[WindowsX]

I found testing paper about measuring changes with power cord here.
 
http://www.johnvestman.com/VDCableNoiseSlewRevised2007.pdf
 
Here's front page about this test.

 

Testing the Virtual Dynamics Power One A/C Power Cable
 
Supervisor: Jeff Archbold, M.A.Sc., P.Eng.
University of Toronto Engineering Students:
Visnu Siritharan
Diana Mollicone
Elnaz Ghasemi
Fatima Asad
 
November 19, 2007
Walters Forensic Engineering Inc., in association with the University of Toronto

 
Please read and let us know what you guys think about this test.

 

[spruce music]

Well I would call it amateurish, but the amateurs I know would be rightly offended.
 
We see the common issues of a graph without knowing what it is a graph of.  FFT, actual sample points, what is it?  Combined with no values on the axis of the graph. 
 
The very small differences they obtained may or may not be for real.  If they just did this once those differences are probably small enough you could run the tests again getting different results.  If they are results run several times each with a consistently lower noise floor using the cable under test it probably means a slightly lower noise floor. 
 
What is a 0 khz signal, silence?  Maybe a language issue if this was translated from French.  It appears for peak levels they are using sample values. That alone could cause the differences they saw if the test is run only once.  What I mean by that is you look for the single highest sample value and call it your peak.  With 1 khz signals and clock drift between two devices, the same exact output signal identical in every possible manner might record different peak sample values with different drifting of two clocks.  The two clocks being the clock in the DVD player and the one in the Maudio USB device.  That might explain why they said they couldn't use the 10 khz signal.  With so few samples the result might make no sense to them viewed this way. 
 
The conclusion that they obtained a 2.6% greater slew rate and 1.55 db quieter noise floor even if real hardly seem extraordinary. 
 
Plus this is the defunct company that was found to be using a very inexpensive cable wrapped in a fancy outer layer and selling for a premium of a few hundred percent.  Which eventually lead them to go out of business. 
 
http://www.xtremesystems.org/forums/showthread.php?181772-Virtual-Dynamics-Power-3-%E2%80%93-2000-Profit-Margin
 
http://www.head-fi.org/t/293165/my-cat-tore-up-my-virtual-dynamics-power-3
 
I will save the embarrassment of the various comments here on head-fi about the greatness of these cables on various threads before the truth was outed by a miscreant feline.  I recall one thread where the cable lowered noise so much they could now easily hear if they rested their gear on sorbothane pucks or simply the floor.  
 
Or you could go to youtube and search for virtual dynamics Rick Schultz where you will see videos highlighting the same miniscule things in this paper.  Lower noise floor (by tiny amounts lower) and increased slew rates again by tiny amounts faster.  Makes me think the paper you linked was commissioned by Mr. Schultz. 
 
Its okay if you are big fan though.  Rick came back with High Fidelity Cables.  I wonder if he made the new covering cat resistant this time?
http://www.hifizine.com/2012/06/rick-schultz-interview/

 

[GuyUnder]

Huh? 1.5 dB isn't minuscule, and obviously these kinds of tests don't tell the whole story. These kinds of cables rely on resonance control -- they don't have have any special technology, no unique geometry or advanced materials. They are just standard cables filled with a dampening material. You could make one yourself with a glue gun and some tubing.

 

[spruce music]

Well true believers gonna believe.
 
The thread on head-fi I linked to where the cat did a feline tear down had 772 posts with many still defending the cable.  Hard to believe.....or well.....I guess not if you are a believer.
 
Is the 1.55 db the difference in peak sample values between two tests?  We don't know as they didn't specify enough information.  You can get that level of difference between any two runs of silence testing even with the same exact configuration. 

 

[castleofargh]

I'm confused is that supposed to be a paper based on only 2 sets of cables ?

 

[U-3C]

Huh? 1.5 dB isn't minuscule, and obviously these kinds of tests don't tell the whole story. These kinds of cables rely on resonance control -- they don't have have any special technology, no unique geometry or advanced materials. They are just standard cables filled with a dampening material. You could make one yourself with a glue gun and some tubing.

I can make cables with special technology, unique geometry or advanced materials with a glue gun and some tubing too... Nobody will doubt me when I give the cables to them to listen...

 

[watchnerd]

 
Please read and let us know what you guys think about this test.

 
It's a student paper.
 
It has about the quality I would expect of a student paper, namely, sloppy:
 
-No scales on graphs
-Didn't take multiple samples over multiple runs, apparently
-No statistical analysis of multiple results (because they only did 1?)
-Weird methodology, using a DVD player with DVD-R + ADC, which replicates a scenario almost nobody will do in real life
 
If they were my students, I'd give them a C+ grade.

 

[WindowsX]

It's funny that I find this test having more tangible information than archimago's tests. I can only see rough graphs without details for each measurement. I also visited main domain today.
 
http://www.johnvestman.com/
 
It was hosted in audio studio mastering company website. Maybe it has something more than a student's project.
 
Regards,
Keetakawee

 

[spruce music]

  It's funny that I find this test having more tangible information than archimago's tests. I can only see rough graphs without details for each measurement. I also visited main domain today.
 
http://www.johnvestman.com/
 
It was hosted in audio studio mastering company website. Maybe it has something more than a student's project.
 
Regards,
Keetakawee

I hate to continually be negative and disagree with so much you post, but how can you read that paper and find it has more tangible information than archimago's tests?
 
