Listen to cable directionality

[christopher3393]

Listened to A/B files twice. Detected no differences.

[plissken]

I can confirm that Alex's file is exactly the same. I expected nothing less.

 

After all his ranting and raving I'm still not sure what his point was. Did he massage the file somehow?

[mansr]

After all his ranting and raving I'm still not sure what his point was. Did he massage the file somehow?

 

Probably saved it to a USB stick with a Regen and linear power supply or something equally pointless.

[Jud]

Hi Dennis and Alex -

 

As you know by now, I don't think A/B tests are effective most of the time, so I didn't want to expend a lot of time and effort on this; in fact I think the more time and effort one expends, the less effective such tests are likely to be, even if there are differences (other than in loudness and perhaps a couple of other quite obvious sorts of differences).

 

So I wanted to run through these "once over lightly" and just note my initial preference, if any. On a couple of these, however, I actually forgot my initial preference by the time the next selection came along (yes, I'm capable of that), and so wound up going back and forth a couple of times. I'll note those in my brief report below. I was going to do this by PM, but as long as everyone else seems to be doing theirs in-thread, here goes nothin'. These are (with noted exceptions) my initial preferences first time through. My guess (putting my science hat on for a minute) is that these preferences could well be governed by loudness differences between the end of one sample and the beginning of the next, though I haven't gone back to check that.

 

1 - B

 

2 - A

 

3 - B (this is one where I was writing "B" as selection 4 was starting to play, but then wondered whether I had misremembered and it was actually "A." I went back and forth a couple of times before coming back to B.)

 

4 - A

 

5 - B (I went back and forth once before settling on B)

 

Mono track - Dennis, I would have had to pay too close attention in a very short time to try to find the point where a switch may have occurred, so I can't tell you that. My loose (nearly subconscious) impression after listening a couple of times is that the right channel sounded like something may have been going on, so I'll pick the right channel on that very informal basis.

 

Alex - My initial impression was that I liked your track better, but after going back and forth a few times I couldn't tell a difference. Unfortunately I think that's probably too vague to really satisfy anyone, but those were my impressions.

 

Finally, and most important - Others may already know, or you may have posted these in a part of the thread I missed, but Dennis, I really liked tracks 2 and 4. Who did them, so I can go buy them?

[Middy]

Hi I have not read any further until posting. No idea which is which.

1b

2b

3b

4b

5b

Only say may be a touch louder and maybe wider. A bit clearer at the edge of the sound stage. Singers seem more in focus. Maybe a touch more bass... They could all be the same and just me bias...

I hope this helps just for the fun factor.

Some lovely music btw. .. Thanks

 

Sent from my SM-G900F using Computer Audiophile mobile app

[Sal1950]

The response is low for only one reason: most people are afraid to be wrong. Am I right?

 

I don't think that's the reason. I'd guess the majority of folks, even among audiophiles, aren't terribly interested in these specific tests or don't think they'll prove anything. And if I'm any indication, many people are, in the words of an old foreman on a long-ago summer job, busier than "a thousand-legger with a hotfoot."

 

Please Jud, that is exactly the reason. Many folks here go on and on for hours at their keyboards bragging about their golden ears and ultra resolving systems. They talk about all the ridiculous things they think they hear like cable directionality, to SATA cables, USB widgets, magic grounding caskets, etc, etc. But any time they're ask to compare a few 30 second files and chose from things they claim make day/night differences they run like roaches from the light.

OH, I know, "blind tests don't prove anything". Sorry but,Yes they do. They highlight what is real against what is imaginary controled mainly by bias.

Too busy is a lame excuse for people that seem to live on this forum 24/7.

[Superdad]

Without wading into the waters of the direction that the wire strands are drawn, twists, and other topology issues, I will point out that most shielded audio interconnects (at least those based on twisted pairs, quads, etc. plus shield) have their shield tied only at one end.

 

So regardless of other above-mentioned factors, turning a cable around (upstream/downstream) does electrically change something.

 

 

John Swenson has recently been schooling me on how incorrectly shielding is done on virtually ALL audio cables, and that neither tying the shield at one end nor tying the shield at both ends are effective. The right way to make a effective shield would be to not have it tied at either end, but to have a single separate wire--OUTSIDE the outer jacket--tied to shield at BOTH ends--joining the two ends of the shield. When he is not frightfully busy, I'll try to get him to jump on and explain the principles.

 

--Alex C.

