is the definition of "better " in this case.

 

I often see the term "resolution" being used. I never did understand what it means. This is first time I saw it being used with low level amplification. Does resolution means ability to produce signal at low level? So when people say the system does not have enough resolution, it means the system cannot reproduce certain sound? I have a recording of Sting playing in a cabin. In the background, there is a fire place with the logs crackling. I never noticed it until someone told me abut it. Now I can hear it on my system. Does this mean I have a high "resolving" system?

Yes, although not entirely. But since the music is ALWAYS present at some low level and only ocasionally does get louder, this is the most critical range. If any audio device is poor at low level, it will never be acknowledged as being good. 

 

If you under switching regulator understand switching power supply, the usual abberations are - or were - intermodulation in the treble. Each generation is better, I have yet to hear one that is completely devoid of this problem. 

 

There is no such thing as perfection in real life. VERY few amplifiers survive the simplest of tests - one channel of the oscilloscope connected to the input, another to the output of the amplifier. Sensitivities of the oscilloscope adjusted so that both are the same in amplitude. Then invert one for 180 degrees and sum the two together ; VERY rarely a perfect line = no difference between the two - will be observed when playing music - not pure sine wave . If you only take this difference and amplify it so that you can listen to it with headphones or speakers, you will learn this difference, which in perfect amp would simply not exist, does not sound the same among different amplifiers.

 

Yes, the cited example of logs crackling in the fireplace is indication of certain level of resolution. Poor equipment would glaze over it. 

 

One could apply "resolution" to the musicians, too. I will never forget the Vienna Philharmonic Orchestra under Ivan Fischer rehearsing Bartok's Concerto for Orchestra; he quietly spoke "Leise" = quietly - and was with that softly spoken word in a big concert hall louder than all  the members of the orchestra playing entrance to a movement of said concert... - they are justly famous for the ability to control their playing to this level !

post #2917 of 3206
Quote:
Originally Posted by analogsurviver View Post
 
Quote:
Originally Posted by dvw View Post
 

Are you saying the difference between amplifier is at low level? The distortion of a "quality" amplifier is lower at low volume than a non-quality amp. How in your opinion does switching regulator affect this low level amplification performance? Would a "quality" power cable help in this case?

 

If two amplifiers are perfectlt faithful to the original sound, how would they sounded different? What is the definition of "better " in this case.

 

I often see the term "resolution" being used. I never did understand what it means. This is first time I saw it being used with low level amplification. Does resolution means ability to produce signal at low level? So when people say the system does not have enough resolution, it means the system cannot reproduce certain sound? I have a recording of Sting playing in a cabin. In the background, there is a fire place with the logs crackling. I never noticed it until someone told me abut it. Now I can hear it on my system. Does this mean I have a high "resolving" system?

Yes, although not entirely. But since the music is ALWAYS present at some low level and only ocasionally does get louder, this is the most critical range. If any audio device is poor at low level, it will never be acknowledged as being good. 

 

If you under switching regulator understand switching power supply, the usual abberations are - or were - intermodulation in the treble. Each generation is better, I have yet to hear one that is completely devoid of this problem. 

 

There is no such thing as perfection in real life. VERY few amplifiers survive the simplest of tests - one channel of the oscilloscope connected to the input, another to the output of the amplifier. Sensitivities of the oscilloscope adjusted so that both are the same in amplitude. Then invert one for 180 degrees and sum the two together ; VERY rarely a perfect line = no difference between the two - will be observed when playing music - not pure sine wave . If you only take this difference and amplify it so that you can listen to it with headphones or speakers, you will learn this difference, which in perfect amp would simply not exist, does not sound the same among different amplifiers.

 

Yes, the cited example of logs crackling in the fireplace is indication of certain level of resolution. Poor equipment would glaze over it. 

 

One could apply "resolution" to the musicians, too. I will never forget the Vienna Philharmonic Orchestra under Ivan Fischer rehearsing Bartok's Concerto for Orchestra; he quietly spoke "Leise" = quietly - and was with that softly spoken word in a big concert hall louder than all  the members of the orchestra playing entrance to a movement of said concert... - they are justly famous for the ability to control their playing to this level !


wow you really have a gift for finding "scientific look alike" to justify your strange views.

maybe that's just me, maybe I forgot how to science.

 

 an amp would actually distort more on the higher voltage signal by simple math. be it good or bad amp. 1% distortion from a signal at 0DB will always make much more noise than even 5% distortion from a sound at -60DB. 

the 1% distortion made from music at 0DB will be noise at -40DB and below, while the 5% distortion generated from a sound at -60DB will make a noise at -85DB and below(something like that right guys?).

also the resulting signal of you suggested experiment(sorry I don't own an oscilloscope) could be the slight difference in alignment. or maybe just a difference in dynamic making the alignment impossible without stretching the signal exactly? that could create up to a few DB signals on your result that wouldn't really be noise per se.

 what load do you place on the outputs of the amp? and even if I imagine this as being done in a correct way, then the difference is ... the distortion of the amp. \o/ 

and there might well be more because of IMD that obviously doesn't show up with a single sine wave, and because all the music will have ripples of harmonic distortions. but most of us know that already. what counts is the level in DB of the distortion, if it's low enough you don't hear it and that's it. even if the shape of the noise looks like mickey.

