• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

RPi USB vs HAT S/PDIF output

mash

Member
Joined
Oct 8, 2020
Messages
69
Likes
62
I recently ran across this article on John Darko's site:

https://darko.audio/2021/03/12-more-thoughts-on-the-raspberry-pi-as-spotify-connect-endpoint/

In particular, his point 3, "Using one of the more carefully engineered S/PDIF DDC HATs will deliver superior sound quality to the Pi4’s USB port when feeding the same DAC" struck me as odd. Beyond the point that most DACs support better Hi-rez functionality through their USB interface, I would think that my high performance DAC can deal with most/all issues in the RPi USB stream and using a HAT to add an extra transform into the path would add as many potential issues as it would solve.

Glossing over the fact, that I shouldn't be spending time reading main stream review sites......I'm wondering what people think the best option for using an RPi streamer is.......HAT with S/PDIF out to your DAC or simple RPi (no HAT) USB out to your DAC?
 

Daverz

Major Contributor
Joined
Mar 17, 2019
Messages
1,026
Likes
1,015
I recently ran across this article on John Darko's site:

https://darko.audio/2021/03/12-more-thoughts-on-the-raspberry-pi-as-spotify-connect-endpoint/

In particular, his point 3, "Using one of the more carefully engineered S/PDIF DDC HATs will deliver superior sound quality to the Pi4’s USB port when feeding the same DAC" struck me as odd. Beyond the point that most DACs support better Hi-rez functionality through their USB interface, I would think that my high performance DAC can deal with most/all issues in the RPi USB stream and using a HAT to add an extra transform into the path would add as many potential issues as it would solve.

Glossing over the fact, that I shouldn't be spending time reading main stream review sites......I'm wondering what people think the best option for using an RPi streamer is.......HAT with S/PDIF out to your DAC or simple RPi (no HAT) USB out to your DAC?

DACs typically have better jitter performance on their USB connections. Just sample some of Amir's DAC reviews. Your instincts about that statement on Darko's site are correct; it's bullshit audiophile "wisdom".

(HATs might have made sense for sample rates over 48kHz on the RPi3, but that issue is irrelevant on the RPi4).
 

Stu23

Member
Joined
Sep 5, 2021
Messages
16
Likes
14
I've documented elsewhere my journey with the Pi, but I was replacing a Chromecast Audio using optical out into a MF V-Dac which sounded sweet but wouldn't play gapless.
First attempt was a Pi 3B+ usb out into the V-Dac. Compared to the Chromecast it was terrible. The sweet highs were gone and so was the separation and soundstage.
Adding a Hifiberry Digi2 Pro HAT and playing optical in to the V-Dac got it all back.
So, I think the point is, it depends on Pi and your Dac's usb implementation.
I've since swapped the V-Dac for the V-Dac ii which has asynchronous usb, and should be better, but haven't tried it that way yet.
 

voodooless

Major Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
1,947
Likes
2,841
Location
Netherlands
I've documented elsewhere my journey with the Pi, but I was replacing a Chromecast Audio using optical out into a MF V-Dac which sounded sweet but wouldn't play gapless.
First attempt was a Pi 3B+ usb out into the V-Dac. Compared to the Chromecast it was terrible. The sweet highs were gone and so was the separation and soundstage.
Adding a Hifiberry Digi2 Pro HAT and playing optical in to the V-Dac got it all back.
So, I think the point is, it depends on Pi and your Dac's usb implementation.
I've since swapped the V-Dac for the V-Dac ii which has asynchronous usb, and should be better, but haven't tried it that way yet.
If you haven't done any blind tests, your observations are quite meaningless. Then again, synchronous USB DAC's should be banned anyway, so there is at least a slight possibility that it might sound different in some way due to the worse jitter. I doubt though that it would change the tonal nature. Also, some measurements done by a fellow member (@LTig) would indicate that there should not be an audible difference.
 

JSmith

Major Contributor
Joined
Feb 8, 2021
Messages
1,101
Likes
1,861
Location
Algol Perseus
First attempt was a Pi 3B+ usb out
There were known issues with USB from a Pi3, this is no longer a concern with a Pi4, however Darko claimed otherwise. So you're not comparing the same thing he is.

Has been going on for a while;

RPi (no HAT) USB out to your DAC?
Yep... but with an Pi4.



JSmith
 

kchap

Active Member
Joined
Jun 10, 2021
Messages
133
Likes
116
There were known issues with USB from a Pi3, this is no longer a concern with a Pi4, however Darko claimed otherwise. So you're not comparing the same thing he is.

Has been going on for a while;


Yep... but with an Pi4.



JSmith
I started with the Pi4 inside the Nanosound ONE case and DAC. I later switched to USB DACs, currently a Kahdas Tone card and the Topping D10 balanced with a Topping E50 on the way. Very happy so far. I think you do have watch power consumption. The D10B and Khadas get power from the Pi4 so you need to carefully consider connecting peripherals like SSDs, CD drives and external displays. In some situation a DAC like the E50 is better because it has an external PSU.

