Are all Zeros created Equal

Are all Zeros created Equal

I consider myself to be an audiophile. It's a bit of a curse, actually. The never ending pusruit of getting better sound can be tiring. On the flip side, I spend a lot of time enjoying the hobby, and so I'd say it's worth it.

I read something the other day that I have not been able to let go.

It was on a software audio player forum. The software loads a piece of music into RAM on the PC, and then "plays" it from there. Playing it actually means sending it to an outboard DAC over USB using an Asynchronous USB approach, where teh DAC is the master clock.

The author made the claim that he tested the following 2 scenarios, and one was "better." (Of course, we could be dealing with a placebo effect here, but bear with me.)

In method A, he connects an HDD using SATA > USB> MOBO, loads the music into RAM, disconnects the drive, and plays the song. (Disconnecting the drive is done to minimize "noise" [ground loops, EMI/RFI, etc.])

In method B, he connects the HDD using SATA > USB > Optical (galvanic isolation) > USB > MOBO.
The claim is that B is better!

He thinks that "how the data gets into RAM" has an effect on the sound.

Could this be? Is the "noise" of the HDD connected in method A stored along with the bits in RAM? Are not all Zeros created equal? Am I missing something? And better yet... how can I test this?
More ?s:

When the computer sends the song to the DAC, does the CPU make any determiniations about the state of the Zeros and Ones? Is it "harder" for a chip to determine 0 or 1 if the signal is "less clean"?

So many questions!

Please help.

10 posts / 0 new
Last post
For more complete information about compiler optimizations, see our Optimization Notice.

"does the CPU make any determiniations about the state of the Zeros and Ones? Is it "harder" for a chip to determine 0 or 1 if the signal is "less clean"?"

Could you explain that part again? Sorry, its just I that I tend to get confused easily 
 

The 0's and 1' will not get confused. Ground loop issues, as well as noise on the AC power lines may be more related to the PC's switching power supply than anything else. If you have an oscilloscope you can check for noise. What is powering your  MOBO?

Can you hear a difference?

Jim Dempsey

www.quickthreadprogramming.com

Hi kdubious,

You can read about the Reed-Solomon error correction codes: https://en.wikipedia.org/wiki/Reed%E2%80%93Solomon_error_correction

It is not 100% related to your question but it will provide some answers to your questions because it will allow you to understand how these error correction algorithms are applied.

 

Gastón C. Hillar

Hi kdubious,

Another important reading is about 8B/10B encoding, the one that Serial ATA uses in its many versions: https://en.wikipedia.org/wiki/8b/10b_encoding

 

 

Gastón C. Hillar

>>...He thinks that "how the data gets into RAM" has an effect on the sound...

I think something else happens ( at a hardware level / possibly at a driver ) and that is why his two tests have different results.

Sergey,

I hadn't read that part... A RAM failure would cause this issue. :)

 

Gastón C. Hillar

>>...I hadn't read that part... A RAM failure would cause this issue. :)

...an Access Violation exception would be thrown and then a media application crashed...

Sergey,

Of course, I was joking. :) Everything would have crashed! It seems that this forum includes many questions that do not make sense. I guess I won't pay more attention to the Watercooler / Catchall thread. :)

Gastón C. Hillar

Thank you both for replying, albeit 5 years later. There's an entire industry of high end audio, much of which has moved into the computer audio space. One takes a computer, with tweaks, and connect it to an external Digital to Analog Converter (DAC), often over USB. Claims are made that various aspects of the computer feeding the signal have a noticeable, sometimes large, impact on the resulting sound output. Things like turning off services, tweaking RAM settings in the BIOS, or powering the computer with a linear power supply. It could be purely placebo, sure. But in an attempt to be scientific, I'm wondering how things at the hardware level might deal with slight noise/distortion/timing issues and whatnot... and how or if that might impact the actual 0's and 1's sent over the USB line.

 

 

 

 

Leave a Comment

Please sign in to add a comment. Not a member? Join today