I don't get where that example is going at all. it's not like any amp will ever measure with 0 distortion. you don't give us any numerical value as if an error was an error too many and that's it. maybe the noise(anything that is not music) is at -90DB and you make it look like it's the end for music.

 

anyway I would really like to see the magnitude of whatever you hope to show in proportion to let's say ... the distortion from your beloved and oh so realistic vinyls, compared to the signal from the masters. that sure would put things in perspective. how come you're never that critical for all the defects of vinyl and DSD? this is sound science, I would expect all things criticized with the same rigor.

 

oh and I think dvw's post was ironic.

post #2918 of 3206
Quote:
Originally Posted by analogsurviver View Post
 

 

If you under switching regulator understand switching power supply, the usual abberations are - or were - intermodulation in the treble. Each generation is better, I have yet to hear one that is completely devoid of this problem. 

 

There is no such thing as perfection in real life. VERY few amplifiers survive the simplest of tests - one channel of the oscilloscope connected to the input, another to the output of the amplifier. Sensitivities of the oscilloscope adjusted so that both are the same in amplitude. Then invert one for 180 degrees and sum the two together ; VERY rarely a perfect line = no difference between the two - will be observed when playing music - not pure sine wave . If you only take this difference and amplify it so that you can listen to it with headphones or speakers, you will learn this difference, which in perfect amp would simply not exist, does not sound the same among different amplifiers.

 

I'm so glad all my equipment are high resolution. But I don't think I can agree with you on switching regulators cause intermodulation distortion on high frequency signal (treble). If that is the case we'll be seeing lots of distortion on high frequency equipment like DSL modem, WiFi router. As far as I know all communication equipment use switching reg, and intermodulation would cause BER.

 

On the next part of signal inverted 180 deg. Do this also apply to differential signal? Does this mean a single ended output is better than differential distortionwise? I mean when the signals are summed there will be more distortion than a simgle ended signal. You would definitely not recommend any bridged amp application, right?.

 

Since no amp is perfect which I do agree. And no two amp are the same even coming off the same production line (at least in theory). Do you think any one can tell the difference between two amps of the same model? If not, what is the deviation before you can hear the difference? 0.5 dB, 1dB? I visited a headphone manufacturer a ouple of months ago. They do use "golden ears" to do the final QA on their headphone. They can hear about 1dB difference. 1 dB iis a lot of variance.

post #2919 of 3206
Quote:
Originally Posted by dvw View Post
 

I'm so glad all my equipment are high resolution. But I don't think I can agree with you on switching regulators cause intermodulation distortion on high frequency signal (treble). If that is the case we'll be seeing lots of distortion on high frequency equipment like DSL modem, WiFi router. As far as I know all communication equipment use switching reg, and intermodulation would cause BER.

 

On the next part of signal inverted 180 deg. Do this also apply to differential signal? Does this mean a single ended output is better than differential distortionwise? I mean when the signals are summed there will be more distortion than a simgle ended signal. You would definitely not recommend any bridged amp application, right?.

 

Since no amp is perfect which I do agree. And no two amp are the same even coming off the same production line (at least in theory). Do you think any one can tell the difference between two amps of the same model? If not, what is the deviation before you can hear the difference? 0.5 dB, 1dB? I visited a headphone manufacturer a ouple of months ago. They do use "golden ears" to do the final QA on their headphone. They can hear about 1dB difference. 1 dB iis a lot of variance.

This makes sense, I have yet to run into a post that mentioned this, maybe because I spend a lot of time outside the science section.  The differential amp utilizes two outputs, which means there will be more amps that will be utilized fro two paths instead of single.  Since the differential is sum of both, distortion from each path would sum.  This depends on what is more significant.  Which is more more significant, the noise or the distortion caused by addition of the amps.  Anybody know more about this?

 

I think that it's pretty reasonable to think that headphone is where the signal accuracy matters as it has the most significant distortion that is caused.  The amps magnatudes less.  What I like to see is signal measurements of various distortion of signals fed to the headphones, and we can get a better idea of the relationship between distortion of the amp to the headphones.


