Several of the papers found when googling do indeed mention also the cables themselves, not just the connectors. However, the major issue seems to be that the cable construction is such that a good connection with the connector can be established, calling for solid centre conductor and non-braided shield. In addition to this, at least the Amphenol paper also mentions issues with the actual cable itself, where sometimes significant PIM can occur due to bad design and/or quality. One of the examples is coaxial cables having a copper clad aluminium conductor in the center, where it sometimes happens that the copper flakes off. Impurities in the dielectricum seems also to be a source of PIM, as well as a few other things. Howeve, it seems that the shield is usually not of any concern here, except for the cable-to-connector assembly issues. While there is discussion of diode effects occuring in connectors, I could not find any mentioning of such phenomenae in cables.
Do remember, however, that these are issues occuring in certain types of microwave applications, and it is not in any way clear that it could have any relevance at all to audio. One first thing to ask would be if the design guidelines for low-PIM microwave cable assemblies have any correlation with what cables people claim to sound better for audio? If PIM is an issue, there should be such a correlation.
Do remember, however, that these are issues occuring in certain types of microwave applications, and it is not in any way clear that it could have any relevance at all to audio. One first thing to ask would be if the design guidelines for low-PIM microwave cable assemblies have any correlation with what cables people claim to sound better for audio? If PIM is an issue, there should be such a correlation.
Folks, I do the measurements, I have had SY over to check my work. It is inconclusive, to be sure, but it is real. If anyone has any questions, please contact SY as well, and ask his opinion.
I was wondering too what measurements and results applicable in the GHz range have to do with audibility in the audio range.
The old adage: everything has an effect somewhere. The question: is it of any significance to the task at hand?
Christer, "In general, for connections and cable assemblies, PIM is not frequency dependent." This is from the Amphenol paper. Why should it have a frequency range dependence? Apparently, the microwave folks found it first.
john curl said:
OK, I missed that sentence, but I had been suspecting that to be the case, since most of the mechanisms discussed do not seem to be frequency dependent. So then the question is, does audio have the similar type of situation where PIM is an issue? In microwave the main problem seems to arise when we have strong transmitter signals that cause IM products at frequencies where we simultaneously try to receive weak signals. I suppose it is not really important whether signals are transmitted or received, that is just the common case in microwave applications. So possibly PIM could be an issue since audio deals with a broad band of frequencies of very varying amplitude, which is a similar situation to the microwave case. Note that I say possibly. There is no evidence that this is the case. We don't even know for sure what type of phenomenon you are measuring, even if we assume the measurements are correct. It could be PIM. It could be something else.
The papers do have a lot of design guidelines for how to assemble PIM-free cables assemblies, so one could try to make two cables, in one case trying to follow these design guidelines and in the other case trying to go against the guidelines. Then try to measure IM in the cables. If any IM can be measured at all, there should be more of it in the second case.
Christer, I have at least 2 one meter audio cable types that measure almost perfectly in my test. They create my 'baseline' for the measurement. Also, please think about 7th harmonic distortion. This is VERY important in audio, and when weighted properly, brings some of the distortion that I measure to very high relative levels.
John, I think you missed my point. You seem to measure something, but we don't know what it is. The idea of making two pairs of cables, one attempted as PIM-free as possible, the other made to deliberately exhibit as much PIM as possible, would be an interesting test. If the "low-PIM cable" shows no or very little distorsion, and the "high-PIM cabel" shows clearly higher distorsion and of a type similar to what you have measured before, then it would be a hint, although no proof, that you are indeed measuring PIM. Since the microwave people already seem to know how to make almost PIM-free cables, why not make use of their knowledge, if it is indeed PIM we are dealing with?
Hi Christer,
I think you posed a valid test idea. I think it's a test well worth doing if anyone has the equipment and time.
-Chris
I think you posed a valid test idea. I think it's a test well worth doing if anyone has the equipment and time.
-Chris
This from a site by David Messinger, aka HIFI AKTIV, and my translation of it.
http://www.hifiaktiv.at/diverses/re...men/realistische_betrachtungen_blindtests.htm
I have deleted the part re: cd players, as it ist not important for the topic under discussion.
http://www.hifiaktiv.at/diverses/re...men/realistische_betrachtungen_blindtests.htm
I have deleted the part re: cd players, as it ist not important for the topic under discussion.
Test results
>Post #629
>This from a site by David Messinger, aka HIFI AKTIV, and my >translation of it.
>http://www.hifiaktiv.at/diverses/re..._blindtests.htm
It's a shame more info wasn't presented as to the rationale and design of the test. There are specific decisions that must be made regarding what constitutes a positive result and what constitutes the null(the criteria and confidence levels). Looks like this test is anecdotal at best, it's a shame a statistical design wasn't used to yield some statistical power. (see post 570)
Nevertheless, it does represent anecdotal confirmation that the null hypothesis has yet to be challenged
auplater
>Post #629
>This from a site by David Messinger, aka HIFI AKTIV, and my >translation of it.