[witchdoctor]

Without wading into the waters of the direction that the wire strands are drawn, twists, and other topology issues, I will point out that most shielded audio interconnects (at least those based on twisted pairs, quads, etc. plus shield) have their shield tied only at one end.

 

So regardless of other above-mentioned factors, turning a cable around (upstream/downstream) does electrically change something.

 

 

John Swenson has recently been schooling me on how incorrectly shielding is done on virtually ALL audio cables, and that neither tying the shield at one end nor tying the shield at both ends are effective. The right way to make a effective shield would be to not have it tied at either end, but to have a single separate wire--OUTSIDE the outer jacket--tied to shield at BOTH ends--joining the two ends of the shield. When he is not frightfully busy, I'll try to get him to jump on and explain the principles.

 

--Alex C.

 

That would be much appreciated, thanks!

[Middy]

That would be nice...I've looked at more bloody shielding than Captain America re runs . What to do with a shield drain wire has been driving me bonkers...Like reading this thread but fun..

 

Sent from my SM-G900F using Computer Audiophile mobile app

[esldude]

How many times have I asked you to explain comments that you push off as simple facts? I couldn't even begin to put a number on it. You either ignore me or talk about something else, and pretend to answer with info that doesn't apply. Now, as far as gracing you with my aural perceptions, you can easily get that answer many times over if you've read my threads.

 

But I guess that's not good enough and you still want me to take your test. My answer to that is no, for 2 reasons. First, I don't trust you as far as Sal can throw you. I think you want to be right so bad that you would have no problem fabricating anything just to make yourself look good. Second, I know this a big deal for you because its about listening and you're all excited, but I've already done this 20+ years ago when those cables were current. I still have 3 pairs of them sitting in my closet. Why would I take your half ass test when I can just listen to the ones I have? That said, even if I took the test, you've already passed judgement, so my answers won't matter one bit. You want the test to show a certain result, and I have no doubt you'll get the answer you want.

 

I wouldn't bother to post except for two things. One is you saying I would fabricate anything to make myself look good. The recordings were done as I described. Everything setup and burned in for about 62 hours. A recording of the 5 tracks and some simple test signals. Cables flipped and recorded again using the same source both times. It seems a better thing to do than threads that become childish bickering which is leading no where. Here at least anyone interested can listen and have a common thing to discuss or express an opinion about. They get to hear two directions of a cable without knowing which is which. That is an at least inconvenient thing to manage in person because of the time switching and the limits of keeping it actually blind.

 

The other point is you saying I "passed judgement". I had an expectation of the results yes. On the other hand if the results had a wide consensus with most choices being the same by most people that would be very interesting and I would attempt to find out why if possible.

 

If you didn't care about results, and won't listen to the files then why did you post in the thread? If my not answering your questions has insulted you then ask whatever you want here. I will either answer, say I can't answer or suggest some sources if the answer is too complex for forum postings. I do sometimes become flippant especially about things that have been banged around too many times here before which may seem rude if you are reasonably recent to the forum which you are. Wasn't anything personal.

[esldude]

I don't think that's the reason. I'd guess the majority of folks, even among audiophiles, aren't terribly interested in these specific tests or don't think they'll prove anything. And if I'm any indication, many people are, in the words of an old foreman on a long-ago summer job, busier than "a thousand-legger with a hotfoot."

 

I understand people are busy. Some of it probably is people not interested in the thing being tested or think it shows nothing. Yet I see threads on topics that get lots of views and lots of posts. Put up a listening opportunity about that very thing, and not many people take part. I can't know for certain why, but it is the result each time.

 

Nevertheless I do think it mostly is people are afraid to be wrong. They have a bit of ego or sense of self invested in their gear and being an audiophile. It looks too much like a test or a challenge (and sometimes it is a challenge). Some of the things I have put up as polls were anonymous. One could listen, vote and no one other than themselves know what their personal results were. But they would know and I think they didn't want to find out they were wrong in choices.

 

Now yes, I realize none of these sorts of things prove anything. Our listening panel is not trained, the gear is different for every person doing this, and we don't know if those people have significant hearing defects. I would have thought it more interesting than arguing that I see across entrenched positions in threads.