Edited
 
Last edited:

Stu23

Member
Joined
Sep 5, 2021
Messages
16
Likes
14
If you haven't done any blind tests, your observations are quite meaningless. Then again, synchronous USB DAC's should be banned anyway, so there is at least a slight possibility that it might sound different in some way due to the worse jitter. I doubt though that it would change the tonal nature. Also, some measurements done by a fellow member (@LTig) would indicate that there should not be an audible difference.
Well I've been here about ten minutes and already my observations are meaningless ;). I don't claim to be a person that notices subtle differences, nor do I particularly care about them. My system is mostly secondhand, no bullshit stuff, set up for convenience as much as anything, and I only really ever listen to Spotify. I do like an open sound though and when listening I tend to pick out different instruments and their position in the "picture". I'm also very ssssensitive to sssssibilance, so can only listen to something with subtle highs.

Now, I was using the same DAC, power supplies, and headphones, so the difference was entirely between the Pi/USBcable/USB input on DAC and CCA/optical cable/optical input on DAC. The difference was not "do I think that's better or not?" so much as "where the f*** have my musicians gone?" Tonal differences were possibly less obvious, but it was more sibilant and less easy to listen to. Swapping back to the CCA gave instant relaxation and a "Oh God, that's better"... As I'd bought the Pi to replace the CCA, my "expectation bias" was pointing firmly in the other direction and I was very disappointed. No, I haven't done blind tests, but for my own use I didn't need to... this isn't a review and I'm not selling anything... one I like a lot, the other was plain unpleasant.

Putting the Digi2 HAT on the DAC got rid of the nasties and it now makes music I can listen to, much like the CCA did. I haven't compared them - I suspect the CCA might still be better and don't want to know!

I'll add 2 points: I was very much a "digital doesn't matter, bits is bits" kind of guy until I experienced this and the VDAC has since been found to not work at all on the digital coax input, so it's possibly not a perfect example, but it still sounds excellent on the optical.
 

Count Arthur

Major Contributor
Joined
Jan 10, 2020
Messages
1,033
Likes
1,987
"digital doesn't matter, bits is bits"
I agree, but it's totally possible to screw up the electronics involved at either the "sending" or "receiving" end or the conversion to analogue, or all of them.
 

somebodyelse

Major Contributor
Joined
Dec 5, 2018
Messages
2,205
Likes
1,640
There were known issues with USB from a Pi3,
There were known potential issues with the USB from a Pi <=3 but many people never experience them, and are happily using them without issue to this day.
Well I've been here about ten minutes and already my observations are meaningless ;).
Funny you should say that - our host did a video about exactly that yesterday ;)
https://www.audiosciencereview.com/...d-testing-you-are-doing-it-wrong-video.26809/

Measurements with different streamers with digital outputs (usb, spdif, toslink) both here and from archimago show that the output of a competent DAC doesn't change significantly with different digital sources. By 'significantly' I mean it's of the same order as the differences from rerunning the test on the same hardware. By 'competent' I mean not sensitive to the normal variations in things like USB power and jitter, which show up clearly in the DAC tests here - see the early modi tests for an example. We do sometimes see DACs that have a problem on specific digital inputs. There are also sometimes compatibility problems, but those tend to cause major malfunctions rather than a sound quality degredation. Ground related noise issues with single ended interconnects may be an audible problem with USB, or with spdif if it's not transformer isolated, but that's a system issue rather than one purely down to the source.
 

Yasser 06

Member
Joined
Dec 4, 2020
Messages
66
Likes
6
There were known issues with USB from a Pi3, this is no longer a concern with a Pi4, however Darko claimed otherwise. So you're not comparing the same thing he is.

Has been going on for a while;


Yep... but with an Pi4.



JSmith
Darko is rogue youtuber vampire, sucking the blood of his followers. I laughed so much about his "design" coffee maker that he inserted at the beginning of most of his video. He is such a typical millennium fraud with probably a fake degree in engineering. Willing to be smart glasses, allegedly neglected liitle beard. He is a full package. Weirdly, outside his propaganda activities, he is probably a nice and decent fellow.
 

Stu23

Member
Joined
Sep 5, 2021
Messages
16
Likes
14
Darko is rogue youtuber vampire, sucking the blood of his followers. I laughed so much about his "design" coffee maker that he inserted at the beginning of most of his video. He is such a typical millennium fraud with probably a fake degree in engineering. Willing to be smart glasses, allegedly neglected liitle beard. He is a full package. Weirdly, outside his propaganda activities, he is probably a nice and decent fellow.
I find him difficult to dislike, but I'm a simple man with simple needs, so I take his audiofoolishness with a pinch of salt. He makes a lot more sense than some people... about the only thing that makes any scientific sense on this link is using a battery as a 12V power supply...

 

abdo123

Major Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
3,557
Likes
3,124
Location
Brussels, Belgium
Well I've been here about ten minutes and already my observations are meaningless ;).

your observations are not meaningless, you just have to provide a plausible scenario / evidence that supports your observations. you don't believe everything that you read online by anonymous profiles do you? :)

Well that's our perspective.
 

