Investigation Into Effects Of PC load On DAC Analogue Output

[March Audio]

2 minutes ago, PeterSt said:

 

How would that work ?

Nobody is changing the time domain. So IMO this can only about jitter and DeltaWave does not deal with that, does it ?

I think it does deal with drift, but that is something else.

Seriously?  Time domain isn't just about jitter.  The point was that FFTs don't see quickly changing signals.  Hence the use of deltawave to make comparisons in the time domain.

 

Again mo significant difference was found between the PC at idle and full load.

[March Audio]

5 minutes ago, PeterSt said:

 

How would that work ?

Nobody is changing the time domain. So IMO this can only about jitter and DeltaWave does not deal with that, does it ?

I think it does deal with drift, but that is something else.

Are you under the impression that noise won't be seen in the time domain?

[March Audio]

Anyway more data from different dacs and PCs coming soon.

[March Audio]

30 minutes ago, PeterSt said:

 

Yes, seriously. I was suggesting that nothing in the time domain changes. So what would be changing in DeltaWave then ? (your idea, not mine).

Let's regard this a misunderstanding. I am fine with that.

Still I would not use something like DeltaWave, because I would not be looking for noise (I am counting the number of times saying this).

Never mind.

Then how on earth are you hearing changes in sound quality?

 

Seriously

[March Audio]

29 minutes ago, PeterSt said:

 

To be more clear: yes, we would see that. But no changes whatsoever in this regard.

Maybe recall that I can't see measurable differences myself, OK ? thanks.

And it is not about noise (still counting).

So what is it about?

 

BTW I was looking for differences between the frequency and tome domain of the unloaded and loaded PC. That could be noise, spurious signals or jitter.

 

 

[March Audio]

30 minutes ago, PeterSt said:

 

Well, the fact that your looped-back RME-RME is something I could do myself without any other variables than the PC.

Not that I'd expect a different result from your tests, but it would free you from things I just could do myself.

 

 

I'm not just doing looped back rme.

 

It sounds just fine.  Is this where you say it sounds crap, like ssds sound crap or any software that doesn't use tweaks 😉

[March Audio]

Just to say I have moderated some nonsense sense posts.

[March Audio]

1 hour ago, Summit said:

 

You mean censorship because I show how subjektiv all your claims are. 

 

Mo  you weren't making any sense and just trying to disrupt the thread

[March Audio]

54 minutes ago, PeterSt said:

 

I already told you - it should be about jitter. But

a. the differences I saw myself were waved away by you because the averaging was not ready (duh);

b. you were fast to quote Julian Dunn with his 20ns whatever jitter "so that can't be it".

 

Ad b.: Which now is your statement.

If you ask in a week of time the same question ("so what is it about") I will give the same answer. Over and over if needed.

I'm sorry Peter but you have just demonstrated how little you understand about the subject.

 

Jitter dies not get averaged away in FFTs.  It appears as sidebands, widening of the base of the tone or if it's really wide band a raising of the noise floor.

 

I demonstrated there was no increase in jitter.

[March Audio]

20 minutes ago, pkane2001 said:

 

 

Seriously, Peter. Jitter is measurable in time domain. In fact, DeltaWave computes it for each individual sample in the file and the overall RMS value in the time domain. And, to be clear, time domain contains all the same impurities, distortions, noise, jitter, whatever else you want to call it, that the frequency domain does. There is no difference. Whatever exists in one, exists in the other.

So Peter thinks nothing changes in the time domain and yet all this is about jitter - a time domain issue.

 

He also doesn't accept the scientific research on the audibility of jitter and levels required which are quite high by the standards achieved in dacs these days.

 

He doesn't accept the demonstration of jitter not changing with PC load.

 

And yet we are supposed to believe his "I hear it therefore it is" claims.

[March Audio]

4 hours ago, PeterSt said:

 

Maybe this is not fully OffTopic, as I realise that I only need to hook up my own and apply all the stuff I normally apply in order to see (hear) what happens. And, If I could describe SQ changes, would you be prepared to believe that (say initially) ?

Considerung the evidence to the contrary, unless you could demonstrate with a convincing blind test, frankly, no.

[March Audio]

7 minutes ago, semente said:

We can't find noise at the DAC's ouput.

Ideally we would be testing the D/A chip's performance with different software, OS, computers, input interfaces, etc. but it is easier to just measure the (whole) DAC.

 

Is there a way to measure noise just before the signal goes into the D/A chip?

You can look at the i2s lines but it's what comes out of the dac that is important.  If there are issues it will be seen in the dac output.

 

I will be trying several different PCs and laptops as sources.

[March Audio]

3 minutes ago, semente said:

 

I understand what you are trying to say but it has been claimed often that noise "travelling along" with the signal stream affects the D/A conversion.

Could there be an aspect of the D/A chip performance that is not shown in a "traditional" set of DAC output measurements?

What youvare really asking here is are there things we can't measure but can hear?

 

Simple answers no.  We can measure way below thresholds of hearing.  However you need to use the correct measurement to find what you are looking for.  Correlation can be difficult. 

[March Audio]

15 minutes ago, botrytis said:

 

Sorry, March Audio is looking at things objectively. Saying you can hear differences, is not an objective observation, it is subjective since we know nothing about the situation, etc.  Measurements are needed.