 

Finally a few people have contacted me saying I only put up comparison files that people can't hear. It mostly works out that way. It isn't by design really. I was very surprised with the 8th generation copy files. I couldn't hear it blind, but I am old enough my hearing isn't superb anymore. I thought maybe others with younger better ears could hear something. The DAC comparo had two that I can hear blinded. Yet with the paucity of results it wasn't very conclusive. Whether anyone voted or not I thought it would be interesting. Reading people's descriptions of how DACs sound you would think them very, very different. Yet when you don't have a label or haven't spent money on them are they really that different? I think those files should cause one to pause before thinking of dropping $2-150K on a DAC for sonic reasons. That might seem like a challenge, but it was really meant to be interesting and thought provoking.

[Axial]

I'd guess the majority of folks, even among audiophiles, aren't terribly interested in these specific tests or don't think they'll prove anything. And if I'm any indication, many people are, in the words of an old foreman on a long-ago summer job, busier than "a thousand-legger with a hotfoot."

 

With almost 150 posts people are interested by AudioQuest audio cable's signal directionality.

The listening test from the OP is just for fun; we have nothing to lose.

If one hundred members do the test, and that we examine our results, we are getting closer to science.

[Don Hills]

... My guess (putting my science hat on for a minute) is that these preferences could well be governed by loudness differences between the end of one sample and the beginning of the next, ...

 

Bears keeping in mind when preparing samples for testing.

[esldude]

Exactly what interconnect did you use? From what I can tell, the Diamond name is only used on digital cables.

 

AudioQuest Cable Archive

 

Here you can look at archival info on AQ cables. Under analog interconnect down the page are links for Diamond X2 and X3.

 

AQ made both analog interconnects and digital cables with the Diamond name. These are probably 10 or maybe closer to 15 years old. AQ Diamond Hyperlitz. These are the X3 version. Same cable as Diamond XLR with the third leg grounded at one end only as George described earlier. I once inquired about getting XLR ends on them and AQ confirmed with the X3 that could be done. I also have a Diamond AES/EBU cable. The construction externally appears to be identical to the analog cable though spec'd by AQ as a digital interconnect. If this has the correct 110 ohm impedance I don't suppose that would cause any trouble using it for analog. Then again, maybe it isn't a proper digital cable.

[Don Hills]

...

John Swenson has recently been schooling me on how incorrectly shielding is done on virtually ALL audio cables, and that neither tying the shield at one end nor tying the shield at both ends are effective. The right way to make a effective shield would be to not have it tied at either end, but to have a single separate wire--OUTSIDE the outer jacket--tied to shield at BOTH ends--joining the two ends of the shield. When he is not frightfully busy, I'll try to get him to jump on and explain the principles. ...

 

I would be interested in that. It sounds to me like a recipe for increasing the noise injected into the shield, and hence coupled to the signal conductors.

[esldude]

Without wading into the waters of the direction that the wire strands are drawn, twists, and other topology issues, I will point out that most shielded audio interconnects (at least those based on twisted pairs, quads, etc. plus shield) have their shield tied only at one end.

 

So regardless of other above-mentioned factors, turning a cable around (upstream/downstream) does electrically change something.

 

 

John Swenson has recently been schooling me on how incorrectly shielding is done on virtually ALL audio cables, and that neither tying the shield at one end nor tying the shield at both ends are effective. The right way to make a effective shield would be to not have it tied at either end, but to have a single separate wire--OUTSIDE the outer jacket--tied to shield at BOTH ends--joining the two ends of the shield. When he is not frightfully busy, I'll try to get him to jump on and explain the principles.

 

--Alex C.

 

My initial response is, why do you think they invented balanced connections? That is really what we need. I know marketing conditions play a part, but any serious audiophile gear should have balanced. I more or less agree with what you are quoting from John here. In fact to get the lowest noise one usually needs to look at the whole system and pick single ended grounding ends based on the effect overall. Yet, while there are differences with much gear they are not big differences. I say that having just done a 5 channel rig where it made a big humming difference. 5 channels with long cable runs of all things in home audio could use some XLR's and they are scarce though not non-existent.

[sandyk]

Hi Jud

We had at least 3 members preferring 2a

Please check your PMs re the Fiona Apple album

Kind Regards

Alex

[One and a half]

Hi Jud

We had at least 3 members preferring 2a

Please check your PMs re the Fiona Apple album

Kind Regards

Alex

 

Unless there's a miscount, there's 4 for 2A and 4 for 2B.

[sandyk]

Unless there's a miscount, there's 4 for 2A and 4 for 2B.

 

In other words the whole exercise proved 5/8 of SFA ?

Subtle differences like this at the analogue output of a DAC or preamp, are highly unlikely to fully survive a pass through (I presume) an inexpensive USB powered A/D converter.

