Head-Fi.org › Forums › Equipment Forums › Sound Science › My cable test enterprise
New Posts  All Forums:Forum Nav:

My cable test enterprise - Page 9

post #121 of 438
In my experience with this matter. If you have had your gear for a good while and you know the sound inside and out any little change you are going to notice like night and day so in this instance any IC should sound different. If you listen to your gear occasionally or always upgrading or changing your gear around I don't think you will hear differences in the IC's. To me IC's are like Tubes, They do their one job but they all sound a little different.
post #122 of 438
Ya gotta do what makes you happy. If spending your money on cables, and other tweeks makes you happy, then go for it. But I think we can all agree that cables (like most other tweeks) is a controversial subject, and will likely always be an issue of personnel experience/belief. Differences can be measured in cables, depending on the applied methods, but the audibility of those differences is were the testing will fail. No two of us has identical hearing and our hearing changes, everyday, minute by minute even, as our bodies respond to internal and external stimulus. Further complication comes from the fact that our ability to memorize sound is very unreliable, making a direct comparison of sounds separated by even a few seconds of time, a very questionable method of getting accurate results.

Makes it nearly impossible to predict what anyones experience will turn out to be. (Now add the variable of most of us having different equipment into that equation......)

my 2 cents- expensive wire is a foolish investment if the results are not predictable.
post #123 of 438
Quote:
Originally Posted by nick_charles View Post
So now you accept that my method is okay, now that it seems to support your view, you do not think that the digitization has destroyed the differences now
My objection was just directed to your planned listening tests, not the measuring methods. Of course I'm ready to accept a result that supports my view, how could I not be! I just feared you'd chose a method (again: for the listening tests) that would make it difficult or impossible to detect possible differences.


Quote:
The differences are in the high frequency range a range which is much less audible and also the frequencies are in the -50 to -57db range which is 30 to 37 db down on the fundamental.
Sure. But don't you ask yourself how cables (note: «passive devices») can cause them? Audibility is still in question, but not that much unlikely anymore according to your measurements.


Quote:
But if you think you can tell the difference PM me your email address I and will send you the most typical sample of each (or all 20 if you like) and you can tell me if you can hear the difference.
No, thanks! I already know that I can hear differences among cables, no need to «prove it» to myself with this specific digitized sample, especially since I had a negative experience with digitized cable sound.


Quote:
The best part is yet to come, the stock cable was better, it had least signal attenuation in these frequencies apart from (*).
Doesn't matter to me at all. I don't need a special audiophile or high-priced cable turning out to be better (in a single criterion) than a standard cable. What's important is that there are measurable differences passably likely to be audible in terms of frequency and amplitude. And as with amps, I don't think there's an absolutely «neutral» cable.
.
post #124 of 438
Thread Starter 

major error found

I am sorry to say that the cymbals tests were affected by a fatal error. I have not yet decided if it materially alters the earlier tests or the square wave test, though that possibility clearly exists, but it does mean that the cymbals test at least is wholly unreliable, sorry.

My protocol involves averaging levels from 10 trials to get rid of random variation. In my first tests I did this and carefully examined the inter-trial difference and found them to always be very low regardless of the cable under test. Thus I took them for granted. This turns out to be a mistake. The spreadsheet was put up on Jan 26, page 1 of this thread so you can see the original results and I have not altered that post since

In my first tests the average level using white noise was in the order of -9 to -10db and the ADC consistently gave comparable results across all 10 trials. Never deviating by more than 0.056db at any frequency for any cable or any trial - ever.

However it seems that my CDP/ADC combo becomes less consistent at lower levels and below ~ - 49db this leads to large variability between trials. Viz (diff is difference between max and min value for 10 trials at a given frequency) - numbers are for the Sidewinder

diff(db)Freq (Hz) ave level (db)
4.242219488-55.1447754
4.129417786-52.9219821
3.628519057-52.1731768
3.598417808-49.8502896
3.548717765-54.305588
3.056217829-49.3324807
3.023419035-52.9519859
2.932019746-53.4425854
2.920319724-57.6342261
2.766219767-49.7677604
2.764519703-56.4358097
2.269517722-51.4190789
2.240019552-49.7014637
2.188217894-50.7446542
2.104419681-50.2375773
2.096219509-52.6722602

This may mean that my ADC is not good enough to reliably digitize sounds below -49db with the level of consistency required. Or the problem may be at the CD player end, since I have three that can read the CD-Rs I have created I can try different ones and see which it is.

If it turns out to be a problem with my ADC (almost certainly the case) and if I am going to carry on I need an empirically much better ADC. Having already invested mucho cash and time I am unsure if I will continue. If anyone can reccomend a quantifiably superor USB DAC for < $150 I will consider this as an option.

This is a big blow to me. It has been an interesting exercise and fun but in terms of advancing knowledge I am afraid my findings are at best questionable.

SORRY
post #125 of 438
Too late, mate! I've already archived your graph!
.
post #126 of 438
Ah, it's ok. The fact is that you tried to come to an objective conclusion, and to me, that's what's important.

Perhaps you'll be able to redo the test with better equipment in time...
post #127 of 438
Thread Starter 
Quote:
Originally Posted by JaZZ View Post
Too late, mate! I've already archived your graph!
.
You have archived something that is not reliable , I was not about to offline the graph anyway , your choice of course but I fail to see what benefit you get from it, certainly if you want to use it as proof of cable differences it does not give you a strong case. The AES would piss themselves laughing if you used the graph

Look at the graph very carefully even with the reliability issue look at the differences between the cables, look at the pattern, how much difference do you actually see ?

How often amnd how much does either cable surface above the reference ?

The answer is infrequently and not by much , but be my guest.

UPDATE: At least some part of the variation may have been due to the Denon CD player. When I switched to the Entech 203.2 fed by a digital feed the variation over 10 trials dropped by 45% and variation at lower levels (-50 to -60db) was much smaller - though still unacceptable for my purposes.

Ho Hum
post #128 of 438
No need to apoligize. Nick, are these short clips your playing? Cymbal crashes?
post #129 of 438
Thread Starter 
Quote:
Originally Posted by olblueyez View Post
No need to apoligize. Nick, are these short clips your playing? Cymbal crashes?
yes, they are chinese cymbals.

Reliability update: Not good I am afraid. I tried a different transport (Marantz) to feed my DAC and well the results were more variable so the best combo I have is a Denon providing a coax feed, this gives rise to an average 10 trial (min to max ) variation or 0.22db per frequency, this is clearly unacceptable.
post #130 of 438
Maybe you need a better coax cable.
post #131 of 438
Quote:
Originally Posted by Uncle Erik View Post

... That's why, in over 30 years of cable controversy, no one has ever passed a blind test.
Is this really true? 30 years is a long time.

Quote:
Originally Posted by olblueyez View Post

One more question, if it is as you say "Placebo" then why would many different people report the same results?
For the same reason the Greek empire and then the Roman empire believed in Mt. Olympus.

Quote:
Originally Posted by odigg View Post

The cable argument has been going on for a long time. James Randi even offered 1 million dollars to
somebody who could tell the difference between cables in a controlled test.
If this is actually true, what are the people who can hear the differences between cables waiting for?
A million dollars is a lot of money.

There is also the Clark amplifier challenge waiting out there for someone to collect on.



There is no spoon !

USG
post #132 of 438
Quote:
Originally Posted by nick_charles View Post
If it turns out to be a problem with my ADC (almost certainly the case) and if I am going to carry on I need an empirically much better ADC. Having already invested mucho cash and time I am unsure if I will continue. If anyone can reccomend a quantifiably superor USB DAC for < $150 I will consider this as an option.
I'm not sure if any of these are quantifiably superior, simply because I haven't seen any RMAA or other tests on them. They are under $150 though.

Some pro audio interfaces
Roland/Cakewalk UA-1G which I strongly suspect is a mostly cosmetic update of the Edirol UA-1EX
M-Audio Transit
Fast Track USB

Audiophile DACS.

Super Pro 707
Trends UD-10.1

Some of those can convert USB to an optical/coax signal to be sent to your Entech.

Is anybody interested in helping nick_charles out with funding? This is an interesting project and I wouldn't mind donating a little bit to help offset the cost of a relatively inexpensive USB DAC or something along those lines.
post #133 of 438
Quote:
Originally Posted by odigg View Post
I'm not sure if any of these are quantifiably superior, simply because I haven't seen any RMAA or other tests on them. They are under $150 though.

Some pro audio interfaces
Roland/Cakewalk UA-1G which I strongly suspect is a mostly cosmetic update of the Edirol UA-1EX
M-Audio Transit
Fast Track USB

Audiophile DACS.

Super Pro 707
Trends UD-10.1

Some of those can convert USB to an optical/coax signal to be sent to your Entech.

Is anybody interested in helping nick_charles out with funding? This is an interesting project and I wouldn't mind donating a little bit to help offset the cost of a relatively inexpensive USB DAC or something along those lines.
Im in.
post #134 of 438
Quote:
Originally Posted by upstateguy View Post

If this is actually true, what are the people who can hear the differences between cables waiting for?
A million dollars is a lot of money.

There is also the Clark amplifier challenge waiting out there for someone to collect on.

There is a strong political reason why we've never heard of a serious attempt.

Let's assume there is a person named "Mozart." Mozart is a self-proclaimed golden ears and has demonstrated that in blind tests he can score well very on tests comparing MP3s to FLAC, distortion audibility tests, etc. Statistically, Mozart is 3 standard deviations above the mean in his hearing ability.

Mozart applies for the Randi challenge.

Problem is, who is going to supply the expensive wire? It's not in the interest of most, if any, audiophile grade wire suppliers to provide wire or even allow their wire to be used in this test. What if Mozart, a person who possesses measurably superior hearing ability, cannot tell the difference?

The news would hit the audiophile world. People would probably run away from purchasing wire from that company. People with that wire might be inclined to switch to something else. There would probably be major financial repercussion to that one wire company.

If the Randi Foundation decided to buy some expensive wire and do the test without the manufacturers permission, there might be a lawsuit either preventing the test or to suppress the information if the Mozart cannot tell the difference.

Anyway, I'm not trying to support either side of the cable argument. It's just easier to talk about and perform tests on a small scale (like on this forum) because they're aren't massive political and economic ramifications. nick_charles' test results might convince a few people that cables aren't worth it. It's not going to become a world-news worthy event (unless you think everything on slashdot is news worthy). Nick_charles - no offense intended.
post #135 of 438
Come on people, lets drum up some cash for Nick!!!
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › My cable test enterprise