Sorry I made an error, apologies @semente.

 

I meant to say that I am removing summits posts where they are inaccurate and just trying to provoke argument

[March Audio]

1 minute ago, semente said:

 

If I remember correctly people were reporting poor sound quality from early solid state amplifiers and engineers said it wasn't possible. And were later proved wrong (Otala et al.).

Could a similar thing be happening here?

 

You provide a set of measurements that does not explain many reports of perceived audible differences.

Isn't there a chance that something else might need looking into / measuring?

 

I don't have a horse in this race, I am merely curious.

Actually you might be interested in the paper below which shows those conclusions about negative feedback were wrong.

 

https://www.google.com/url?sa=t&source=web&rct=j&url=https://linearaudio.net/sites/linearaudio.net/files/volume1bp.pdf&ved=2ahUKEwjp3IWGvZLxAhW6wjgGHbMSA4gQFjAAegQIBBAC&usg=AOvVaw3TQHX7dKj3pUiCoPCahlPA&cshid=1623514050065

 

It's a deliberately leading question you are posing which has no answer. 😉

 

Anyway, it's straying from the purpose of this thread.  This is about seeing if we can measure differences in a dac output, when a PC is unloaded and when loaded.

 

If you have some different measurements you would like tovsee performed we can try them.

 

Controlled subjective listening tests are the next step and I am planning to do that.

[March Audio]

1 minute ago, Summit said:

 

If they would be "inaccurate", why not explain how instead of removing them?

 

Yes I questioning your statements and have asked you to post evidence of what you claim to be facts.  That is the very core in objective audio discussions. 

Because he is going round in circles arguing black is white. .  The tests performed are objective.  Him repeatedly saying the opposite is just disruptive noise

[March Audio]

1 minute ago, PeterSt said:

 

Ah, we're on that tour again. Quote me where I said that.

 

All I see is that you mentioned DeltaWave all right, but did not use it for Jitter measurement. But, I could have missed that.

 

So ... we are back at putting word's in other's mouths in order to next happily use that. Wait, abuse that.

 

 

A new one. Just make things up, throw it over the fence and wait for buddies to upvote. You could try to quote me on that one again, but I'm afraid you are too dizzy.**

 

**): Now you can keep on repeating for the next two weeks that I called you a dizzy. No wait. I didn't say that.

"Yes, seriously. I was suggesting that nothing in the time domain changes"

[March Audio]

Also

 

"Nobody is changing the time domain. So IMO this can only about jitter "

[March Audio]

1 minute ago, PeterSt said:

 

Ahem, and where did *I* say that ?

Maybe I did. Some times I get dizzy too. A quote again would help. Preferably from this thread because then it is in-context.

So are you now saying that pc load variations are not audible, or that SSDs don't sound crap after all?

[March Audio]

4 minutes ago, PeterSt said:

 

Mind boggling.

Can you try Dutch please ?

Peter, you said the following:

 

"Nobody is changing the time domain. So IMO this can only about jitter "

 

This is a contradiction and demonstrates you don't understand the subject.

 

If you continue with posting pointless noise  I will just moderate and remove them.

 

[March Audio]

30 minutes ago, semente said:

 

You provide a set of measurements that does not explain many reports of perceived audible differences.

Isn't there a chance that something else might need looking into / measuring?

 

I don't have a horse in this race, I am merely curious.

Let's be clear, I am not saying that circumstances can't exist where dac outputs are affected negatively by the upstream PC.  I have already mentioned ground loops and lack of isolation.  I am planning to go on and demonstrate some of those problems.

 

However let's also be clear that unless those reports you are referring to were obtained in controlled conditions, they simply can't be trusted.  

[March Audio]

3 minutes ago, PeterSt said:

 

Wait ... that is a good one. So yes please, demonstrate for example the output of two bit perfect players and how they change the output of the DAC you measure relatively to each other.

That should be doable.

Haha

 

Peter, havent you grasped that it's not the players creating the difference?

 

I have always mentioned the issues of ground loops etc.

[March Audio]

7 hours ago, PeterSt said:

 

That depends who is thinking about it. I only work 15 years on players doing just that.

I work for 13 years on DAC's being able to withstand that. I still fail.

 

You ?

No wait, I already know the answer. Haha.

 

You know, it is such a pity that you keep on thinking I know shit.

And that makes it the other way around. Without words now, but you know ...

Appealing to your own authority is an unconvincing argument.  It's your own statements that keep betraying your lack of knowledge.  Basics that needed to be explained to you by others in this thread.

 

All we have seen is "hand waving",  zero objective evidence that supports your position and an awful lot of noise. 

[March Audio]

5 hours ago, jabbr said:

Thanks for this. Lots of arguments that are incomprehensible to me. For a real counterargument perhaps people could reproduce your measurements and show different results, or give us other good measurements that are comprehensible. I agree -- bottom line is 1) output of DAC or perhaps 2) output of speakers assuming volumes can be normalized.

Please ask questions, it's a complex subject.  I'm sure between myself, @pkane2001 and @idiot_savant we can make it more understandable.

 

Indeed, controlled subjective listening tests are something I intend to move on to.  Will take a bit of time and organising however.

[March Audio]

@pkane2001 would you be able to give a description of your DeltaWave software, what is does, but not too technical?  It would help others to understand the analysis being performed.

 

Thanks