We all think highly of our abilities but the more controlled the test, the poorer those abilities are proven to be. If you have any knowledge of what you're listening to, i.e., you can see it or control the comparisons yourself, the results are suspect at best.
Sure, we can hear frequency response differences and a variety of other things, but identical signals sound identical. I don't waste time even debating that with people.
If a differential comparison on the test bench shows two different volume controls and balance arrangements yield the same electrical signal to a precise degree, there isn't a difference, no matter what our flawed ears and brain tell us. Those are cases where the listening test has to be examined in detail because it's very likely flawed. OTOH, if there is an electrical difference, it can be described and quantified and then we have something to work with to improve the equipment and make good choices.
The problem with relying on listening comparisons alone is one never makes true progress, instead wandering from change to change forever. The debates go on forever as well because almost no one has the time and knowledge to set up truly unbiased comparisons. Or, maybe it's just more fun to remain in denial of the bias sources or to believe that if one tries really hard they can be overcome by force of will alone.
IMO, good design requires both controlled listening and precision measurements. Either one alone won't get you where you want to go.
Sure, we can hear frequency response differences and a variety of other things, but identical signals sound identical. I don't waste time even debating that with people.
If a differential comparison on the test bench shows two different volume controls and balance arrangements yield the same electrical signal to a precise degree, there isn't a difference, no matter what our flawed ears and brain tell us. Those are cases where the listening test has to be examined in detail because it's very likely flawed. OTOH, if there is an electrical difference, it can be described and quantified and then we have something to work with to improve the equipment and make good choices.
The problem with relying on listening comparisons alone is one never makes true progress, instead wandering from change to change forever. The debates go on forever as well because almost no one has the time and knowledge to set up truly unbiased comparisons. Or, maybe it's just more fun to remain in denial of the bias sources or to believe that if one tries really hard they can be overcome by force of will alone.
IMO, good design requires both controlled listening and precision measurements. Either one alone won't get you where you want to go.
Hi Conrad, based on your respected seasoned view on subjective evaluation, you mean that what" the test bench shows "and the "precision measurement" can
discriminate or ascertain via measurement parameters " an electrical difference " which is needed along with controlled test in order to conclude an audible difference in componentry? Is that what you do to discern your preferences in the audio gear that you listen to?
discriminate or ascertain via measurement parameters " an electrical difference " which is needed along with controlled test in order to conclude an audible difference in componentry? Is that what you do to discern your preferences in the audio gear that you listen to?
Whoever said that crosstalk is the only issue? If the circuit has poor PSU IM then separating into dual mono could improve sound even with zero crosstalk. Of course, the real solution is to solve the PSU IM - then you don't need dual mono and you get better sound in each channel as it no longer suffers from its own IM.
In the end there is always an engineering solution to an engineering problem. If it isn't an engineering problem then it probably is all in your head.
In the end there is always an engineering solution to an engineering problem. If it isn't an engineering problem then it probably is all in your head.
Just for interest,
ESP - A Better Volume Control
Material type is not the question because its back to does carbon sound better than conductive plastic...or what type of resistor is used in a stepped attenuator. Tantalum, carbon, metal film, etc Switch type...I'm selective, however thats just choice.
Regards
M. Gregg
ESP - A Better Volume Control
Material type is not the question because its back to does carbon sound better than conductive plastic...or what type of resistor is used in a stepped attenuator. Tantalum, carbon, metal film, etc Switch type...I'm selective, however thats just choice.
Regards
M. Gregg
Last edited:
I fully agree with you. One should be very skeptical indeed if one thinks one is hearing differences that cannot be correlated with precision measurements.
Chris
I'm not entirely sure of the question but too many times I've suspected a sonic difference that turned out to be either a figment of my imagination or some other issue, not even the thing I thought made the difference. Thus, I'm always skeptical of sonic differences until I track down the specific cause or causes on the test bench.
Listening sessions are great but I don't have the resources to do them blind, and I'm even more susceptible to suggestion and my own opinion if I know what the circuit or construction are. There is a danger- it's still possible to have test bench and listening sessions reinforce an incorrect conclusion if one is stubborn and not open to what the measurements are really saying.
It's worse with my DIY stuff because, depending on the day, I'm insecure enough to doubt everything I do, knowing its complete crap, or sufficiently convinced of my genius to know my creations are superior to all else.
Thanks to you all for your input: Apparently respondents to this thread believe the sonic virtues of audio equipment can be correlated to measured parameters or specification. Wow what a breakthrough! this mean that measurements on the test bench can determine which equipment will be sonically superior. this is essentially what can be deduced from your responses. Doesn't semiconductors designs have superior measured parameters than tubes yet many prefer the sound of tubes? What measurement accounts for this discrepancy.
I will assume sarcasm is being used as a substitute for rational argument based on understanding and reasonable evidence.
The 'sound of tubes' is not a well-defined concept. To some it means a poorly designed circuit which adds euphonic distortion yet maintains plausible deniability (at least in their own eyes). For others the 'sound of tubes' is aurally indistinguishable from competent SS systems; their preference for tubes is based on other issues.
I certainly don't want 'tube sound' in my system; I want good sound from tubes.
The 'sound of tubes' is not a well-defined concept. To some it means a poorly designed circuit which adds euphonic distortion yet maintains plausible deniability (at least in their own eyes). For others the 'sound of tubes' is aurally indistinguishable from competent SS systems; their preference for tubes is based on other issues.
I certainly don't want 'tube sound' in my system; I want good sound from tubes.
thank you DF96 for your response. yes you are correct. My sarcasm stems from the contention that one can correlate test bench measurement to determine sonic differences. please elaborate regarding your reasonable evidence. If you are incorporating tubes in your design you are hearing tubes amplification, period.
Negative feedback largely obliterates the characteristics of the active devices, which is precisely what it is intended to do.
By 'reasonable evidence' I mean either published and accepted data, or personal tests which make some effort to ensure that the apparent differences were actually heard. It is largely the former which tells us roughly what levels of distortion (~1%) and bandwidth (~ 20-20kHz) are required for good sound reproduction. Some people think they need much more than this, when in reality they may not even be approaching this (mainly due to transducer issues).
By 'reasonable evidence' I mean either published and accepted data, or personal tests which make some effort to ensure that the apparent differences were actually heard. It is largely the former which tells us roughly what levels of distortion (~1%) and bandwidth (~ 20-20kHz) are required for good sound reproduction. Some people think they need much more than this, when in reality they may not even be approaching this (mainly due to transducer issues).
Sonic differences are ALWAYS measurable. If not, then the difference heard is because of a poorly done listening test or a matter of bias, prejudice and other psychological effects.
Last edited:
Yes sure funk1980
I have an article called Commentary for Consumers. In this article it states and I quote " As a consumer, you want to get the best sound you can get. You can accomplish this through critical listening " It also states what I also stated in post#24 " Mono operation is very desirable in high end systems, but of course it is expensive. A modest compromise is offered by dual-mono operation " He must be as naïve as me. the person I quoted is NELSON PASS you know THE ONE AND ONLY!
Last edited:
It seems to me that two very different questions often get conflated in debates of this kind.
The first is one which, to me, should be essentially uncontroversial, namely that if two amplifiers genuinely sound different, then there will be differences in their output signals that can be measured by precision equipment. Ultimately, an amplifier is just an electronic device that is producing a time-dependent output voltage Vout(t) in response to an input voltage Vin(t). If two amplifiers produce a different sound (all other items, like signal source and loudspeakers, being equal), then it must be possible to see, using precision measurements, that Vout(t) from one amplifier is different from Vout(t) from the other.
The second, and much more subtle question, is whether one can, so to speak, "predict" what the sonic qualities of a given amplifier will be simply by poring over the data acquired by making precision measurements. I don't think anyone in this thread was making such a statement. But I think maybe when people react against "objectivists" they tend to assume that this much more contentious and subtle kind of claim has been made.
Personally, I am agnotic or somewhat skeptical about the second question. But to me, there is really nothing controversial or contentious about the first. And that is why I would say that if sonic differences are claimed, and yet they cannot be backed up by precision measurements, then there is a strong likelyhood that the claimed difference is imagined. At the very least, one should then be very careful to make sure that the claimed differences can still be heard in properly conducted double blind testing.
Chris
The first is one which, to me, should be essentially uncontroversial, namely that if two amplifiers genuinely sound different, then there will be differences in their output signals that can be measured by precision equipment. Ultimately, an amplifier is just an electronic device that is producing a time-dependent output voltage Vout(t) in response to an input voltage Vin(t). If two amplifiers produce a different sound (all other items, like signal source and loudspeakers, being equal), then it must be possible to see, using precision measurements, that Vout(t) from one amplifier is different from Vout(t) from the other.
The second, and much more subtle question, is whether one can, so to speak, "predict" what the sonic qualities of a given amplifier will be simply by poring over the data acquired by making precision measurements. I don't think anyone in this thread was making such a statement. But I think maybe when people react against "objectivists" they tend to assume that this much more contentious and subtle kind of claim has been made.
Personally, I am agnotic or somewhat skeptical about the second question. But to me, there is really nothing controversial or contentious about the first. And that is why I would say that if sonic differences are claimed, and yet they cannot be backed up by precision measurements, then there is a strong likelyhood that the claimed difference is imagined. At the very least, one should then be very careful to make sure that the claimed differences can still be heard in properly conducted double blind testing.
Chris
In the absence of rational argument, and when sarcasm has been rebuffed, why not try quoting a 'guru'? For any audio statement which it is desired to make, it is always possible to find someone who has said it before. It is a strange feature of human nature (seen, for example, in emergency evacuations) that people are happy to be wrong provided they are not alone in being wrong.yero said:
Last edited:
I can't predict what some error sounds like from measurements. What I can predict is that when the commonly known and understood errors are reduced to a very low level, electronics all sound the same, even to the most golden of ears. Again, identical signals sound identical. The "commonly known and understood errors" covers a lot of ground, not just response and THD into resistive loads, so you have to do your homework. I'm not saying that reducing all errors is a trivial task- most amps have flaws, even if minor.
Thus, as I've probably said on several occasions, any amp (or other electronic component) that sounds "different", even if it sounds better, probably has something wrong with it, contains a design error, or was purposely built to produce a less accurate sound.
BTW, if anyone needs support for their viewpoint, for a modest fee I'll say pretty much anything you want.
Thus, as I've probably said on several occasions, any amp (or other electronic component) that sounds "different", even if it sounds better, probably has something wrong with it, contains a design error, or was purposely built to produce a less accurate sound.
BTW, if anyone needs support for their viewpoint, for a modest fee I'll say pretty much anything you want.
Last edited:
Instead of using a standard balance arrangement with the wiper to ground, simply use a dual gang linear pot which each as a separate volume control contra-wired. You keep the practicality and remove the inter channel link, but to suggest two volume controls is either a practical solution or one that gives a valid improvement is.... well not wanting to be rude here but "bonkers" springs to mind...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.