Any perceived differences were NEVER going to be night and day differences, even before an A/D conversion.

Neither were we able to prove that any perceived differences were attributable to swapping the cables around, and could have been due to removing the cables and reinserting them again, possibly improving the plug to socket contact resistance.

[esldude]

In other words the whole exercise proved 5/8 of SFA ?

Subtle differences like this at the analogue output of a DAC or preamp, are highly unlikely to fully survive a pass through (I presume) an inexpensive USB powered A/D converter.

 

Well it is the same ADC which was used when 8 passes thru it were apparently not heard vs the digital original with no passes thru the analog world. I am comfortable the ADC isn't a big bottleneck though others might still question that.

 

Any perceived differences were NEVER going to be night and day differences, even before an A/D conversion.

Neither were we able to prove that any perceived differences were attributable to swapping the cables around, and could have been due to removing the cables and reinserting them again, possibly improving the plug to socket contact resistance.

 

So you are telling me, that leaving a cable connected for 62 hrs, and then removing and replugging it has a good chance of being audible? The 1 khz distortion had it highest harmonic (the second) at -109.4 db in the bad channel in both directions referenced to the - 2.0 dbfs test tone. The good channel was about a decibel lower. So this .00032 % distortion level still hides something related to replugging the cable that is audible? The peak sample level before replugging was .00012 db higher than the highest peak after replugging in the 10,000 samples I looked at from the test signal.

 

I know you were saying it was possible and not concluding it was a sure problem. I think seeing these distortion numbers unchanged with direction says contact resistance was low enough to be a non-factor. The peak values of the signal being so nearly identical points to the same conclusion. I believe contact resistance differences can safely be marked off the list.

[sandyk]

So you are telling me, that leaving a cable connected for 62 hrs, and then removing and replugging it has a good chance of being audible?

 

Neither scenario would be capable of a night or day difference !

Any difference would be subtle at best, and passing it through another A/D conversion would make it even harder to hear any differences. Perhaps it may have been better to save as 24 bits (24/44.1) at those VERY low levels ?

[Jud]

Neither scenario would be capable of a night or day difference !

Any difference would be subtle at best, and passing it through another A/D conversion would make it even harder to hear any differences. Perhaps it may have been better to save as 24 bits at those VERY low levels ?

 

Then you're into how/whether dither if used changed the sound, and if it wasn't used, whether it should have been.... Sources of possible questions are pretty well never ending. I'm sure Dennis realizes that full well, and proposed the test out of curiosity and whatever entertaining (enlightening?) discussion material it might provide.

[sandyk]

. I'm sure Dennis realizes that full well, and proposed the test out of curiosity

 

Dennis never posts a test out of curiosity. They are invariably designed to destroy Subjective credibility, whether with boutique cables, 16/44.1 vs high res PCM or DSD etc..

I wish this wasn't the case, because if he decided to investigate the reasons behind the reports instead of trying to shut them down, we would all benefit from his undoubted expertise.

I doubt that Dennis could honestly say that the results obtained here weren't as he expected.

I also doubt that he has ever heard a difference between his own cables with either orientation.

 

Dennis set out here to try and demonstrate that it didn't matter a damn which way around the cables were connected.

[mansr]

Perhaps it may have been better to save as 24 bits (24/44.1) at those VERY low levels ?

 

The files Dennis provided are 24-bit.

[Jud]

Dennis never posts a test out of curiosity. They are invariably designed to destroy Subjective credibility, whether with boutique cables, 16/44.1 vs high res PCM or DSD etc..

 

Judging by the objective vs. subjective numbers in the forum, at least among relatively frequent posters, it appears Dennis's evil plans for world objectivist domination have been foiled.

 

Slightly more seriously, I've found Dennis to be curious about many things in audio, and as his explorations occasionally pique *my* curiosity, that seems to me to be a good thing on the whole.

 

I wish this wasn't the case, because if he decided to investigate the reasons behind the reports instead of trying to shut them down, we would all benefit from his undoubted expertise.

I doubt that Dennis could honestly say that the results obtained here weren't as he expected.

I also doubt that he has ever heard a difference between his own cables with either orientation.

 

Dennis set out here to try and demonstrate that it didn't matter a damn which way around the cables were connected.

 

I would guess that's what Dennis expected to find, and that's how he would interpret the results. As I've said before, I don't think A/B tests are effective for these sorts of differences, if they exist, so to me it's really not a big deal either way. The test wasn't terribly long or involved, so it wasn't a bother, and most important I discovered some good music.