[dougms3]

Ahh yes, apologies, what I mean by "bandwidth" is that SPDIF is limited to 24bit/192kHz, whereas USB can support up to 32 bit/768 kHz. The maximum communication throughput is totally irrespective of the audio file size. USB can support digital formats like native DSD for instance, whereas SPDIF has limited support (i.e. only DSD over PCM). Just one example...

And when you say that USB is "very noisy," what are you basing this on, exactly? Because from what I have seen, USB usually offers lower jitter, and ultimately less noise and distortion than SPDIF. Example--

https://www.thewelltemperedcomputer.com/Intro/SQ/USB_SPDIF.htm

In fact, if you follow Amir's posts on ASR, he demonstrates this fact regularly when he evaluates the noise and disortion coming from various DAC's inputs. So, I'm curious to learn where you are seeing SPDIF so routinely outperforming USB.

Ah ok, you're referring to the sample rate. Either way, it still wouldn't come close to maxxing out the usb bandwidth.

I think spdif is limited to dsd64 if I'm not mistaken but i2s can support higher levels of DSD. There are alot of people that don't like oversampling and will tell you that NOS sounds better. I prefer dsd upsampling but thats just me.

This may ruffle some feathers but what the hey. When I say noisy, I'm referring to electromagnetic interference. This is a sensitive subject because there are people who don't believe such a thing exists.

PC components like the cpu, ram, gpu, PS, etc, are very noisy, they generate noise and spread it around everywhere including into your audio system. Basically, anything that carries an electrical signal, has an EMF around it, and it moves with the flow of power. Most things are not affected by this noise but audio is very sensitive to this noise and on top of that its difficult to hear when its mixed in with music. If you play some white noise mixed in with your music, it'll be extremely difficult to isolate it by ear, most people won't even notice its there. However, once its removed, its very easy to tell as the sound improves.

I didn't say anything about spdif outperforming anything.

Amir is an amusing character, example, he reviews and judges the sound quality of audio devices without using his ears. It'd be the equivalent of a TV reviewer that refuses to look at the TV and only rates it based on measurements.

His methodology is quite silly, he measures the same metrics for devices that operate differently. All tube amps measure poorly, does that mean all tube amps don't sound good?

“If it measures good and sounds bad, — it is bad. If it sounds good and measures bad, — you've measured the wrong thing.” - Daniel von Recklinghausen

This doesn't really apply to amir since he doesnt use his ears at all.

 

[hsg120]

Ah ok, you're referring to the sample rate. Either way, it still wouldn't come close to maxxing out the usb bandwidth.

I think spdif is limited to dsd64 if I'm not mistaken but i2s can support higher levels of DSD. There are alot of people that don't like oversampling and will tell you that NOS sounds better. I prefer dsd upsampling but thats just me.

Not just sample rate, but bitrate. I'm just making a point of fact--that USB has a higher bandwidth than SDPIF, i.e. bits/second. USB can also transport native DSD (not DSD over PCM).

And I2S is neat and all, but I'm still not sure I understand what the advantage is if your source isn't I2S. And even then, I think the advantage is debatable at best.

This may ruffle some feathers but what the hey. When I say noisy, I'm referring to electromagnetic interference. This is a sensitive subject because there are people who don't believe such a thing exists.

PC components like the cpu, ram, gpu, PS, etc, are very noisy, they generate noise and spread it around everywhere including into your audio system. Basically, anything that carries an electrical signal, has an EMF around it, and it moves with the flow of power. Most things are not affected by this noise but audio is very sensitive to this noise and on top of that its difficult to hear when its mixed in with music. If you play some white noise mixed in with your music, it'll be extremely difficult to isolate it by ear, most people won't even notice its there. However, once its removed, its very easy to tell as the sound improves.

I didn't say anything about spdif outperforming anything.

Amir is an amusing character, example, he reviews and judges the sound quality of audio devices without using his ears. It'd be the equivalent of a TV reviewer that refuses to look at the TV and only rates it based on measurements.

