"I don't need blind A/B tests ore any fancy measurements to tell me"
I thought we already came to an agreement on this depedant on your objective?
"There are many audio components that measure terrible yet sound great" - "different interconnects sound different, as do different tubes, as do speaker wires, as do output tranformers, as do coils, caps, etc"
Yes, you would be referring to euphonic design. Your examples of tube amps and speakers is hardly relevant, as the high output impedance of most tube amps is not able to work accurately in conjunction with speakers having broad impedance swings, resulting in FR changes that are within the threshold of audibility. The same is true with speakers, where again no one is giong to argue the differences of different speaker drivers having an audible difference. These relatively HUGE differences are not a debate. These characteristics are used to an advanatage to 'voice' these devices in most cases.
When you discuss coils, resistors, wires, caps, etc. the objective is a device that adds or subtracts nothing, excepting an intented function such as resistance, capacitance, etc.. Any correctly applied variants of these devices in an application perform so closely that bias is the ruling factor in evaluation. Thus, controlled testing must be used to find if their are actual differences between like-value components. Not using controlled testing is perfectly O.K. IF YOUR OBJECTIVE is to find what you THINK sounds best. In this circumstance you must be willing to accept purely psychological effects as a feasible method to change your perception. If you agree with this, then what are we arguing about?
-Chris
I thought we already came to an agreement on this depedant on your objective?
"There are many audio components that measure terrible yet sound great" - "different interconnects sound different, as do different tubes, as do speaker wires, as do output tranformers, as do coils, caps, etc"
Yes, you would be referring to euphonic design. Your examples of tube amps and speakers is hardly relevant, as the high output impedance of most tube amps is not able to work accurately in conjunction with speakers having broad impedance swings, resulting in FR changes that are within the threshold of audibility. The same is true with speakers, where again no one is giong to argue the differences of different speaker drivers having an audible difference. These relatively HUGE differences are not a debate. These characteristics are used to an advanatage to 'voice' these devices in most cases.
When you discuss coils, resistors, wires, caps, etc. the objective is a device that adds or subtracts nothing, excepting an intented function such as resistance, capacitance, etc.. Any correctly applied variants of these devices in an application perform so closely that bias is the ruling factor in evaluation. Thus, controlled testing must be used to find if their are actual differences between like-value components. Not using controlled testing is perfectly O.K. IF YOUR OBJECTIVE is to find what you THINK sounds best. In this circumstance you must be willing to accept purely psychological effects as a feasible method to change your perception. If you agree with this, then what are we arguing about?
-Chris
SOUND and Measurements
Brad,
I agree with many of the thing you have said, and your right we have biases and preferences. However a company or an engineer must have a way of qualify the components he is using.
OK THD this tells us about the quality of a sinewave. This only give us non-descriptive information about non-linearity's of a signal, but it a good reference point or place to start. However we don't listen to sinewaves , do we. A impulse wave tells us much more about an amp's rise and fall time, bandwidth and more. Of course, noise measurements are important but not the whole picture. So we need many different tests to qualify a part, component or a systems. In addition, each sub-system will have it own types of measurements ie. Power supplies, amps, DAC, transistors, caps, filters, crystals, laser, fiber on and on it goes.
Let talk about resistor, I'm sure your aware that you really don't want to use carbon composition resistors in the signal path for audio. The fact is carbon resistors sound bad and they generate noise. Metal film resistors sound better, they a have lower noise and better stability. However, at high frequency circuits with carbon comp function better than metal films. The reason is the metal film resistor has a spiral cut raising the inductance, decreasing their bandwidth. Ok I've added a link for you to read that written by Dr. Marsh and Walt Jung on testing.
http://www.capacitors.com/pickcap/pickcap.htm
I can go on making these point, however that fact is that components companies and large one make a living out of high performance resistors, caps, transistor, lasers PCB's diode and even auto parts. Most of these companies will even tell you how to measure the differences in components.
So yes we can make measurement that can characterize the sonic performance of parts and systems CHRIS. It all in knowing what you need to measure.