>http://www.hifiaktiv.at/diverses/re..._blindtests.htm
It's a shame more info wasn't presented as to the rationale and design of the test. There are specific decisions that must be made regarding what constitutes a positive result and what constitutes the null(the criteria and confidence levels). Looks like this test is anecdotal at best, it's a shame a statistical design wasn't used to yield some statistical power. (see post 570)
Nevertheless, it does represent anecdotal confirmation that the null hypothesis has yet to be challenged
auplater
I think your analysis did a fine enough job challenging it. It was from all appearances a poor test. Which sounds 'better'? What was that about?
This question could not be adressed due to the fact that no differences could be discerned.
I agree - proper statistcal anlysis is important. But what is there to analyse if the "hit rate" was 50% as he stated? At least that is how I interpret his statement that the results equal chance.
audio-kraut said:
Ahhhh .... but THAT system doesn't quite have resolution of THIS system ... and THOSE people obviously don't have as golden ears as THESE people .... and cables used are known crap as everyone knows that TruePremiumPerfectCableWithAlot (tm) eats those crappy Cradas and VdH for brunch ....
and so ad nauseam.
Yes, cables withstood a test of time very well - just as clairvoyancy, astrology, UFOs and religion.
All still going strong - in fact stronger than ever.
Bratislav
PS some very vocal individuals also claim to be able to hear cable directivity (another "day and night" difference), whether amplifier feet are square or round, type of a single resistor used, type of alloy heathsink is made of; what's left ? Hear a color of the anodizing dye ? Where is the car parked ? Whether two or there possums are walking my powerline ? 😉
audio-kraut said:
That misses the point. The only question of importance was: 'was a difference reliably discerned'. Asking which is better confuses the testing process by overlaying an additional value judgement. It's like the DBTs designed to pick 'which is which'. These are secondary to the real test, 'is a difference reliably heard'. I didn't bother getting into the other crippling faults - that it was sighted, that it took place in unfamiliar acoustics with unfamiliar equipment, that the tester was apparently in the room and aware of which source was playing, that judging from what was posted the latter constituted the presence of a skeptical authority figure in the room challenging the subject to hear a difference, etc. etc.. It was a poor test on multiple levels and wouldn't survive seconds in a scientific peer review environment.
Use science to hold science to the feet of the subjectivists. This test was junk science and arguably worse than the Hi-Fi + article lambasted earlier.
I was reading an article in the UK hifi industry trade rag the other week. 52% of turnover in the average shop is due to sales of accessories. Take that as you will. 😉
Right on. And it also biases the test; the question assumes that there IS a difference, and the listener is clearly being coached. Bad design for all the reasons you say and more.
pinkmouse said:
Another anectodal point is that markup margins on accessories are far higher than on a 'real' component.
Makes kind of sense, when you can buy TEN complete DVD players, with many mechanical, moving bits that have to be made with high precision, plus lots of not so trivial electronic bits (including dual wavelenghth laser module), for a price of ONE nicely colored piece of wire and a plastic 'Impedance Termination (sic)' box that is either empty or has a single resistor (and maybe a capacitor) in it.
Madness ?
I thought it was clear from the result that NO (NO, you understand, NO) difference was discerned (to repeat, NO difference).
The test was also blind - I guess you either have trouble reading or my translation ain't worth ****.
It is also a very convenient excuse, always forwarded by the believers that the circumstances were not familiar. That argument is so worthless - if there is a difference, then she will be heard in A/B comparison or not) as to be wasting just space.
I also agree with the point about "better". I am becoming quite convinced that many prefer the sound of distortion, perhaps without realizing it. DACs (no (') is that better SY?) with linear interpolation algorithms... amps with reduced rail capacitance... really gets me wondering.
You have to at least commend their goal... while the effort and results were lacking...
After seeing "powercords, and demagnetized CD's", it was way too much for me to read on a nice Sunday afternoon.
🙄
You have to at least commend their goal... while the effort and results were lacking...
After seeing "powercords, and demagnetized CD's", it was way too much for me to read on a nice Sunday afternoon.
🙄
poobah said:
What is even more surprising is that most people nowadays seems to listen to, and prefer, "music" with deliberate distorion. I am afraid that most people no longer knows what acoustical instruments sound like in reality, in a live concert with no electronics involved. What reference do they have for what sounds better or worse if there is no "reference version" of their music, but even live concerts are at the mercy of audio reproduction equipment, and the instruments themselves don't function properly without microphones, or even use synth circuits. Even a moderately good grand piano out of tune sounds better than most electric pianos, i fyou ask me, and that both live and in reproduction.
This is, by the way, one reason why I treat most peoples listening opinions as probably useless, at least for me. Unless the listen to acoustic music of some form and know what it should sound like, then I don't see why their opinions should be of much relevance to me. OK, certain aspectes might still be relevant, with some other types of music, to be fair.
- Status
- Not open for further replies.