His methodology is quite silly, he measures the same metrics for devices that operate differently. All tube amps measure poorly, does that mean all tube amps don't sound good?

“If it measures good and sounds bad, — it is bad. If it sounds good and measures bad, — you've measured the wrong thing.” - Daniel von Recklinghausen

This doesn't really apply to amir since he doesnt use his ears at all.

Maybe you're conflating quantifiable noise/distortion, with the perception of sound? I agree with you that one is measurable, and the other is not.

You said that USB is an inherently poor communication protocol for audio and mention sources of noise--so, my probably poor assumption was that you were referring to something quantifiable, and not just subjective. Because as I've pointed out, measurements show almost invariably that USB audio exhibits less jitter than SPDIF. And I2S is neat, but I can count digitial sources with a native I2S output on just one hand.

 

[dougms3]

Not just sample rate, but bitrate. I'm just making a point of fact--that USB has a higher bandwidth than SDPIF, i.e. bits/second. USB can also transport native DSD (not DSD over PCM).

And I2S is neat and all, but I'm still not sure I understand what the advantage is if your source isn't I2S. And even then, I think the advantage is debatable at best.



Maybe you're conflating quantifiable noise/distortion, with the perception of sound? I agree with you that one is measurable, and the other is not.

You said that USB is an inherently poor communication protocol for audio and mention sources of noise--so, my probably poor assumption was that you were referring to something quantifiable, and not just subjective. Because as I've pointed out, measurements show almost invariably that USB audio exhibits less jitter than SPDIF. And I2S is neat, but I can count digitial sources with a native I2S output on just one hand.

The advantage of i2s is there is no conversion of the audio signal. Reserve your judgement of either till you hear it.

If by conflating, you mean that thing where you thought I said spdif outperforms something, where in reality I didn't say anything about spdif. I'm not conflating anything with anything. I listen with my ears.

USB is not an inherently poor protocol for audio, it IS a poor protocol for audio. Read my previous post regarding the noise injection and 5v power intertwined with data.

Not sure why you keep bringing up spdif, I didn't say anything about spdif. Conflating issues I suppose.

Jitter is not the metric involved in audio quality.

Read the quote by Recklinghausen about measurements.

Congratulations on the counting.

It seems like you're not really looking for answers, you're looking for confirmation bias regarding your preference for usb. If you want to use usb, you're free to do whatever you want but no its not the best option for audio quality. And thats the reason why DDCs don't have usb output.

 

[hsg120]

The advantage of i2s is there is no conversion of the audio signal. Reserve your judgement of either till you hear it.

"Reservation of judgement" isn't really required though. Because the point is moot if there are like, even half a dozen(?) sources that offer a native I2S output. It's so uncommon to the point of being essentially irrelevant.

If by conflating, you mean that thing where you thought I said spdif outperforms something, where in reality I didn't say anything about spdif. I'm not conflating anything with anything. I listen with my ears.

No, I think you're unwittingly conflating "measurements," with "what sounds pleasing to the ear."

On the one hand, you mention how USB is an inherently noisy digital communication protocol--you know, things that are easily quantifiable with an APX555 Audio Precision Analyzer (which, Amir has).

Then on the other hand, when I showed you evidence that suggests that in practice USB is less noisy than SPDIF, you went off on a tangent about how "Amir doesn't use his ears".

And, I keep mentioning SPDIF because native I2S is so rare. Even then, I think the advantage is debatable. Then, you know, there's this--

https://www.audiosciencereview.com/forum/index.php?threads/study-is-i²s-interface-better-for-dacs-than-s-pdif-or-usb.7105/

USB is not an inherently poor protocol for audio, it IS a poor protocol for audio. Read my previous post regarding the noise injection and 5v power intertwined with data.

Neat, so USB carries a 5v supply along with the signal--except, to what detriment, exactly? What data do you have to show that carrying a 5v supply along with the data actually imparts any measurable interference in practice that modern USB DACs cannot handle?

I'm not trying to be a d*ck--I'm just asking for you to back up your theory with actual data.

