Investigation Into Effects Of PC load On DAC Analogue Output

[March Audio]

BTW @manueljenkin how are you getting along in your Juniabs thread?  We are awaiting the explanation as to how that works. 😉

[manueljenkin]

1 hour ago, March Audio said:

Your hearing works spatially by assessing timing, amplitude and frequency response.  Responses that are modified by your head, torso and ear shape.

This is only an extremely small subset of the whole process (not to mention your vague use of "timing" and "amplitude", without mentioning what particular properties/structure of that is relating to spatial perception). You can't insert your random assertions everywhere and get away with it. That is neither objectivity nor science.

 

1 hour ago, March Audio said:

OK Manuel, would you like to tell us all about your knowledge of the subject?  Tell us how it works?

Not falling for this bait as it'll take the thread off topic.

 

1 hour ago, March Audio said:

How do you think that things like Q sound Work?  They manipulate these factors.

I have no reason to predict how a particular tool goes about doing its job, or how accurate it is on this aspect, since the whole domain is constantly evolving and better solutions from better models are often released.

[March Audio]

1 hour ago, manueljenkin said:

This is only an extremely small subset of the whole process (not to mention your vague use of "timing" and "amplitude", without mentioning what particular properties/structure of that is relating to spatial perception). You can't insert your random assertions everywhere and get away with it. That is neither objectivity nor science.

 

Not falling for this bait as it'll take the thread off topic.

 

I have no reason to predict how a particular tool goes about doing its job, or how accurate it is on this aspect, since the whole domain is constantly evolving and better solutions from better models are often released.

Really?  Then please go ahead and explain what these other major factors are.

 

There is nothing random or vague about what I have said.  Its quite the contrary, it is yourself that is coming up with random hand waving assertions.

[One and a half]

7 hours ago, March Audio said:

Must to add, for soundstage to change the frequency response, phase (timing), or amplitude would need to change.  The tests look at all of those things.

Many a subjective experiment concentrated on reducing common mode noise has made improvements in soundstage. The worst scenario is a laptop connected directly to a DAC via USB (with a printer cable). Common mode noise can ruin an image very easily.

[opus101]

40 minutes ago, March Audio said:

Really?  Then please go ahead and explain what these other major factors are.

 

One of the major factors you omitted was depth perception. Could that be because you dismiss soundstage depth as 'an artificial construct' ? In truth its no more artificial than any other audible illusion portrayed by a stereo system.

[March Audio]

6 minutes ago, One and a half said:

Many a subjective experiment concentrated on reducing common mode noise has made improvements in soundstage. The worst scenario is a laptop connected directly to a DAC via USB (with a printer cable). Common mode noise can ruin an image very easily.

Really?  So post the evidence.  Saying "I hear it therefore it is" has no credibility in this objective thread.

[March Audio]

7 minutes ago, opus101 said:

 

One of the major factors you omitted was depth perception. Could that be because you dismiss soundstage depth as 'an artificial construct' ? In truth its no more artificial than any other audible illusion portrayed by a stereo system.

Why do think its ommitted.  Why do you think 'spatial perception" doesnt cover it?

[opus101]

Are you saying 'spatial perception' does cover it? To me 'spatial perception' would most certainly include L/R imaging - that's perceived by a completely different method than depth. So you want to lump them together?

[March Audio]

7 minutes ago, opus101 said:

Are you saying 'spatial perception' does cover it? To me 'spatial perception' would most certainly include L/R imaging - that's perceived by a completely different method than depth. So you want to lump them together?

Eh? Since when has spatial perception only been limited to 2 dimensions? 

[March Audio]

So back to measurements and the Gustard X16 we started on.

 

Below are the jitter measurements.

 

 

PC unloaded.  We see a smattering of 50Hz spaced components some of which are likely to be from the linear power supply in the unit.  Otherwise some widening of the tone base is seen indicating some close in phase noise.  Some of this will be due to the fact that the clocks in the RME and Gustard arent locked together.  They both will have some noise so the result will inevitibly be slightly worse than the RME in loopback.

