Diretta Measurements and Listening Tests

Sure, thanks for pointing that out:

  • DAC: Topping D70
  • Analyzer: E1DA Cosmos ADC Grade A + ARTA software
  • Scope: Siglent SDS1052DL

Also, forgot to mention that all tests were against Diretta version 147_13-1 (both host and target), which was the latest at the time of my setup.

Let me know if you need additional details.

2 Likes

But it does, to my set of ears and perception. I do have a reasonable understanding of how async usb audio works though, so the question you’re asking is relevant, and the core of this.

Something in the shape or form of the data flow to the DAC changes, but how does this affect the transformation from packets into an audio signal?
I don’t have the answer to that, but there are a lot of good thinking here, so i have my hopes that this can be verified by measuring something that correlates to the listening experience.

If the USB transfer mode is async, the target controls the flow of the data, so if the source did not honor that, you’d get drops (buffer underflow) or ā€œskipsā€ (buffer overflow). If the transfer was controlled by the source, then the DAC would have to adjust its clock to the source rate, and would have to use methods like PLL to reject jitter. That would be an unnecessary complication for USB. Measurements don’t show any difference in jitter response, so either the USB is still used in async mode, or the DAC rejects jitter very well, or the Pi’s USB clock is very good. Perhaps @David_Snyder knows how Diretta target uses the USB. I can’t make much sense of what they say on their site.

3 Likes

Your interpretation is correct.

1 Like

I agree, but I think that effort would not be justified at this point. I’m not saying Diretta is not changing the flow as they say they do, just that those changes don’t seem to matter.

2 Likes

If my understanding is correct, that’s Diretta’s claim; not mine in any case. Otherwise, why would there be a need for smoothness in the packet transmission?

Jitter test is supposed to be quite sensitive to time domain variations. Even if inaudible, the FFT should reveal some differences, and I don’t see any.

3 Likes

Your Brain. We all not only hear differently due to individual physical reasons, we also perceive and process the sound differently due to inherent different memories and personal experiences.

Here is an article about the topic; ignore that article, however, you should follow the link in the article TO the originating article, and read it. Fascinating stuff.

Personally, I find that the visual cortex of the brain produces too much interference to actually ā€œlistenā€ to music. I need darkness with closed eyes, to fully focus on the music I’m listening too.

6 Likes

Yawn

And frankly, I’m disappointed that you’ve over stepped your remit as a moderator

A moderator is a volunteer and fellow enthusiast who will have views and experiences that may differ from yours. In agreeing to take on the task of moderation he doesn’t become a silent drone there to make life easy for you. I’d calm down. The rest of us will take you more seriously if you can deal with opposing criticism of which there may be a lot as there usually is in this type of discussion.

6 Likes

ā€œYour Brainā€

Yes, and that was a direct response to your question, but, meant in the same joking spirit of the question. So, let me put the full quote of what I was responding to, grinning face and all…

Besides, I clarified that short answer with little more fleshed out reasoning and then a link to the article/study from whence that answer was distilled, and gave an example from my personal experience in support.

I’m re-posted the direct link because I think it was pertinent to the discussion, and had gotten a bit lost.

ā€œhttps://assets.cureus.com/uploads/review_article/pdf/184566/20240227-2971-ae3iu8.pdfā€

2 Likes

Thanks for the links. I’ve read the first linked article (I’ll read the study later), and I think that’s also worth reading. I was thinking that one of @David_Snyder’s points - that the perceived improvements of Diretta are not due to the novelty factor because it keeps sounding better the more you listen to it - might fit the pattern of ā€œbrain burn-inā€ described there, but in Diretta’s case, nothing actually changes in the sound signature, so there’s no acclimatization of the brain to something objectively different. In any case, it seems ā€œour brainsā€ :face_with_tongue: work in vastly more mysterious ways than networking protocols or USB connections.

2 Likes

If like me you are short sighted and usually wear glasses, you may find that you can hear much better with your glasses on. Whilst counter intuitive, to me it makes perfect sense. If the brain is working overtime trying to make sense of an out of focus world it is diverted from other tasks. Hearing and other senses do suffer without my glasses on, too many distractions. Critical listening is best done on headphones with eyes closed.

If you have the appropriate gear, it would be good to see an inferred jitter comparison as well: Wiim Ultra Jitter Technical Investigation | Super Best Audio Friends

Thanks Fernando. I don’t have a dScope and it seems those start around $4K, so I’ll pass…

1 Like

A post was split to a new topic: Pi Zero 2W configured as a Roon endpoin

A post was merged into an existing topic: Pi Zero 2W configured as a Roon endpoint

I was in touch with an electronics designer here in Sweden, to see if there was further tests to be suggested.
He supported my thesis that it is very difficult to find measurements that are qualitative, repeatable and first and foremost: That correlates with the listening experience…

Anyways, a theory he suggested was that minute variations in periodicity (due to smoother power consumption) of the USB transfer actually could modulate the audio signal in the DAC, into the audible spectrum.
Well, you are looking at exactly that and see nothing of relevance, at least not with synthetic signals..

So the dilemma hasn’t changed, none of the measurements you are able to perform shows any significant differences whether with or without Diretta in the transfer. As a side note there are a fraction of audiophiles that would constitute that as absolute transparency, which is considered premium, in those circles! :nerd_face:

So, still looking forward to if you and the kids have found any preference in listening!

4 posts were merged into an existing topic: I used to think the DAC was 90% of the sound. This sub-$300 DIY project proved me wrong

Please keep this thread on-topic. It’s a disussion about the measurements and listening tests of Diretta.

By all means, reply as a related topic from this thread (the reader will see a link here.)

Accordingly, I will move (rather than delete) posts that the moderators consider off topic, even if related to Diretta.

1 Like