Anybody measured jitter from a computer?

[Keith_W]

As some on ASR would know, I use a Windows 11 PC which does everything - stream and store media, perform convolution, and outputs digital via Ravenna to an 8 channel DAC. I recently had a discussion with a subjectivist who claims that computers are noisy environments and that they do not sound as good as a dedicated streamer or high end CD transport. He said the only way to get a computer to sound good is (insert usual audiophile nonsense) ... linear PSU, audiophile USB cards, reclocking the ethernet output, etc.

I then realized that I have never seen any measurements of output from a PC and whether "noise" and "jitter" are indeed lowered by all these audiophile tweaks (although I would highly suspect not) and whether bit perfect playback from a PC is the same as any other digital transport. As a good scientist, I don't think I can categorically say that it won't make a difference until I see measurements that say there is no difference. So: is anybody aware of any measurements of the output from a PC?

 

[Blumlein 88]

A good start is here>

Review and Measurements of Intona USB Isolator for Audio DACs

This is a review and measurement of Intona USB Isolator. In the last few years there has been a proliferation of USB filters, cleaners, regenerators, feeding on fear of audiophiles of noisy computer ports. Intona doesn't actually play in that market directly. They built this device for...

www.audiosciencereview.com

Intona USB 3.0 Isolator Review

This is a review and detailed measurements of the Intona USB 3.0 (compatible with 2.0) Isolator. It is on kind loan from a member. It costs US $360. If you have been around here you know of a number of "audio" companies selling USB filters, cleaners, regenerators, etc. I have shown all of...

www.audiosciencereview.com

Jitter from an asynchronous USB connection leaves all timing to the DAC. It will use a free running crystal next to the DAC chip. This is the best way as you have a free resonating crystal at its own frequency not having to track or match another clock. What you can see sometimes is the basic 8 khz switching bleeding thru. There are a whole world of audiophile nonsense about how this is dirty. Funny how when asynch USB came out it was praised for its low noise and very low jitter. And sure enough over a few years when it is the norm someone says the old way was better which is BS. Records are better than CDs, and then CDs are better than audio over USB. Blah, blah, blah......

Read the above reviews by Amir and see what difference it makes having a USB cleaner vs not. All the measures are so low it cannot matter which does not show for certain that a streamer might in theory have a chance to be better. What it does show in fact is there is so little damage over USB it could never be heard if it were better.

 

[Blumlein 88]

Then look at the results from some CD players.

Emotiva ERC-4 CD Player (DAC) Review

This is a review and detailed measurements of the DAC portion of the Emotiva ERC-4 CD player. It is on kind loan from a member and costs US $649. The display lights up in classic, intense dark blue colors of Emotiva. I am not a fan of it as it is old fashioned. Not much functionality is here...

www.audiosciencereview.com

Marantz SA-10 Review (SACD Player & DAC)

This is a review and detailed measurements of the Marantz SACD player and USB balanced DAC. It is on kind loan from a member and costs US $7,499. The industrial design is good although usability is poor with super tiny fonts for various settings (e.g. filter number top left). Disc tray opens...

www.audiosciencereview.com

Tascam BD-MP1 Review (CD/Blu-ray Player)

Continuing our theme of testing "older" formats, here is a review and detailed measurements of the Tascam BD-MP1. It was purchased new and kindly sent to me by a member. The BD-MP1 costs US $499. Consistent with its marketing theme, this is a "professional" rack mounted optical player: I...

www.audiosciencereview.com

 

[charlielaub]

Audio inside a computer does not have "jitter". That term just does not apply to what is happening to the audio samples. Computers process audio in chunks and the audio is buffered multiple times. The audio subsystem controls the size of these buffers.

Audio sample rates are very, very, very slow compared to the clock of the computer. Also processing/moving of buffered samples happens much, much faster than the audio is being spooled out at the end of the playback chain. Once N samples have been received, from a file or other source, the computer groups them together. These audio samples may sit there doing nothing for a few msec while the computer waits for the appropriate time to send them to the sink (e.g. DAC) or to another process thread within the OS, etc. Meanwhile the computer has other stuff to do within the OS...

The only clock source that is controlling the playback timing of the samples is located in the DAC. An asynchonous DAC has its own clock, so that is the clock that is completely responsible for playback jitter. It has nothing to do with the computer. I believe that some older DACs (e.g. ES9023) that derive their clock from the USB bus clock do indeed get their playback rate from the computer, however, the mechanism for deriving the playback clock from the USB bus clock typically has relatively good jitter suppression.

Recently an entire thread developed over at DIYaudio in which it was claimed that tweaking "micro timing" and other aspects of the computer OS improved playback sound quality and reduced noise, e.g. insinuating that the OS processing of samples was degrading them somehow. It was complete rubbish from start to finish. I pointed out to the OP author that the OS has nothing to do with playback timing or noise and he replied that disbelievers like me didn't need to post - he was just looking for like minded people for a circle jerk. He could not and did not provide any measurement or objective evidence that the result of his tweaks improved anything. He only made lots of claims that is was "so obviously" sounding better, sounding more "analog", yadda yadda, under his sighted testing that was not even A-B. I am spending less and less time "over there" when these sorts of threads are the popular ones.

EDIT:
For an example of the mindset from those posting in the thread over at DIYaudio that I mentioned:

Over here at DIYA we usually talk about experiences and findings. That's not some theoretical hocus pocus.
People have been doing and trying it. And spent long hours and quite some money on it. A few people
even share some of the work with the community. Most problems over here at DIY-A start when
certain people start hijacking great opportunities by asking for explanations and measurements.
These people know very well that nobody can and would be willing to afford a scientific lab to
prove the point.
That's why it is so important to run the empirical approach. The more inmates confirm
a certain result or behaviour, the more you can rely on the validity of a subject.