 

PC unloaded

 

PC loaded - no significant difference

 

 

 

here are the time domain measurements from deltawave.

 

PC Unloaded.  Like the RME loop back random noise at the limits of the DAC/ ADC performance. -120.2 dB

 

 

 

 

 

PC loaded - no significant difference -120.17dB

 

 

 

 

Next testing the Gustards RCA output.

[March Audio]

Oh just to add that we are measuring down to ridiculously low levels and what we see there on the gustard plots is in no way audible and should not be taken as any issue with the dac.

[March Audio]

I have just noticed that the jitter plotsxabove are round the wrong way.  Makes no difference to the comments.

[March Audio]

Also have to mention that I have moderated a number of off topic and circular argumentative posts from some that don't appear to understand that threads in the objective area require more than unsubstantiated opinion.

[idiot_savant]

So I suppose something I would say is that (again) we have a fair amount of hand-waving from some quarters. If you are unhappy with FFT analysis, and have a better way of objectively proving something, please feel free to put it forward. I'm not going to say @March Audio will peform any test you throw at him but just saying "I don't like the way you're testing" isn't going to get us anywhere.

 

 

He's put some thought and effort into this, which should be appreciated - somewhat more than waving his hands in the air and making "what-if" statements

 

 

your friendly neighbourhood idiot

[jabbr]

15 hours ago, One and a half said:

The images of the waveforms don't tell the whole story. A PC is a common mode generator par excellence. There are so many smps and pwm controllers on the motherboard, that anything connected will receive a fair dose of noise, this includes USB, AES3 cables.

The images don't show how the soundstage is set, how wide or narrow the instruments are placed in music. From my humble experiments with the DX ISO filters on ethernet transmissions, there's more up the upper end of the conducted EMC spectrum that causes soundstage errors in DACs. Yes these are heard and are subjective. By removal of upper audio band noise does the audio soundstage improve.

 

Here's a paper on common mode noise on AES3 transmissions for interest.

 

 

AES2003.pdf 834.39 kB · 4 downloads

 

Moreover non-linearities are caused by system responses to complex inputs that are not the mere sum of sinewave sweeps. 

[jabbr]

On 6/14/2021 at 6:53 PM, idiot_savant said:

Erm,

 

sorry to be pedantic, but this thread is about PC load? Can that change common mode noise? If so, surely we can measure it? As for ringing, it’s not showing up on the DAC output? The thread is about common conceptions  - if the DAC used in the test is “too good”, can someone suggest one more likely to show differences?

 

 

 

 

 

your friendly neighbourhood idiot 

 

PC load most certainly can and has been shown to affect common mode noise. There is a measurement of this in one of my classic textbooks on high speed digital design/RF mitigation etc. I can't recall which off hand. I'm going to guess Ott but if not then Johnson.

 

In my experience the Pre Box Digital S2 which is an outstanding though very cost effective DAC is more input sensitive than the iFi iMicro DSD to keep this discussion to commonly available and relatively inexpensive DACs that I like the sound of.

[semente]

Is this of any relevance?

 

Perception of mid frequency and high frequency intermodulation distortion in loudspeakers, and its relationship to high-definition audio.

 

 

 

Griesinger_IMD-perception.ppt

[March Audio]

8 hours ago, jabbr said:

 

Moreover non-linearities are caused by system responses to complex inputs that are not the mere sum of sinewave sweeps. 

Again please read the thread.  You missed the complex multitude test which also showed no differences when the PC was loaded.

[March Audio]

8 hours ago, jabbr said:

PC load most certainly can and has been shown to affect common mode noise. There is a measurement of this in one of my classic textbooks on high speed digital design/RF mitigation etc. I can't recall which off hand. I'm going to guess Ott but if not then Johnson.

 

In my experience the Pre Box Digital S2 which is an outstanding though very cost effective DAC is more input sensitive than the iFi iMicro DSD to keep this discussion to commonly available and relatively inexpensive DACs that I like the sound of.

Again please show the evidence.