The END
Brad,
I agree with many of the thing you have said, and your right we have biases and preferences. However a company or an engineer must have a way of qualify the components he is using.
OK THD this tells us about the quality of a sinewave. This only give us non-descriptive information about non-linearity's of a signal, but it a good reference point or place to start. However we don't listen to sinewaves , do we. A impulse wave tells us much more about an amp's rise and fall time, bandwidth and more. Of course, noise measurements are important but not the whole picture. So we need many different tests to qualify a part, component or a systems. In addition, each sub-system will have it own types of measurements ie. Power supplies, amps, DAC, transistors, caps, filters, crystals, laser, fiber on and on it goes.
Let talk about resistor, I'm sure your aware that you really don't want to use carbon composition resistors in the signal path for audio. The fact is carbon resistors sound bad and they generate noise. Metal film resistors sound better, they a have lower noise and better stability. However, at high frequency circuits with carbon comp function better than metal films. The reason is the metal film resistor has a spiral cut raising the inductance, decreasing their bandwidth. Ok I've added a link for you to read that written by Dr. Marsh and Walt Jung on testing.
http://www.capacitors.com/pickcap/pickcap.htm
I can go on making these point, however that fact is that components companies and large one make a living out of high performance resistors, caps, transistor, lasers PCB's diode and even auto parts. Most of these companies will even tell you how to measure the differences in components.
So yes we can make measurement that can characterize the sonic performance of parts and systems CHRIS. It all in knowing what you need to measure.
The END
"http://www.capacitors.com/pickcap/pickcap.htm"
Hardly unbiased listening tests. However, if you can overlook the subjective listening tests it is a very good source of information I agree. I was wondering when someone was going to hyperlink this often cited page. Regardless of the sugjective content, the published data they present shows insignifiant differences between their sampled metallized polypropylene and polyprolyene type capacitors. This link hardly helps your argument. Even with the polyester capacitors that certainly have inferior specifications, you must still determine the threshold that is audible. Again, how shall we determine actual differences at this level? Hmmm.
"however that fact is that components companies and large one make a living out of high performance resistors, caps, transistor, lasers PCB's diode and even auto parts"
LMAO! Yes, once again the old 'if it costs more and people buy it, it MUST be better' routine. In this case, 'buy it' has two very applicable meanings. I don't understand how you come to these conclusions. Must be a result of completely subjective evaluation. It is not an issue if someone uses what they like. I don't have a quibble with that. But because they paid more for it and like it, it MUST be better?????? Would I be correct in assuming you also percieve differences in cables/interconnects? I noticed that your specific statments earlier concerning polyester, metallized polypropylene and polypropylene capactiros coincided directly with pricing, not the relative 'measured' performance differences of the different types. Funny isn't it? Bias that is.
-Chris
Hardly unbiased listening tests. However, if you can overlook the subjective listening tests it is a very good source of information I agree. I was wondering when someone was going to hyperlink this often cited page. Regardless of the sugjective content, the published data they present shows insignifiant differences between their sampled metallized polypropylene and polyprolyene type capacitors. This link hardly helps your argument. Even with the polyester capacitors that certainly have inferior specifications, you must still determine the threshold that is audible. Again, how shall we determine actual differences at this level? Hmmm.
"however that fact is that components companies and large one make a living out of high performance resistors, caps, transistor, lasers PCB's diode and even auto parts"
LMAO! Yes, once again the old 'if it costs more and people buy it, it MUST be better' routine. In this case, 'buy it' has two very applicable meanings. I don't understand how you come to these conclusions. Must be a result of completely subjective evaluation. It is not an issue if someone uses what they like. I don't have a quibble with that. But because they paid more for it and like it, it MUST be better?????? Would I be correct in assuming you also percieve differences in cables/interconnects? I noticed that your specific statments earlier concerning polyester, metallized polypropylene and polypropylene capactiros coincided directly with pricing, not the relative 'measured' performance differences of the different types. Funny isn't it? Bias that is.
-Chris
It is the norm on cost
Christ,
On several occasions, I spent a lot of time characterizing and testing capacitors for high accurate high speed sample & hold I designed at Texas Instruments. I found two capacitor that I could use for the design one being a Teflon and the alternate being a precision Polystyrene. The Key here to the component is the DA, the ESR of and temp performance of the caps. Of course the Teflon part cost more and had the best performance. You see the designed specification called for the highest accuracy. the cost of this one cap was not important since it effected the over all system performance.
So you seem to have an issue for paying quality and precision, whats that about. What components have you found that justify you statements.
My experience comes from design and testing high performance systems. I get samples first if they meet my expectations and the price point is right I use it. When you start a new design for production, you don't cut corners. After you get the performance you expect, then you cost reduce the design to meet your target performance and price, of course digital assemblies are a little different. This way you know what your margins are.
For my home audio projects I don't work that way, I use what is know is best, however I use those part were needed. Yes I have a bias and it's not having to go back and modify or change something I've already completed. I have done before and don't like wasting the money or the time.
Good luck
Christ,
On several occasions, I spent a lot of time characterizing and testing capacitors for high accurate high speed sample & hold I designed at Texas Instruments. I found two capacitor that I could use for the design one being a Teflon and the alternate being a precision Polystyrene. The Key here to the component is the DA, the ESR of and temp performance of the caps. Of course the Teflon part cost more and had the best performance. You see the designed specification called for the highest accuracy. the cost of this one cap was not important since it effected the over all system performance.
So you seem to have an issue for paying quality and precision, whats that about. What components have you found that justify you statements.
My experience comes from design and testing high performance systems. I get samples first if they meet my expectations and the price point is right I use it. When you start a new design for production, you don't cut corners. After you get the performance you expect, then you cost reduce the design to meet your target performance and price, of course digital assemblies are a little different. This way you know what your margins are.
For my home audio projects I don't work that way, I use what is know is best, however I use those part were needed. Yes I have a bias and it's not having to go back and modify or change something I've already completed. I have done before and don't like wasting the money or the time.
Good luck
"I spent a lot of time characterizing and testing capacitors for high accurate high speed sample & hold I designed at Texas Instruments. I found two capacitor that I could use for the design one being a Teflon and the alternate being a precision Polystyrene."
Look at your statement. WHAT does this have to do with the relatively low frequencies in an analogue circuit covering the audio band? Are the components you cited of superior measurement? No question they are, when did I ever say they were not? Any audible difference in an analogue circuit as compared to using metallized poly, poly tin foil, comparables? You may speculate, and draw biased conclusion on subjective tests all that you wish in that regard. That is your right. But don't preach it as if it's not refutable here. A board like Madisound, you would likely not find anyone to disagree with you, as their are lots of 'believers' of subjective component selection over their.
"So you seem to have an issue for paying quality and precision"
Again, with you it's always about the money. More expensive MUST be better, huh? You miss the point again.
"What components have you found that justify you statements. "
JUSTIFY my statements? AGAIN, you are looking at this entire thing in the wrong perspective. What is there for me to justify? Am I to justify the need for controlled testing to insure accurate testing of like components? You believe you are not subject to judgement errors from psychological biases that everyone else is? Your statements would certainly indicate the you are a true believer. That I will NEVER argue. LMAO!
This thread has past it's usefullness. But I will be glad to keep responding to you as long as the board moderator leaves the thread open.
-Chris
Look at your statement. WHAT does this have to do with the relatively low frequencies in an analogue circuit covering the audio band? Are the components you cited of superior measurement? No question they are, when did I ever say they were not? Any audible difference in an analogue circuit as compared to using metallized poly, poly tin foil, comparables? You may speculate, and draw biased conclusion on subjective tests all that you wish in that regard. That is your right. But don't preach it as if it's not refutable here. A board like Madisound, you would likely not find anyone to disagree with you, as their are lots of 'believers' of subjective component selection over their.
"So you seem to have an issue for paying quality and precision"
Again, with you it's always about the money. More expensive MUST be better, huh? You miss the point again.
"What components have you found that justify you statements. "
JUSTIFY my statements? AGAIN, you are looking at this entire thing in the wrong perspective. What is there for me to justify? Am I to justify the need for controlled testing to insure accurate testing of like components? You believe you are not subject to judgement errors from psychological biases that everyone else is? Your statements would certainly indicate the you are a true believer. That I will NEVER argue. LMAO!
This thread has past it's usefullness. But I will be glad to keep responding to you as long as the board moderator leaves the thread open.
-Chris
Chris,
The biggest problem I have with what you're saying is that you seem to say that without blind testing you can't determine if something sounds better or not. Now I won't argue that there are lots of people out there effected by biases and imagine changes in sound that aren't there. But I don't argree that EVERYONE is effected by this and cannot determine differences through extended listening. To the contrary, I think some people can detect changes in sound that are difficult or impossible to measure.
The biggest problem I have with what you're saying is that you seem to say that without blind testing you can't determine if something sounds better or not. Now I won't argue that there are lots of people out there effected by biases and imagine changes in sound that aren't there. But I don't argree that EVERYONE is effected by this and cannot determine differences through extended listening. To the contrary, I think some people can detect changes in sound that are difficult or impossible to measure.
"But I don't argree that EVERYONE is effected by this and cannot determine differences through extended listening. "
Sorry, bias is a universal problem for all humans. However, I agree with you in one regard: their may very well be people with freakish 'super' hearing that could possibly hear a difference that is below normal hearing abilities. However, for the other 99.99% of the population it would not be a factor. Again, IF they were able to hear that well they would still be subject to bias also, so even their 'sighted' listening is of no value, worthless. But, if their hearing is this good, then they will be able to readily identify the differences in a blind test won't they?
-Chris
Sorry, bias is a universal problem for all humans. However, I agree with you in one regard: their may very well be people with freakish 'super' hearing that could possibly hear a difference that is below normal hearing abilities. However, for the other 99.99% of the population it would not be a factor. Again, IF they were able to hear that well they would still be subject to bias also, so even their 'sighted' listening is of no value, worthless. But, if their hearing is this good, then they will be able to readily identify the differences in a blind test won't they?
-Chris
There are a two topic here. First is considering measurements, second is about sonic differences between (film) caps.
I wouldn’t say measurements could tell us everything we want to know. Many times everything is OK with measurements, but there are problems with sound. There could be many reasons for this. Maybe we don’t have appropriate interpretations of measurement. Maybe because all known measurement techniques are still inadequate. But perfect measurement results are not a guarantee for good (especially not for perfect) sound. Why tubes sound better than solid-state in many parameters? Because they have better specs? OK, I know some can say “they not sound better in any parameter” or some can debate importance of even and uneven harmonics, but I’ll be free to ignore such approaches now. This clumsiness between objective (measured) and subjective (heard) parameters applies to caps as for any other component. Unfortunately, but it is so. In this regard I don’t think the measurement of caps will tell us what to use. Blind (short) test or long term listening? If goal is to compare components, then blind test is very useful. But for reveal of real (unique) nature of component’s sound I think long term listening will tell us a lot more.
Pedja
I wouldn’t say measurements could tell us everything we want to know. Many times everything is OK with measurements, but there are problems with sound. There could be many reasons for this. Maybe we don’t have appropriate interpretations of measurement. Maybe because all known measurement techniques are still inadequate. But perfect measurement results are not a guarantee for good (especially not for perfect) sound. Why tubes sound better than solid-state in many parameters? Because they have better specs? OK, I know some can say “they not sound better in any parameter” or some can debate importance of even and uneven harmonics, but I’ll be free to ignore such approaches now. This clumsiness between objective (measured) and subjective (heard) parameters applies to caps as for any other component. Unfortunately, but it is so. In this regard I don’t think the measurement of caps will tell us what to use. Blind (short) test or long term listening? If goal is to compare components, then blind test is very useful. But for reveal of real (unique) nature of component’s sound I think long term listening will tell us a lot more.
Pedja
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.