|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
65.19.76.104
In Reply to: RE: Cable shootout.. posted by tomservo on June 21, 2014 at 19:29:11
I nulled out two of the cables using the WAV files. The null was at -75 dB relative to the original music. By cranking up the volume by +26 dB and listening carefully I could hear the difference, which sounded like quiet music in the background, not noise or distortion.
I would have to see an indication of repeatability (multiple tests with a given cable, with and without disconnecting and reconnecting and done in a random order by cable) before I would attribute any differences to the cables themselves and not some other aspect of the experiment. It's all too easy to conduct experiments that one believes to be measuring one thing only to discover one was actually measuring something else.
To put things in perspective (not sure that's quite the word) I've seen a number of audiophiles report big differences in files that differed only in the low order bit of a 24 bit format and other files that were identical except for a few hundred extra leading zero samples at the start. In one case a file that differed only at -138 dB was described as "completely trashed".
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Follow Ups:
I went back and repeated my difference test (mogami vs. gotham). This time I boosted the level digitally so I could compare the original and the difference without running the risk of blowing out my ears. The difference had a high noise level, and the music was obviously thin and gritty, extremely gritty on peaks. I'm thinking that the DAC was putting out more or less the same signal, subject to a different load, especially a different cable capacitance, but most of the differences (especially the noise and level dependent grit) are probably artifacts of the ADC being boosted by 75 dB. It would be nice to know what ADC was used to capture the output of those cables. I'll bet that the difference signal would have been quite a bit more musical if a clean ADC like the Grimm and been used.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Hi Tony
Cool, somebody looked!
With a null -75dB down and IF the spectrum you heard was that of the recording, for a signal cable like these, if driving a 5K load, a difference of about 7 Ohms in series R would do it.
If this were a speaker cable, that much difference could be the result of one cable having about 1/1000 of one Ohm more resistance than the other.
(Assuming an average load Z of 6 Ohms).
If the null had more high end, that would be from one having a little more series L than the other or signal cable, parallel C. Any form of non-linearity would be a "difference" and be most audible (like when you use this on electronics etc).
“I would have to see an indication of repeatability (multiple tests with a given cable, with and without disconnecting and reconnecting and done in a random order by cable) before I would attribute any differences to the cables themselves and not some other aspect of the experiment. It's all too easy to conduct experiments that one believes to be measuring one thing only to discover one was actually measuring something else.”
Sad but true and same for the -138dB.
Best,
Tom
As far as I could tell, the levels were matched to a tiny fraction of a dB, i.e. to all the significant digits displayed by Soundforge. This was surprisingly good for analog comparisons. A slight drift in temperature could cause a gain differences needed to get a better null. One thing that I noted was that the two files were perfectly synchronized. This indicates that the DAC and the ADCs involved were run off of a common clock for both runs. Otherwise, there would have been clock drift across the file with the samples going in/out of phase.
I've done a lot of null tests, usually digital to digital involving comparison of sample rate conversions of one sort or another. These rarely give 75 dB nulls, even when there is no theoretical reason for a null to be more than a few dB from the resolution given by the bit depth. (An example is resampling 44.1 to 88.2 and back to 44.1. Getting good nulls was impossible due to sub-sample delays in the filters, fractional gain errors, phase shifts and filter pass-band ripple. Eventually, I figured out how to do this kind of conversion without excessive error, except in the pathological cases where there was energy too close to Fs/2.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Hi Tony
Man you are lucky to be working in the digital domain where the noise floor is so low. A -75dB null is very low indeed, if one were using an old style HP analogue distortion analyzer, a -75dB null on a Voltage signal would be unreadable.
Of course there are other complications there too haha.
One thing that was puzzling at first but made sense later was that measuring a low distortion loudspeaker, one needed to subtract the amplifiers voltage and harmonics from the loudspeaker (used vector subtraction with an Hp3562) because the amplifiers harmonics could either add OR subtract from the loudspeakers distortions depending on their phase.
Best,
Tom
Some of the best digital SRCs nulled out in the -40 dB range.
I worked with an analog computer in the early 1960's. The individual "operational amplifiers" were "chopper stabilized" and were rated to be accurate to 0.01%. By the time you were done solving a problem, e.g. integrating the rocket equation for an ABM system, you would be very lucky if the accuracy was 0.1%.
This machine was part of a "Hybrid" computer. You can see what this looked like in the attached brochure. A real fun teenage summer job, for which I was paid the minimum wage at the time of $1.00 / hour. (Years later, I learned that my boss charged the U.S. Air Force $7.00 / hour.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
At this late date, I don't know if it was Analogue or Digital but I once visited the Sinclair (petroleum) research labs outside of Chicago. A friend of my dads was a chemist in the hi pressure lab and had some missing digits to prove it.
The computer in question was a vacuum tube model with a huge plugboard as in the EAI brochure.
One of the brain guys my dad knew had written a great circle navigation program for this machine. And it worked! Somewhere in my 'stuff' I may still have the 1/4 MILE of punch tape of the program. Only later (mid/late 70s) could you get a great circle navigation program for the then wacky revolutionary HP handhelds.