 

How did you measure the effect in the dac output?

[March Audio]

2 hours ago, semente said:

Is this of any relevance?

 

Perception of mid frequency and high frequency intermodulation distortion in loudspeakers, and its relationship to high-definition audio.

 

 

 

Griesinger_IMD-perception.ppt 1.16 MB · 3 downloads

Not really.  Without discussing the content of the paper, uktrasonic intermodulation products need to appear in band, ie below 20kHz to be audible.  So far We have seen nothing even vaguely audible change between the PC unloaded and pc loaded states.

[March Audio]

So back to some more measurements of the Gustard X16, this time from the single ended RCA output.

 

This is with the PC unloaded.  This time, compared to the balanced XLR output we see spuria at 50Hz and harmonics at 100Hz, 150Hz etc.  Several reasons behind this; using the RCA output has reduced the CMR of the ADC input, so those mains spuria will have less rejection.  Also, noise currents that in XLR connections will be (mostly) harmlessly carried in the separate shield wire,  can end up in the RCA signal low (shield).  They can range from classic ground loops to leakage currents from SMPS to general noise from the PC.

 

In this case we can see that the spuria are still only at extremely low and inaudible levels.

 

 

 

PC Loaded - no significant difference.

 

 

 

Jitter - PC unloaded.  Finally we have something interesting to look at.  We see the same 50Hz sidebands as with the XLR connection but also some addtiional spuria at 11kHz and 13kHz.

 

 

lets widen the view. Its spread throughout the ranhge.  Note higher spikes at 8kHz and 16 kHz which coincide with USB packet noise.

 

 

With PC loaded - No significant difference

 

 

So the RCA connection is showing some very minor issues - we are down below -150dB.

 

Lets put a galvanic isolator in the Gustard USB connection.  The major spuria have gone.  However we still see some 50Hz contamination.

 

 

 

Next I will look at another RCA single ended output DAC.

[March Audio]

Been a few days but here is another RCA output single ended DAC (I will let you guess what it is - cough) on the same PC.  It is USB powered.

 

PC unloaded.  We see a tiny amont of 50Hz mains pick up and note the 8kHz and 16kHz spuria.  These are almost certainly not the DACs harmonic distortion, but USB packet noise.

 

 

PC loaded.  No significant difference

 

 

 

Jitter PC unloaded.  We see some very low level 50Hz sidebands.  Note the DAC board is not in its case so may be worse than normal.

 

 

In a wider view we can see 8kHz and 16kHz spuria. Again almost certainly USB packet noise.

 

 

PC loaded.  No significant difference.

 

 

 

 

 

The next plots are with a galvanic isolator in circuit.  PC Loaded.

 

No surprises here, the USB packet noise has gone.

 

 

 

I will try and get time to perform the Deltawave time domain tests later but from what we have seen so far this PC causes no difference in DAC output whether unloaded or loaded.  The minor issues that are seen are present in both circumstances and are caused by noise currents going from PC to the DAC.  They are resolved by a galvanic isolator.

[Allan F]

On 6/16/2021 at 3:50 AM, March Audio said:

Also have to mention that I have moderated a number of off topic and circular argumentative posts from some that don't appear to understand that threads in the objective area require more than unsubstantiated opinion.

 

Not if they are replying to a thread posted in the objective forum by an individual citing his experience as "proof" that confirmation or expectation bias explains "differences" heard by individuals who "trust their ears".

[March Audio]

3 hours ago, Allan F said:

 

Not if they are replying to a thread posted in the objective forum by an individual citing his experience as "proof" that confirmation or expectation bias explains "differences" heard by individuals who "trust their ears".

Who is doing that?

 

This is simply an objective investigation into whether any effects can be seen in a dac output when the Source PC is loaded compared to unloaded.

 

You may want to extrapolate that to mean other things, but that's not what's happening here.

 

However if you want to look at the very real effects of bias there are a couple of other threads that cover this subject.

[March Audio]

I have a couple of laptops and another NUC pc to try these tests on  but it will take a bit of time to run through them.