 

[fpitas]

Blaming jitter is the audio equivalent of blaming the dog.

 

[Keith_W]

Thank you for that reply. I have a Merging DAC, and I send audio to it via Ravenna. Am I correct in saying that the computer is sending packets to the DAC, which stores it in memory then reclocks it for DA conversion? It should be completely immune to any jitter from the PC, or if I understand you correctly, there is no such thing as jitter from the PC since audio is not being streamed, it is being sent in packets?

 

[thegeton]

Blaming jitter is the audio equivalent of blaming the dog.

It's that open source? Am I free to use it?

Additionally, we should can your expression and use it as the first response to any new subjectivist jitter thread.

Just sayin'

 

[notsodeadlizard]

As some on ASR would know, I use a Windows 11 PC which does everything - stream and store media, perform convolution, and outputs digital via Ravenna to an 8 channel DAC. I recently had a discussion with a subjectivist who claims that computers are noisy environments and that they do not sound as good as a dedicated streamer or high end CD transport. He said the only way to get a computer to sound good is (insert usual audiophile nonsense) ... linear PSU, audiophile USB cards, reclocking the ethernet output, etc.

I then realized that I have never seen any measurements of output from a PC and whether "noise" and "jitter" are indeed lowered by all these audiophile tweaks (although I would highly suspect not) and whether bit perfect playback from a PC is the same as any other digital transport. As a good scientist, I don't think I can categorically say that it won't make a difference until I see measurements that say there is no difference. So: is anybody aware of any measurements of the output from a PC?

It has sense only if your PC is a bit special and has some kind of S/PDIF outputs (coax or TOSLINK typically).
But because S/PDIF also typically is implemented as USB to S/PDIF converter, there is no sense at all.
At receiving async USB side there are some kind of good DAC's clock generator with low jitter and internal process of resynchronization, so...

 

[charlielaub]

Thank you for that reply. I have a Merging DAC, and I send audio to it via Ravenna. Am I correct in saying that the computer is sending packets to the DAC, which stores it in memory then reclocks it for DA conversion? It should be completely immune to any jitter from the PC, or if I understand you correctly, there is no such thing as jitter from the PC since audio is not being streamed, it is being sent in packets?

Correct, except that data sent in packets is no longer "clocked" per se, so techinically speaking the DAC is not "reclocking" the data stream.

See:

What is Asynchronous USB? | Do You NEED It?

What the heck is "Asynchronous USB?" Is this yet another audio scam designed to trick people into buying sh** they don't need? All of these answers and more, comin' up!! Greetings bass head, Stuart Charles

homestudiobasics.com

Asynchronous USB Audio / Asynchronous Digital to Analog Converters - Hifime Audio

USB transfer modes has received a lot of interest and attention recently and “Asynchronous DAC’s” has been praised and hyped for it’s superior performance. However there seems to be a lot of confusion around asynchronous USB Audio. There is a good and explaining article at EDN network about USB...

hifimediy.com

Your DAC is very likely running its own clock, and so is an asynchronous USB device (does not share a clock with the computers' clock).

 

[voodooless]

Am I correct in saying that the computer is sending packets to the DAC, which stores it in memory then reclocks it for DA conversion?

Essentially yes, the DAC will request new samples in regular intervals.

The OS will still need to know the sample rate, so this is still derived from the rate the packets are requested. Things like the OS mixer might resample audio to that rate.

For exclusive audio access applications however, the software will send the audio samples bit perfect.

It should be completely immune to any jitter from the PC, or if I understand you correctly, there is no such thing as jitter from the PC since audio is not being streamed, it is being sent in packets?

Everything is packets… like turtles, it’s packets all the way down . There isn’t really a distinction between the two really. Network streaming audio of video is also just packets… even the I2S interface you could see as having packets.

 

[Sokel]

There are a couple of measurements here:

"Audiophile" motherboards

I'm talking about the ones who advertise "audiophile" grade constructions,with separate USB paths,etc. Truth or snake oil?And do we have any measurements?

www.audiosciencereview.com

 

[Phorize]

MEASUREMENTS: Intel i7 PC and Raspberry Pi 3 B+ Audio Streamer - XLR / RCA, Noise and Jitter. Do digital transports / streamers really make a difference? Do USB cables?

A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.

archimago.blogspot.com

 

[phofman]

It has sense only if your PC is a bit special and has some kind of S/PDIF outputs (coax or TOSLINK typically).

Onboard SPDIF outputs are virtually always outputs of the onboard IntelHDA codec which is a standard soundcard, though clocked by onboard HW PLL clock. Certainly no OS jitter or other SW voodoos apply

 

[potfur]

When I connect my PC to the SMSL C200 DAC using an optical Toslink cable, I have to set DPLL (digital phase-locked loop) on the DAC quite high (7 out of 9 for 44.1 kHz and 8 out of 9 for 192 kHz), otherwise the DAC cannot lock on to the signal, which is manifested by an intermittent stutter. This suggests that the clock stability of the optical signal output from the motherboard (which uses the ALC4082 chip) is not that great. When connected via USB, this issue is of course absent.

 

[phofman]

@potfur : Your internal soundcard is not a standard internal codec hooked via Intel HDA bus to the chipset. ALC408x is a USB audio device, with output running in USB isochronous adaptive mode. I.e. it uses PLL to recover clock from the internal USB bus. The recovered clock is used to generate the SPDIF stream, with the clock being recovered again in your DAC.

All USB details of ALC4080 are e.g. http://alsa-project.org/db/?f=22aed0d76705b354a59364e6e8e41a4fde20e4da