I don't know how much floor space the computer took up, but the AC system EASILY was enough for a small apartment building!
The provided link shows a video of a WORKING 'Difference Engine #2' designed but NEVER completed by Charles Babbage. I believe the Vid shows the machine built by the Smithsonian to the original plans. One intent was to make a CORRECT trig table for navigation. Up to that time trig tables were hand copied and computed. Lots of errors crept in which made for problems, as you can imagine. Babbage ALSO designed a printer which I don't think has been built to this day.
An early, but certainly NOT the earliest analogue computer.
Too much is never enough
There are two Difference Engines in existence.
Both were built by the London Science Museum. One is still there and the other, the one in the video linked, is owned by Nathan Myrhvold and on loan to the Computer History Museum in Mountain View, California.
What happened to the one built by the Smithsonian? That's where I first read about it….and they were in process of building it and trying to find the money to build the PRINTER to go with it.
Do I mis-remember?
Too much is never enough
The two engines were built by the London Science Museum in '89-'91 for Babbages 200th birthday and in '00 they also built the printer.
The Smithsonian may have one of the engines based on the Babbage design and built by the swedish lawyer and inventor Per Georg Scheutz and his son Edvard. They sold one to the US gov in 1860. I can't find anything that may indicate that the Smithsonian ever built one for themselves.
Must be that pesky brain cell again.
Too much is never enough
I think a lot of the nonsense about directionality comes from cables (Monster comes to mind) with direction arrows on them. It has nothing to do with signal flow, but everything to do with good shielding practice. There's a rule that says signal current should never flow in shields. The usual practice is to use 2-conductor shielded cable, and only connect the shield at one end. The usual end to connect the shield is at the driven end because that's usually the quieter ground. Thus, they put an arrow on the cable that points to where the load should be connected. You can plug them in either way and they'll sound exactly the same, though possibly noise will be a tad higher or lower one way or the other.
FWIW, I can't hear the difference between interconnects. Did a test some weeks ago on the matter here-
Geoff:
wanna let me scramble your cables and randomly reconnect or NOT? You than come into the room and listen. Tell me what is or isn't different.
We can do this 3 or 4 times, but 20 would be statistically valid. Let me know when I can drop by and we can begin testing your hypothesis.
Too much is never enough
Unfortunately your little experiment would prove absolutely nothing. A much better experiment is take an unshielded pair of interconnects and reverse their direction. Which direction sounds best? Or even easier for the uber skeptics among us just flip a stock fuse. It's not rocket science.
Edits: 01/31/16 01/31/16
I can hear directionality in interconnects VERY clearly. One direction relatively sucks. Same with fuses. Didn't you get the memo? All wires are directional, in the audible sense. Maybe you are "insensitive" to directionality. Lol
Darn, that memo was obviously sent through a cable in the "wrong" direction, so it never arrived here, or perhaps it arrived before it was sent and I missed it ;-)
That analogue computer reminds me of this:
Edits: 06/26/14 06/26/14 06/26/14
Wow, now that is cool!!
What a neat first job too, that must have been so cool (even at minimum wage haha).
My first job was taking furniture apart for an upholstery shop and the guy chewed tobacco which was everywhere Yucko.
I knew they had analogue computers but i never saw any of them beyond the simple adder circuits.
Do you get out to any audio trade shows?
Try to speak clearly and simply, are you suggesting people should not hear difference in this particular comparison?
Never trust an Atom, they Make Up everything!
What a -75 dB null means is the difference between the two was 1 part in about 30,000,000.
If one were to change the level of the signal you were listening to by that amount, it would be a change of about 1.4^-8 dB, or "pretty small".
If you're working up to say something then by all means spit it out.
Never trust an Atom, they Make Up everything!
Hi
I am not sure what you mean but I can explain without using dB which is a log scale and can be confusing.
As the dB scale is commonly used, maybe explain that would be better.
dB in this use refers to the difference between two things, a 3dB difference is a factor of two in power, 6dB is 4 times, 12dB is 16 times, 20dB is 100 times, 60dB is a million times.
So, if the distortion in an amplifier were -40dB down from the real signal, the THD would be about 1/10,000 the power or level of the desired signal or 1/100 the Voltage level.
-75 dB here means the difference between the two is about 1/30,000,000 the power of the desired signal.
Is that much change audible?
Now this is more complicated because ears are not microphones, what we experience as “audible reality” is FAR more than just the sound pressure reaching ones ears and includes what we see and know as well.
It is the lifetime of hearing that has allowed us to learn to use all the weird things our ears do as a function of angle and height, to create the 3d mental image we experience as reality.
That stuff is usually disconcerting to audiophiles as we know nothing else but that experience and yet it is also well known to audiologists that our ears aren’t “flat” in their frequency response at all and have all kinds of complications.
Hope that helps
I reported what I heard and what I measured. It is not my place to say what other people "should" or "should not" be able to do.
I doubt there would be agreement on what it means for someone to "hear" these differences. (It is necessary to agree on the size of the pin before debating how many angels will fit on it.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
If you ever grow a pair let me know, I might then try asking again.
Lol
Never trust an Atom, they Make Up everything!
I am not an easy troll, so I won't take your bait.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: