Yes, I admit I pay little or no attention to things like MIT or Cardas wire or Geoff Kait's brilliant pebbles, myrtle blocks, the list goes on and on. Embracing all of them uncritically does not flatter your abilities. You don't need them, you are better than that.
I am only a critic of inaccurate or just plain mis-information or out and out fraud. I even admit I pay so little attention to the outsider audio industry that Dick's involvement with Rel Cap escaped me, which is probably a good thing.
without signal theory, physical device equations, models, numbers, correlated with controlled listening results, any psychoacousitic data I really can't see the cause for your smugness
"I often say that when you can measure what you are speaking
about, and express it in numbers, you know something about
it; but when you cannot express it in numbers, your knowledge
is of a meagre and unsatisfactory kind ...
--- Lord Kelvin
"I often say that when you can measure what you are speaking
about, and express it in numbers, you know something about
it; but when you cannot express it in numbers, your knowledge
is of a meagre and unsatisfactory kind ...
--- Lord Kelvin
Last edited:
That is completely beside the point -- (that line up 1db is an artifact of the crude software graphing being used on this old MLSSA). The point should be clear that I am talking about the sum total energy at the crossover area. And, as I said the description of the sound in that area where the decay is slow is similar to masking and distortion .....it would sound peaked anyway. If you add up each curve, the total power curve is VERY peaked.
We talked this topic to death in Lounge.
There is a dedicated topic, about sound quality and measurements. Your point is, group delays is what matters. My point is, much more. A first, nobody proved that we recognize sounds by harmonic content, so we may assume that waveform matters. It includes your postulate as well, about group delay. But a waveform is more than a group delay.
You are banging in already wide opened gates. Come in!
Low distortion cap sounds compressed on music --
Another view of how things sound and a test measurement that correlates: The peak current thru two caps of same voltage applied and same value. One film and foil and one the popular bipolar cap --- listening says the bipolar makes the sound compressed in comparison. This is yet another view of cap behavior
- especially useful info if making speaker crossovers. -RNM
Another view of how things sound and a test measurement that correlates: The peak current thru two caps of same voltage applied and same value. One film and foil and one the popular bipolar cap --- listening says the bipolar makes the sound compressed in comparison. This is yet another view of cap behavior
- especially useful info if making speaker crossovers. -RNM
Thank you. It is NOT my point that group delay matters. My point is simply that when listeners say they hear something and describe it, consistantly .... there are measurements that seem to correlate to that very audible description.We talked this topic to death in Lounge.
There is a dedicated topic, about sound quality and measurements. Your point is, group delays is what matters. My point is, much more. A first, nobody proved that we recognize sounds by harmonic content, so we may assume that waveform matters. It includes your postulate as well, about group delay. But a waveform is more than a group delay.
You are banging in already wide opened gates. Come in!
Last edited:
Time smear on a 1us time scale, you have to do better than that. Please submit these results to an unbiased listening test, no measure and send to audio buddies. I have seen no controlled by any standards test to support these conclusions.
Sooooo literal --->
This was discussed in more detail at DSPx in 1995. You dont have to believe it or even do a listening test - already done years ago. Its the effect of esr of the bipolar & film that is being shown. The effect is greater with larger cap values, of course......and heating due to repeated high currents transients only makes the esr worse. With bipolar caps the dielectric is significant part of the esr as well. I am sure you can extrapolate the time scale yourself with a typical (100+ mfd) cap used in speaker crossovers. Larger values (than this example using 1mfd) like those used in speaker crossovers would be subject to possible esr modulation as well...and thus phase modulation of the filter slope....though I didnt measure or hear from others a sound description that indicates it is audible or significant that way. Live and learn. I put it here for all now and in the future to do their own eval. its another test current impulse test (non sine wave) besides THD on caps. -RNM
Last edited:
References please, maybe I missed them. Personally I don't think this sound of capacitors stuff holds up under careful scrutiny. Private club tests where "it was obvious to everyone in the room" don't cut it. If these effects are not at least subtle, you understand my suspicion.
BTW Barrie Gilbert turned 75 this week and Paul Brokaw is back at ADI (77) 152 years combined and churning things out like 20 somethings.
Wima metallized caps are great caps... especially in low current apps and using polyprop dielectric. -RNM
Last edited:
Scott you know I don't believe in any kind of snake oils. But I found that when about 500 microfarad total electrolytic caps in 800V filter of push-pull tube amp was not shunted by 4 (four only!) microfarad of film cap soundstage was smeared, and highs sounded less natural. The difference was only in different lengths of sidebands on IMD test, hard to spot if don't know what to look for. I.e. the cap intermodulated by phase. Microseconds matter! And oxide caps due to their semiconductor and reactive properties can do that.
We all need to have our observations judged objectively. Soundstage was smeared to you expanded to another, or more likely here today gone tomorrow.
I recall Walt Jung changing small caps with better dielectrics (lower DA) on a ladder network of a CD player DAC and making, what he and others thought was a nice increasing in detail of sound. I just look for explanations that can be measured and let others do the listening. I am in JC's camp in regards to Listening Tests. Skeptical. Too many measurements that correlate with listeners descriptions.
(yes, I have done double-blind listening tests using abx box and did better than 50-50 under the conditions exposed to. But couldnt say i could always do it. Its very hard and yes a lot of things ARE very minor or hard to detect. So I rely on others and a general consensus of listeners on a variety of equipment and conditions -- if they all hear the same thing then something real could be happening. Worth looking into, I think. What tests might show what they are talking about. It is all to make the music system more real.
Its not totally unlike the European or even Chinese medicine..... typically Americans will say thier (other culture) medicine or cure doesnt work and to prove it with their (western) methodology, only. Something that has worked/cured people for centuries. I've been all over the world and find that attitude to be extremely narrow minded. Maybe there are other useful models that work as well . . --RNM
(yes, I have done double-blind listening tests using abx box and did better than 50-50 under the conditions exposed to. But couldnt say i could always do it. Its very hard and yes a lot of things ARE very minor or hard to detect. So I rely on others and a general consensus of listeners on a variety of equipment and conditions -- if they all hear the same thing then something real could be happening. Worth looking into, I think. What tests might show what they are talking about. It is all to make the music system more real.
Its not totally unlike the European or even Chinese medicine..... typically Americans will say thier (other culture) medicine or cure doesnt work and to prove it with their (western) methodology, only. Something that has worked/cured people for centuries. I've been all over the world and find that attitude to be extremely narrow minded. Maybe there are other useful models that work as well . . --RNM
Last edited:
No, it obviously was smeared. I could not understand why 4 microfarads could matter when I had hundred times more. But I soldered it there anyway and heard that all is in place. Then I checked frequency response, it did not change. I checked harmonic content, it did not change. And only what was different, heights of sidebands on IMD test.
What you had done was to improve the filtering out of very high frequency noise; all those electro's would have had high ESL, and so their effectiveness as capacitors at high frequencies was severely curtailed. Straightforward bypassing of the power supplies was achieved by the 4 uF unit, and the amount of intermodulating of high frequency grunge with the audio signal was reduced ...
Frank
Last edited:
Im sure you are right about higher atten of high freqs.
Are you sure that is the the Only thing affected which had audible consequences? Or the explanation of the change in character?
How does that correlate with any audible sound change description made?
Last edited:
- Status
- Not open for further replies.