Edited by SilverEars - 7/12/14 at 8:10pm
post #2920 of 3206
Quote:
Originally Posted by castleofargh View Post


interesting read when you're curious about the subject thx. but as a pure audiophile(sound sound sound!!!!!!) I don't get the point of looking after capacitors. of course for a product manufacturer, or a few DIYers it's mighty important, but for me the product user? if my product can give under 100 or 110DB in distortion, then obviously the capacitors are also below that value(or they did something to correct the signal afterward). in any case it's taken care of. so isn't that a false problem?
I know I could give the same argument for half our discussions here where people struggle between audibly perfect and audibly night and day perfecterer(what I now call the yellow M&M's), but if the end result is good(measurement good), why cares about the individual components? obviously the engineer did something right and the "problem" is solved for at least audio purpose.
A pure audiophile is interested in caps, plugs or whatever to get better sound.
The way to get great sound is by using great parts. Bose, for instance, can't engineer audiophile products because he uses cheap materials. Engineering is not how to get audiophile. Engineering with junk parts gets you Bose, at least. Engineering with quality parts gets you to what audiophiles desire. No audiophile would try to make a Bose speaker sound good by re-engineering it, adding foam, or anything because it's not good to start with. A few big companies put out products with great parts, but the overall product was not so good. All they needed was a someone to come along and use those parts right. But using parts right doesn't get too much better than Bose unless you actually care about every bit. The same goes with all products.
post #2921 of 3206
Quote:
Originally Posted by JamesHuntington View Post


A pure audiophile is interested in caps, plugs or whatever to get better sound.

 

The problem is, the pure audiophile does this assuming that these parts will get better sound. But ALL the testing, indicates this is pretty much not the case. That's where then engineer has the advantage, and can actually create things that do sound better - by knowing which things offer material improvements, and which are just higher cost with no audible improvement. 

post #2922 of 3206
Quote:
Originally Posted by JamesHuntington View Post


A pure audiophile is interested in caps, plugs or whatever to get better sound.
The way to get great sound is by using great parts. Bose, for instance, can't engineer audiophile products because he uses cheap materials. Engineering is not how to get audiophile. Engineering with junk parts gets you Bose, at least. Engineering with quality parts gets you to what audiophiles desire. No audiophile would try to make a Bose speaker sound good by re-engineering it, adding foam, or anything because it's not good to start with. A few big companies put out products with great parts, but the overall product was not so good. All they needed was a someone to come along and use those parts right. But using parts right doesn't get too much better than Bose unless you actually care about every bit. The same goes with all products.


An audiophile should want the best sound possible, not the best imaginary sound. Just because you imagine an electrolytic capacitor inside an amplifier will be a detriment to the sound doesn't make it true.

 

Cheers

post #2923 of 3206
Quote:
Originally Posted by ab initio View Post


An audiophile should want the best sound possible, not the best imaginary sound. Just because you imagine an electrolytic capacitor inside an amplifier will be a detriment to the sound doesn't make it true.

Cheers
just because you don't/can't hear it doesn't make it imaginary. Quality Capacitors are very important. The problem comes with the use of junk materials and the problem is real. The answer Many suggest to the problem is to avoid going backwards, so to say, by using cheap material or products that put profit first.
post #2924 of 3206
Quote:
Originally Posted by JamesHuntington View Post


just because you don't/can't hear it doesn't make it imaginary. Quality Capacitors are very important. The problem comes with the use of junk materials and the problem is real. The answer Many suggest to the problem is to avoid going backwards, so to say, by using cheap material or products that put profit first.

The beautiful thing about science though is that we can measure how much of a difference it makes, and study what kinds of differences are perceptible to humans. Evidence matters here, not just blind assertion of superiority.

post #2925 of 3206
Quote:
Originally Posted by cjl View Post
 
Quote:
Originally Posted by JamesHuntington View Post


just because you don't/can't hear it doesn't make it imaginary. Quality Capacitors are very important. The problem comes with the use of junk materials and the problem is real. The answer Many suggest to the problem is to avoid going backwards, so to say, by using cheap material or products that put profit first.

The beautiful thing about science though is that we can measure how much of a difference it makes, and study what kinds of differences are perceptible to humans. Evidence matters here, not just blind assertion of superiority.


but that's not even the problem, the signal gets ringing or overshoot or takes too much time to reach a voltage or to go back to zero, in probably several parts of a complex circuitry. a good deal of those circuits are in fact implemented just to counter those unwanted artifacts(filters, feedback ...). it's a very unrealistic idea to think that the signal is what it is and that each component with slight imperfection will add an irremediable defect to the signal. nothing could ever be done in electronic if it really worked like that. that's an oversimplified wrong idea of a circuit. having one component misbehaving isn't like going to mp3. a deviation doesn't automatically mean a signal loss.

 

say we need one special type of component to build a particular system. that component always overshoot the desired voltage on first impulse and even the best component does it a little. if we have a way to supress that overshooting without bringing something worse whatever the value, why should we pay for the expensive slightly better one? I think the MIT videos on feedback show that brilliantly. take something super bad, add stuff, end up with a close to perfect sine wave. job done done!

I would even bet that often times, changing a system for a component with better specs could actually have a negative effect on the output signal. because the design and component values were taking the bad component response into account and compensating for it adequately. so when the error is out, or smaller, the correction components might bring it back in the opposite direction.

 

anyway, as always, if the output is transparent, it's transparent.  I believe that tubes are useless relics of the past in audio, but when I see a tube amp with great specs, it's still a great amp. how a good result is obtained doesn't matter.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Testing audiophile claims and myths