Stu23

Member
Joined
Sep 5, 2021
Messages
16
Likes
14
Please yourselves... I was just trying to help. I don't have to prove anything to anyone. I don't have evidence to support my hypothesis, it's a statement of personal fact. Take it or leave it, just don't tell me my experience is meaningless.

OP was asking about differences between USB and HAT... the opinion of this forum seems to be there can't be one, my own experience suggests otherwise. I'm sure it's highly device dependent, but it would appear some USB implementations are not perfect.

I'm a scientist and an engineer. But if I'm buying hifi equipment, I'd rather buy something I enjoy listening to than something that has a perfectly flat frequency response or vanishly low SN ratio.

The USB may be perfect between my Pi and the VDac, I don't care, it is unpleasant for me to listen to.
 

voodooless

Major Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
1,947
Likes
2,841
Location
Netherlands
statement of personal fact.
That is a contradiction. Facts don't have feelings. "Personal facts" are no facts, they are beliefs.
Take it or leave it, just don't tell me my experience is meaningless.
Well, I might have been a bit harsh with the term "meaningless". Could have worded it better. Will do next time.
OP was asking about differences between USB and HAT... the opinion of this forum seems to be there can't be one, my own experience suggests otherwise. I'm sure it's highly device dependent, but it would appear some USB implementations are not perfect.
I didn't say there can't be one, I just said that the likelihood is small and if there is it's probably due to some other issue.

The point is not to just ignore your statements but to more closely look at what is going on so we can actually find out what's going on. That is much more informative than just blatantly stating that USB vs HAT is a win for the HAT. Because without knowing why this is, it is also of little help to anyone.
 
Last edited:

Stu23

Member
Joined
Sep 5, 2021
Messages
16
Likes
14
That is a contradiction. Facts don't have feelings. "Personal facts" are no facts, they are beliefs.

Well, I might have been a bit harsh with the term "meaningless". Could have worded it better. Will do next time.

I didn't say there can't be one, I just said that the likelihood is small and if there is it's probably due to some other issue.

The point is not to just ignore your statements but to more closely look at what is going on so we can actually find out what's going on. That is much more informative than just blatantly stating that USB vs HAT is a win for the HAT. Because without knowing why this is, it is also of little help to anyone.
OK, it's a personal truth.

I didn't state that HAT beats USB. In other applications asynchronous USB is excellent. I've experienced it myself with a laptop and an Audiolab 8200CDQ I used to own. Superb.

I said that in my own experience, with my components, it made a huge difference.

it did. It wasn't remotely subtle. Far bigger than replacing the VDac with a cheapie DAC from ebay (£11) that I bought to test the coax output of the HAT... I'd rather listen to that than the VDac/pi through USB. That one surprised me the other way... it's pretty good for a few quid and I'd be worried about blind testing it!

For some reason you have made it your mission to demean my observations despite the fact that you weren't here, and didn't hear it one way or the other. I'm not promoting a belief system here, just reporting an observation. You, on the other hand, seem to have some preconceived idea that I can't possibly know what I'm talking about because my experience doesn't agree with your understanding.

What is a fact is that MF sold a V-Link unit for the V-Dac which took a USB output and converted it into a (I think) coaxial S/PDIF signal. Perhaps they knew there was a problem and came up with a solution. A cynic might say they engineered it badly in the first place to sell v-links. Either way they were openly saying that there was improvement to be had on their USB Implementation.

The VDac 2 I now have has the asynchronous v-link built in. Sometime I'll set up a comparison... you're welcome to come and have a listen. Bring your favourite headphones!
 

voodooless

Major Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
1,947
Likes
2,841
Location
Netherlands
OK, it's a personal truth.
That is better indeed :)
I said that in my own experience, with my components, it made a huge difference.

it did. It wasn't remotely subtle. Far bigger than replacing the VDac with a cheapie DAC from ebay (£11)
Well, for one, we’ve heard this more than once. In many cases, it turns out people were just wrong: when removed of all biases, all differences disappear like magic. In the remaining cases, it’s always interesting to find out what’s going on. Sometimes this will lead to some interesting insights.
You, on the other hand, seem to have some preconceived idea that I can't possibly know what I'm talking about because my experience doesn't agree with your understanding.
An experience is just that: an experience. It does not necessarily correlate with reality. That means such observation should always be judged with a lot of care. If I were to experience the exact same thing, I would react the same way. So it has nothing to with anyone personally.

If you have something to record the output of your DAC, we could investigate.
 
Last edited:

Stu23

Member
Joined
Sep 5, 2021
Messages
16
Likes
14
I have now removed the VDac from my system and chucked it in a store pile. I would be happy to send it to you or anyone else that can test it with a pi3 via USB and compare it against the optical input. If they can fix whatever is stopping the coax input working you can test that too. You can also have the minidac from ebay if you like, I suspect it's quite good for the money.
I'd like them back eventually, but I'm in no hurry, the VDac2 is doing the job very well!
 
Top Bottom