There literally is very little info in the paper.  As you say rough graphs without details.  No real explanation, nothing close to comprehensive for even one measurement made. Too much supposition without any support.  While archimago has quite a few tests with clear complete explanations of methodology and gear used.  Graphs in abundance always clearly marked.  If something looks odd archimago will try different gear or a different approach to probe more fully whatever the topic of the test is.  The paper you linked looked one and sloppily done.  I don't understand how you think this paper offers very much. 

 

[bfreedma]

  I found testing paper about measuring changes with power cord here.
 
http://www.johnvestman.com/VDCableNoiseSlewRevised2007.pdf
 
Here's front page about this test.

 
 
Please read and let us know what you guys think about this test.

 
I'd be more interested in knowing, specifically, what you think this about this test and what it means.

 

[WindowsX]

Comparing to archimago's tests, I find this test having more tangible information for evaluation. I understand archimago did tests in deeper levels but most of the time he showed only RMAA and graphs without table to see individual data clearly enough. Let's see his recent Rasberry Pi 3 here.
 
http://archimago.blogspot.ca/2017/01/measurements-raspberry-pi-3-as-usb.html
 
Like all others, data from devices are plotted on top of each other in graph with his usual narrative lines saying "No significant difference" in here and there without measurable information like X db difference or Y percentage. But that's not all. He also added audiophile recommendation like this.
 
Lower clock speed --> less power use --> less CPU noise --> better sound!
Better aligned clock speeds --> less random & periodic jitter --> better sound!
Less power use --> less strain on power supply --> less noise & jitter (cuz it's like that) --> better sound!
 
Better sound and here there even though it's supposed to be bits are bits making no difference in measurements. He also said confirmation line like this.
 
But more importantly, for audio streaming, it's still just as responsive (while sounding better of course!).
 
Yes, optimizing Linux OS makes audio streaming sounding better now too. Would be great if he could have said the same to any Windows apps. It looks like he can comment this will sound better without having to put any scientific measurement to backup his claim like everyone else too.
 
I do agree that optimizing Linux OS can affect audio performance and sound quality but I'd like to see his measurements to explain why it sounds better too. Will you guys be OK with "Lower clock speed, better aligned clock speed, less power usage and such" as measurable information to justify better sound?
 
Back to original test paper, I only find his measurements easier to read with table telling individual information without zooming in graph and try to guess the numbers myself. I'd like to see archimago explaining how CRAAP config makes better sound in scientific ways too.
 
Regards,
Keetakawee

 

[watchnerd]

 
He also added audiophile recommendation like this.
 
Lower clock speed --> less power use --> less CPU noise --> better sound!
Better aligned clock speeds --> less random & periodic jitter --> better sound!
Less power use --> less strain on power supply --> less noise & jitter (cuz it's like that) --> better sound!
 

 
You are completely mischaracterizing what he wrote.
 
He said:
 
"Therefore, based on Convoluted Rationalizations And Audiophile Perceptions (CRAAP)™, some out there might say:"
 
He did not say it was necessarily true, or that he recommended it.  He is saying others may say so.
 
See the joke acronym CRAAP?  

 

[bfreedma]

  Comparing to archimago's tests, I find this test having more tangible information for evaluation. I understand archimago did tests in deeper levels but most of the time he showed only RMAA and graphs without table to see individual data clearly enough. Let's see his recent Rasberry Pi 3 here.
 
http://archimago.blogspot.ca/2017/01/measurements-raspberry-pi-3-as-usb.html
 
Like all others, data from devices are plotted on top of each other in graph with his usual narrative lines saying "No significant difference" in here and there without measurable information like X db difference or Y percentage. But that's not all. He also added audiophile recommendation like this.
 
Lower clock speed --> less power use --> less CPU noise --> better sound!
Better aligned clock speeds --> less random & periodic jitter --> better sound!
Less power use --> less strain on power supply --> less noise & jitter (cuz it's like that) --> better sound!
 
Better sound and here there even though it's supposed to be bits are bits making no difference in measurements. He also said confirmation line like this.
 
But more importantly, for audio streaming, it's still just as responsive (while sounding better of course!).
 
Yes, optimizing Linux OS makes audio streaming sounding better now too. Would be great if he could have said the same to any Windows apps. It looks like he can comment this will sound better without having to put any scientific measurement to backup his claim like everyone else too.
 
I do agree that optimizing Linux OS can affect audio performance and sound quality but I'd like to see his measurements to explain why it sounds better too. Will you guys be OK with "Lower clock speed, better aligned clock speed, less power usage and such" as measurable information to justify better sound?
 
Back to original test paper, I only find his measurements easier to read with table telling individual information without zooming in graph and try to guess the numbers myself. I'd like to see archimago explaining how CRAAP config makes better sound in scientific ways too.
 
Regards,
Keetakawee

 
Appreciate your response.
 
Your assumptions are that the changes, even if they are real, are audible.  Without scale on the graphs, I don't know how you can determine audibility.
 
Based on the details available in the paper and making some assumptions that their testing model was legitimate, I don't see audible improvement.  There really aren't enough data points in the paper you reference to make any actual conclusion, but that's my takeaway based on what is available.
 
IMO, archimago has far more data and while not perfect, is significantly more comprehensive.

 

[WindowsX]

   
You are completely mischaracterizing what he wrote.
 
He said:
 
"Therefore, based on Convoluted Rationalizations And Audiophile Perceptions (CRAAP)™, some out there might say:"
 
He did not say it was necessarily true, or that he recommended it.  He is saying others may say so.
 
See the joke acronym CRAAP?  

 
I understand those comments aren't directly from him but the fact he put it on his website means he at least agreed to put it on himself.

 

[Speedskater]

Why go any farther than the abstract? Almost every point is incorrect, electricity just doesn't work that way.