Audible difference in high-end capacitors? - ABX samples

[fpitas]

The only thing curious was the absence of a meaningful test.

I think that's kind of SOP for those guys.

 

[fpitas]

And I hate to be cynical (OK, I really love it) but you won't move high-end rubbish without repeating various snake oil claims.

 

[pogo]

You can hear that?

I assume. For example, I can enjoy very quiet music with my M33. For my old class ABs, I needed a higher level to hear details.

 

[fpitas]

I assume. For example, I can enjoy very quiet music with my M33. For my old class ABs, I needed a higher level to hear details.

You should spring for modern amps. Seriously.

 

[GXAlan]

Back to the audiophiles asking science to prove a negative "test everything or you don't know"

The burden of proof is other way around. If someone believes that an expensive cap makes a difference, they only have to test that one cap to prove the difference really exists.

+1000. The burden of proof is on the capacitor makes a difference side.

They're very no-nonsense, so when the designer of their new flagship wanted hifi nonsense, they decided to use blind testing as their methodology for decision-making. Everyone liked the higher end caps. They still went with their standard Z-caps, but the Duelund caps won every time.

It would be nice to see the results. Often, but not always, when someone says “with all due respect…” they proceed to say something disrespectful. A sound designer saying that he didn’t want hi-fi nonsense often precedes hi-fi nonsense.

What may be fair is Classé Audio and this video about 1:30 in.

When talking about sound quality, you see their measurements and then they make a comment about leaving no stone unturned and some small change might not be audible in isolation, may be audible when the small change is repeated multiple times. I recall this being their decision to use Mundorf capacitors in a different video, but I cannot find that.

What I don’t know is the idea of tolerance. Cheap and expensive capacitors may perform identically but if you bought 100, and they are +/- 3%, would one batch of cheap capacitors always be -3% and a handful even falling out of spec while the expensive ones are just truly +/- 1% and sold as +/- 3%?

 

[uwotm8]

Sure, here's the difference file: https://app.box.com/s/xcgtrzqj1jgh2z8z76i48h0bue1ssnl5

Listen at your normal listening levels -- let us know what you hear

Most part of source record but 10x quieter, pretty much as expected.
Listening to source again, well, there might be a slightest of slightest difference in favor to Alu-Z but I will not catch it in ABX.
VERY uncomfortable record to listen tho, I'm saying that again.

 

[voodooless]

What I don’t know is the idea of tolerance. Cheap and expensive capacitors may perform identically but if you bought 100, and they are +/- 3%, would one batch of cheap capacitors always be -3% and a handful even falling out of spec while the expensive ones are just truly +/- 1% and sold as +/- 3%?

I’m actually quite surprised the 3 DUT caps are so close together. Also mind that the tolerance includes some temperature dependence.

 

[lashto]

I participated in a blind and sighted amp comparison. Way too many details to go into, but my final outcome, was I felt braindead, forcing myself to be able to tell if there was ANY difference in sound, instead of some huge sound signature change as some guys went in expecting to hear.

In the end, the outcome, including the guys expecting "Big Changes" , was pretty much non conclusive. No one could even tell the amps apart from each other, (For sure), let alone hearing some big huge change.

Yep, that everything-sounds-the-same 'virus' is spreading quite fast on ASR lately. I wonder why are all those people spending so much time/effort with their measurements. That does not seem particularly bright of them.
Same as bright for me to pay a monthly 'fee' here when I can just repeat the everything-mantra and all will be roses... also post it a few times every day and become a very popular 'audio-scientist'
/sarcasm

 

[lashto]

Back to the audiophiles asking science to prove a negative "test everything or you don't know"

The burden of proof is other way around. If someone believes that an expensive cap makes a difference, they only have to test that one cap to prove the difference really exists.

Yet, amazingly, no-one ever does it.

If a (more) complete set of measurements is done, this thread could be a very useful reference. One may quote it as solid proof everytime someone says "blabla cap sounds so much different/better". As it stands, without much else than a single FR measurement, it's not much of a proof.
It's just a very small +1 for the null hypothesis that all/most (properly sized) caps sound the same. Kind of useful too and many thanks for the effort. Just saying that it could be much more useful/better.

Anyway, looks like asking for more is not particularly welcome here. Apparently that brings up some sort of audiophile-paranoia. And other useful answers like "go DIY" or "just leave us alone, we're perfect and know it all". Sorry for 'crashing' whatever party was going on here and have fun!

 

[pkane]

As it stands, without much else than a single FR measurement, it's not much of a proof.

Maybe you missed null, phase, frequency, psychoacoustically-weighted audibility and impulse response measurements? These were all posted earlier today.

 

[SMAC88]

I run the test in my studio,

Sundara + Topping L50

There is a clear difference between the Alumen-Z Cap and the Standard one.

I A/B tested with volume match (there is a 2.8db difference between the 2 pink noise recordings and 2.4db the music samples).

The Alumen does sound slightly smoother in the pink noise test, as well as in the music samples, at 0:35 sec the tambourine sounds smoother and clear compared to the standard one which sound like dirty and with harmonics.

This difference goes completely unnoticed in a room that is not treated or only partially treated, especially with background noise, reverb, and AAC codecs. I can conclude that the difference is so minimal that it could be attributed to sample variation. Even two Alumen samples may not sound exactly the same.

So the real question is, does it really make sense to pay X 44 times more for something that you'll only notice in a direct A/B test? I don't think so. I also believe that the small audible difference is represented in your tests, they don't align 100% in all the spectrum so, yes the differences are measured, it's just not worth considering them.

 

[Doodski]

for me to pay a monthly 'fee' here when I can just repeat the everything-mantra and all will be roses... also post it a few times every day and become a very popular 'audio-scientist'
/sarcasm

Nobody is expected to pay a monthly fee to contribute here. I give my contribution in about February of each year and I consider it money well spent for many reasons. It gives us a place to gather on common interest and we all get to chat about whatever subjects we want to join in on. It's a great place!

 

[fpitas]

If a (more) complete set of measurements is done, this thread could be a very useful reference. One may quote it as solid proof everytime someone says "blabla cap sounds so much different/better". As it stands, without much else than a single FR measurement, it's not much of a proof.
It's just a very small +1 for the null hypothesis that all/most (properly sized) caps sound the same. Kind of useful too and many thanks for the effort. Just saying that it could be much more useful/better.

Anyway, looks like asking for more is not particularly welcome here. Apparently that brings up some sort of audiophile-paranoia. And other useful answers like "go DIY" or "just leave us alone, we're perfect and know it all". Sorry for 'crashing' whatever party was going on here and have fun!

You pay to hang around idiots like us?

 

[Keith_W]

This discussion about the microphonics of capacitors is quite interesting. Some capacitors are mounted inside speaker cabinets as part of the crossover. If it is true that capacitors are affected by microphonics, it may be plausible that a capacitor with more robust mechanical construction would resist changing its value. Perhaps we should do a test to see how the capacitance (and maybe frequency response?) varies when subject to a loud test signal. Also, it would be interesting to see if heat makes a difference.

 

[beagleman]

Back to the audiophiles asking science to prove a negative "test everything or you don't know"

The burden of proof is other way around. If someone believes that an expensive cap makes a difference, they only have to test that one cap to prove the difference really exists.

Yet, amazingly, no-one ever does it.

First off, one needs a "Highly resolving" system, and the "Ears" to hear it.




(The typical audiophile response)

 

[Doodski]

This discussion about the microphonics of capacitors is quite interesting. Some capacitors are mounted inside speaker cabinets as part of the crossover. If it is true that capacitors are affected by microphonics, it may be plausible that a capacitor with more robust mechanical construction would resist changing its value. Perhaps we should do a test to see how the capacitance (and maybe frequency response?) varies when subject to a loud test signal. Also, it would be interesting to see if heat makes a difference.

Most caps are not very microphonic although I have seen some failing caps that when subjected to me tapping on them with something they show their fault and it is microphonic or a complete failure. That test suite that you suggested is a excellent test though.

 

[fpitas]

First off, one needs a "Highly resolving" system, and the "Ears" to hear it.




(The typical audiophile response)

Not too resolving. Then it will sound dry and analytical.

 

[levimax]

I think it would be even more interesting to take in another step. I would like to see a test for some higher value / higher voltage crossover caps comparing Film and Fancy Film caps to lowly Bipolar electrolytic caps.

 

[pablolie]

The test is interesting - but we should also consider other factors like long-term performance and reliability when assessing the suitability of components for long term use in audio applications. We all like high signal quality, but let's admit we also don't enjoy sending our equipment out for repair because someone took shortcuts and/or tried to increase profitability by packaging in unnecessarily cheapo components.

 

[GXAlan]

Yep, that everything-sounds-the-same 'virus' is spreading quite fast on ASR lately. I wonder why are all those people spending so much time/effort with their measurements.

Not sure where your sarcasm starts and stops. At least for me, I don’t think everything sounds the same. Spending time and effort with measurements helps try to sort out sighted and blind testing. Half of the fun is measuring.

My biggest argument is that when someone says “night and day” difference, it may only be during very specific portions of a specific song. So the pink noise or things like pkmetric do have some limitations. With the 300B SET review, I have showed how volume level matters also.

The biggest difference I found ended up being that a Topping PA-5 sounded different. But then it turned out the channels were reversed. I no longer had the setup when this was discovered. Then, someone showed the PA-5 II had improperly wired polarity, so I am officially going to say that mine was incorrectly wired.

This thread shows how a Wima “10%” capacitor was actually way better than stated while an Epcos/TDK 5% met the spec but had greater variation (not a direct comparison)

New capacitors out of tolerance... what would you do ?

Hello, I've recently buy some 5% MKP capacitors (very expensive : 0.46$) The serie is pretty bad and some of them are out of tolerance. Is it a common problem ?

www.diyaudio.com