Computer Audio: Is isolation as good as optimization?

Jul 1, 2010
8,713
0
0
#1
There is a pretty lively discussion going on at a computer audio forum I frequent. A gentleman there who sells, among other things, computers optimized as audio sources, firmly believes that the only way to remove the noise from a computer music source is to hunt it down everywhere it occurs and get rid of it: minimize OS operations, hard disk drive, linear power supply, etc. He upgrades computers for this purpose, so he has a stake in it, but I believe he is sincere. The thing is, I've gotten great results from simple isolation. Well, not really simple. I send USB out of my MacBook Pro to a Trends UD 10.1. The Trends galvanically isolates, re-clocks then puts out coax, toslink, AES/EBU and BNC. I believe it is working for two reasons: 1) It sounds great; I'm getting a super clean signal to my active speakers and my headphone system. 2) it makes sense. By isolating and re-clocking outside of the MacBook, I have left behind all the noise, interference and jitter that is in my Mac. Right? Or am I missing something?

Tim
 
Apr 3, 2010
16,022
0
0
Seattle, WA
#4
Unless you can modify the operating system and have equipment to confirm the effects of such modifications, the person is shooting in the dark. Here is an example. Let's say you a have a thousand things running in your PC. You optimize it now and it is only two things. Is the latter better? Not necessarily. The many things being different may regenerate a random signature whereas the other, predictable sources of timing jitter. This is the opposite of what someone thinks may be the case.

While we can argue about the audible effects of such tweaks, the measurements should give us useful data. I realize the equipment is expensive but at some point, there needs to be a data point for us to use as the starting point. Wonder if Stereophile would be open to measuring a few of these systems.
 

Vincent Kars

WBF Technical Expert: Computer Audio
Jul 1, 2010
860
0
0
#7
Less is more.

This is a very strong sentiment in the audiophile world.
If an audible difference between WAV and FLAC is reported, invariably WAV is preferred.
It is umcompressed so it uses less CPU so it is better.
(That you get about double the I/O is safely ignored)

Likewise a lot of PC tweaks, if it is less it is better.
So underclocking is a must.
What does it do?
The frequency lowers and the density of the harmonics increases.
Why would audio improve?
There is a very nice post by John Swenson about this subject: http://www.audioasylum.com/forums/pcaudio/messages/5/56899.html

Shooting in the dark indeed.

I must admit the concept of isolating the DAC (galvanic and a asyn protocol) appeals to me.
But as usual in engineering there is no free lunch.

Say we have a USB DAC using async protocol and galvanic isolated from the PC using opto-couplers. Sounds great.
If we phrase it slightly different: a USB DAC with an async USB receiver converting USB to Toslink.
Does it still sound great?
 

FrantzM

Member Sponsor & WBF Founding Member
Apr 20, 2010
6,469
0
0
#8
Hi

Must one use USB at all? It is practical, true, but it seems t be a noisy interface. In the case of a MacPowerbook isn't SPDIF available from the get go?
 

Vincent Kars

WBF Technical Expert: Computer Audio
Jul 1, 2010
860
0
0
#9
No, it is an option (but a lot of PCs don't have a Toslink out)
But if we talk isolation than:
Yes, the Toslink shields the DAC from the electrical noise going on inside the PC
No, the clock driving the SPDIF is not shielded from the electrical noise at all.
 
Jul 1, 2010
8,713
0
0
#10
Couple of points: can you shut down the internal DAC of the Trends while passing the digital through to your actives; and have you tried driving it on battery power, as is possible?

Frank
It is isolated, anyway, but I don't believe it turns on until you plug something into the analog output.

Tim
 
Jul 1, 2010
8,713
0
0
#11
This is a very strong sentiment in the audiophile world.
If an audible difference between WAV and FLAC is reported, invariably WAV is preferred.
It is umcompressed so it uses less CPU so it is better.
(That you get about double the I/O is safely ignored)

Likewise a lot of PC tweaks, if it is less it is better.
So underclocking is a must.
What does it do?
The frequency lowers and the density of the harmonics increases.
Why would audio improve?
There is a very nice post by John Swenson about this subject: http://www.audioasylum.com/forums/pcaudio/messages/5/56899.html

Shooting in the dark indeed.

I must admit the concept of isolating the DAC (galvanic and a asyn protocol) appeals to me.
But as usual in engineering there is no free lunch.

Say we have a USB DAC using async protocol and galvanic isolated from the PC using opto-couplers. Sounds great.
If we phrase it slightly different: a USB DAC with an async USB receiver converting USB to Toslink.
Does it still sound great?
Depends on how well the receiver at the other end handles toslink, I suppose. Toslink has inherently higher jitter...which just gets us into another great computer audio debate -- when is jitter audible? -- but implementation is everything, they tell me, and toslink can be great.

The less is more ethos has always been interesting, because it is so inconsistent. If less is really more, would the typical audiphile system be a separate component with its repititive power supply, cables, internal wiring, resistors, capacitors, etc, etc. Or would the high end have embraced integration long ago and worked to minimize all of the above in simplified, integrated designs?

Not, evidently, if there is more money in selling boxes and wires...

Tim
 
#12
A good thread, Tim.

> Computer Audio: Is isolation as good as optimization?

If you want to make some informed choices for gear and software and use that system to listen to music for several years, then establishing good isolation in the path to the DAC (or in the DAC itself) that lets you use a standard computer is a smart choice.

If you want to hear differences, make tweaks and hear more differences, then messing around to create a minimal hardware and OS configuration opens new possibilities. It is a lot cheaper than buying and selling gear like interconnect cables to provide subject matter for hearing differences.

Bill
 
Jan 18, 2012
547
0
0
Drobak Norway
#15
hello first of all
why go via toslink?
there´s async usb to I2S interfases out there like e.g. EXA devices in Canada
I run this into SABRE based Buffalo DIY dac with excellent results
lots of enthusiasts in Norway run this excellent solution and the sonic result is staggering on my AG Trio based all out approach
highly recommended
best
Leif
 
May 30, 2010
13,967
42
48
Portugal
#16
Galvanic isolation usually only means that there are no resistive paths between both parts - you can get it using transformers or optical coupling. Transformers break the resistive path, but are susceptible to capacitive coupling, that can be a Trojan in such systems. Also, logic signals per se can introduce noise in the receivers - if the 0 and 1 DC levels are not noise free, noise can enter in the system. In order to avoid these noise signals power supplies must be noise free and PCBs must be properly designed . Just because you see some toroidal couplers in the PCB does not mean the audio section is a separate universe.
 
Apr 3, 2010
16,022
0
0
Seattle, WA
#18
Optical is certainly a viable method to get isolation. While on stand-alone systems S/PDIF usually outperforms Optical, it is not always the case for PCs due to much increased potential for noise from the PC.

On my current system, Toslink underperforms async USB. So I don't use it but as me, you definitely want to test it to know.
 
Jul 1, 2010
8,713
0
0
#19
How about wireless? An obvious advantage is isolation. What are the drawbacks?

Tim
 

Vincent Kars

WBF Technical Expert: Computer Audio
Jul 1, 2010
860
0
0
#20
You might use Wi-Fi or Ethernet (also galvanic isolated) but you need a processor to do all the networking.
Might be a small one (SqueezeTouch) but it is a kind of mini PC.
So the question arise how to isolate the DAC from this computer!
 

About us

  • Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing