PMA;
More to the point you and John seem to get somewhat opposite results, hence my question about possible feedback loop interaction. I never meant to imply you didn't understand, I was trying to point out that just reflections of and by themselves didn't seem like good candidates for any sound changes.
More to the point you and John seem to get somewhat opposite results, hence my question about possible feedback loop interaction. I never meant to imply you didn't understand, I was trying to point out that just reflections of and by themselves didn't seem like good candidates for any sound changes.
SY said:
Sorry SY, you'll have to do better than that. There are two huge problems with your post:
a) Your original post claiming that you could see a physical change was made *before* Scott's post saying that the parts *might* have been thermally shocked. So you are talking out of both ends at once.
b) If you reference the post Scott made, he said "the 'cryo' was probably being dropped into LN and dumped back out. Bob Adams related the story and there was at least some evidence to him that the customer was able to separate the groups honestly."
So the thermal shock part is unsubstantiated third-hand information. But what is more important is that Bob Adams (AD's digital guru for over a decade) felt that the test results were honest. I have to assume (unless Scott informs us otherwise) that this means they did it via listening tests and not via looking for "thermal shock" with a scanning electron microscope.
When you recover from your peyote trip, you might try actually *listening* to a cryo'ed chip. Who knows. You might surprise yourself...
john curl said:
We've been through this before, John.
For those who aren't already aware of it, let me give a summary of this whole saga.
Some years ago in a thread here, there was a discussion about "diode effects" in wires and cables. Someone pooh-poohed the notion and John responded that person saying there were more diodes in cables than that person would ever measure. John said he knew because he had measured them.
What John eventually produced to support this claim were some distortion measurements (spectra) he'd made using several different interconnect cables. The measurements showed significant levels of distortion (well, significant with respect to what one would expect to see produced by a cable) which varied somewhat from one cable to the other.
It was precisely because the distortion levels were so high that I was skeptical of his measurements. John took issue with this and we'd gone a number of rounds over it here and elsewhere.
Some time later I saw a post by Bruno Putzeys over on one of the newsgroups (I believe it was rec.audio.high-end) where he'd done some distortion measurements on some cables he had on hand and reported that he found no distortion even when measuring down to -145dB.
I brought this up and John dismissed it by saying that Bruno hadn't measured the same cables that he had measured.
Ok. Fair 'nuff.
So I arranged to send both John and Bruno a number of identical cables so that each could measure them and report their results.
It's not mentioned in the Audioholics article, but that's how Bruno came to receive the cables he mentions in the article.
For the freebie cables and the two Radio Shack Gold cables, I split them each down the middle like zip cord and sent one to Bruno and one to John. The RG174 cables I'm pretty sure were each made by Dan Banquer.
John's measurements were still turning up significant levels of distortion, similar to the levels he'd measured with his cables. However Bruno's measurements, which again went more than 20dB below where John was able to measure, showed nothing.
It became rather evident that the distortion John was measuring wasn't being produced by the cables but rather were being produced either by John's test equipment or by way or some sort of interaction his test equipment was having with the cables.
Charles Hansen tried to defend John's measurements by suggesting that Bruno wasn't doing the same kind of measurement as John was. By this he meant notching out the fundamental before doing the FFT spectral analysis.
Charles said he seriously doubted that Bruno's System Two Cascade could do that, as he was absolutely certain his (Charles') System One could not.
I downloaded the System One manual from Audio Precision, spent about 20 minutes reading through it and showed Charles that the System One that he'd owned for 10 years was indeed capable of notching out the fundamental.
After that Charles said he was going to do some of his own measurements, but I don't recall his ever reporting back on that issue.
Why should one do that, John, when you're getting distortion where others are not when you're measuring the same cables?
And what exactly are YOUR examples of problem cable? How much worse can you get than copperweld, perhaps aside from making a set of cables out of bailing wire?
Worse example of what, though? Worse examples of cables that your test system can't cope with?
That's a low blow, John. SY came over to your place to HELP you! And if I'm not mistaken, did he not reduce the levels of distortion you were getting with some strategically placed aluminum foil, further pointing to problems within your test system?
Yes. You do that. Show him your cable distortion measurements and then show him Bruno's measurements of the same cables.
Yes, I suggest others do exactly that. Have John show you the distortion distortion he's measuring with the cables I sent him and Bruno, and then take a look at Bruno's measurements.
se
Steve Eddy, you have NO IDEA what I measured, even with your sample cables. It is true that I detected some residual in SOME cables, but the AP and the ST have different grounding systems and this seems to matter. However, I bet that the ST grounding system is closer to MOST audio products, so it would appear to be preferred, IF you are trying to emulate realistic conditions. I personally don't make cable, have never been a cable consultant, and I could care less about cables, except for measurable and listenable differences. My test equipment has been laughed at, even though it cost about $40,000 new, because it is decades old now and a little tired. I have upgraded with Scott's IC's in critical areas, but the $30,000 AP system will measure lower and more consistently. I even have to bang the side of the THD analyzer, to get it to settle down. This is difficult what you are looking for subtle comparisons.
I met Bruno at the last SF, AES and we discussed our results together. WE have no problem with it, and I must admit that Bruno is a very good engineer.
I met Bruno at the last SF, AES and we discussed our results together. WE have no problem with it, and I must admit that Bruno is a very good engineer.
Steve, John's test equipment is elderly, but quite capable. The diode thing doesn't strike me as anything real, however. And as John is aware, elderly stuff has to be nursed a bit more than the youngsters. (My own is a mix of the elderly and new)
Charles, I'm really puzzled by your post. Grab a data sheet for a D/A. I have Analog's AD1955 sheet in front of me. Not cherry-picked, it was the first one on the list from their site. Scroll down to "temperature range." What do you see for the lower limit for storage? What is the temperature of LN2? How badly has any cryo treatment abused that spec? (back of the book answer: by over 140 degrees C) Why do you think they have that spec?
The difference the shock treatment makes is that the changes are likely to be gross and much easier to detect with simpler test gear. I was being too subtle initially, not knowing how poorly that experiment seems to have been done.
Charles, I'm really puzzled by your post. Grab a data sheet for a D/A. I have Analog's AD1955 sheet in front of me. Not cherry-picked, it was the first one on the list from their site. Scroll down to "temperature range." What do you see for the lower limit for storage? What is the temperature of LN2? How badly has any cryo treatment abused that spec? (back of the book answer: by over 140 degrees C) Why do you think they have that spec?
The difference the shock treatment makes is that the changes are likely to be gross and much easier to detect with simpler test gear. I was being too subtle initially, not knowing how poorly that experiment seems to have been done.
Just google conductivity of metals or talk to an experienced metalurgist.
http://en.wikipedia.org/wiki/Electrical_conductivity#Classification_of_materials_by_conductivity
http://en.wikipedia.org/wiki/Electrical_conductivity#Classification_of_materials_by_conductivity
OT
While i was sure to have read about the ´cryo-thing´ quite some years ago at Ed Meitner´s website, i wasn´t able to find it again there.
But it appeared also in an article of Keith Howard, HFN+RR,2001, Vol.46, The Coolest Cables
Ed Meitner told this:
´ Another thing that happened which was probably even more interesting was that Analog Devices came to us and we treated some 20-bit DAC chips. They sent out untreated and treated chips for people to try and again the same thing happened: the treated ones sounded better. Again, you have mechanical resonances and they are attenuated by reducing the residual stress.`
Jakob2
P.S. found at http://www.frozensolidaudio.com/Freezing Issue.htm
While i was sure to have read about the ´cryo-thing´ quite some years ago at Ed Meitner´s website, i wasn´t able to find it again there.
But it appeared also in an article of Keith Howard, HFN+RR,2001, Vol.46, The Coolest Cables
Ed Meitner told this:
´ Another thing that happened which was probably even more interesting was that Analog Devices came to us and we treated some 20-bit DAC chips. They sent out untreated and treated chips for people to try and again the same thing happened: the treated ones sounded better. Again, you have mechanical resonances and they are attenuated by reducing the residual stress.`
Jakob2
P.S. found at http://www.frozensolidaudio.com/Freezing Issue.htm
SY said:There's about 10 years of literature on it, most recently a paper in Revue d'Oenologie. It doesn't make wine "taste better," it just prevents it from tasting worse. Just like an amplifier, which shouldn't make music "sound better," it should just preserve what's there and not alter it.
In fact haven't there been reports that they work too well at preserving? The analogy of tubes vs. solid state comes to mind. I guess you're on the other side this time.
Jakob2 said:
I would call this explanation pure conjecture without something to back it up. It would be easy to put the CD player in an isolation chamber and remove any vibrational excitation of the "mechanical resonances".
IC's have numerous problems with package stress (big list actually). Extreme temperature stress can also cause micro-cracking or crazing along the leads (moisture incursion) or die de-lamination so it's also easy to think of mechanisms where cryo would make the chip perform worse.
I do have some doubts myself about this explanation, but it always seems wise to seperate experimental results from theoretical thoughts about the nature of the effects.
Funny enough sometimes things do work even if any explaining hypothesis is utterly wrong.
(A quick look in the history confirms this)
But it seems that ADI was indeed the company involved and that might mean the cryogenic treatment was a bit more controlled as expressed before.
Jakob2
Charles Hansen said:
For the record guys I paid extra to put Roedersteins in my stepped attenuators. I think it's just some Greyhill switch. I had a set of the most beautful military surplus rotary switches with Kelvin connections everywhere but it was one of those projects that was never going to happen.
- Status
- Not open for further replies.