Not sure why you keep bringing up spdif, I didn't say anything about spdif. Conflating issues I suppose.

=)

Jitter is not the metric involved in audio quality.

Well I mean, the last time I checked, electrical interference like the kind you're describing would be exhibited as jitter in any digital signal. So, maybe you can elaborate for me?

It seems like you're not really looking for answers, you're looking for confirmation bias regarding your preference for usb. If you want to use usb, you're free to do whatever you want but no its not the best option for audio quality. And thats the reason why DDCs don't have usb output.

Confirmation bias, what?? LOL I have no skin in the game. I'm just merely asking questions to try and learn, and understand your perspective. But it seems as though you're just getting agitated by my line of inquiry.

 

[dougms3]

"Reservation of judgement" isn't really required though. Because the point is moot if there are like, even half a dozen(?) sources that offer a native I2S output. It's so uncommon to the point of being essentially irrelevant.



No, I think you're unwittingly conflating "measurements," with "what sounds pleasing to the ear."

On the one hand, you mention how USB is an inherently noisy digital communication protocol--you know, things that are easily quantifiable with an APX555 Audio Precision Analyzer (which, Amir has).

Then on the other hand, when I showed you evidence that suggests that in practice USB is less noisy than SPDIF, you went off on a tangent about how "Amir doesn't use his ears".

And, I keep mentioning SPDIF because native I2S is so rare. Even then, I think the advantage is debatable. Then, you know, there's this--

https://www.audiosciencereview.com/forum/index.php?threads/study-is-i²s-interface-better-for-dacs-than-s-pdif-or-usb.7105/



Neat, so USB carries a 5v supply along with the signal--except, to what detriment, exactly? What data do you have to show that carrying a 5v supply along with the data actually imparts any measurable interference in practice that modern USB DACs cannot handle?

I'm not trying to be a d*ck--I'm just asking for you to back up your theory with actual data.



=)



Well I mean, the last time I checked, electrical interference like the kind you're describing would be exhibited as jitter in any digital signal. So, maybe you can elaborate for me?



Confirmation bias, what?? LOL I have no skin in the game. I'm just merely asking questions to try and learn, and understand your perspective. But it seems as though you're just getting agitated by my line of inquiry.

TLDR

Looks like you've been spending time with gregario.

ASR is the funniest comedy site I've ever seen, they review audio equipment without ever listening to anything.

 

[zukem]

I have owned several 'hi-fi' DDCs (e.g. Denafrips Gaia, Topping U90, Singxer Su6, etc) and often wondered why they almost invariably have a USB input, but never a corresponding USB output.

S/PDIF is fabulous, if you're exclusively doing 24bit/192kHz or less.

And if the proposed advantage of I2S is fewer digital conversion steps, then the argument is sort of moot if the source is USB in the first place xD

But USB--love it or hate it, undeniably offers the best support for all of the 'latest' digital audio formats.

So then, why do most DDCs go from a USB input, to all sorts of other arguably inferior digital outputs? For example, what if I wanted a DDC that just ensured that my DAC is getting a good USB input signal irrespective of the quality of the source? It doesn't seem as though there are 'hi-fi' DDCs that offer this(?).

I assume there's a reason why, but would be so grateful if someone would please fill me in.

Do you have any opinions on the U90? I'm trying to figure out if there are any real downsides to it over the gustard U18? I haven't really seen any ones actual opinion of it on here?

 

[dougms3]

Do you have any opinions on the U90? I'm trying to figure out if there are any real downsides to it over the gustard U18? I haven't really seen any ones actual opinion of it on here?

The audio-gd di24 is the best ddc out there.

 

[zukem]

The audio-gd di24 is the best ddc out there.

Looks like that just came out, I was looking to pick something up used and really like the form factor of the U90. I'll do some research on the audio-gd line though. Was the di-20 similar in performance to the 24?

 

[dougms3]

Looks like that just came out, I was looking to pick something up used and really like the form factor of the U90. I'll do some research on the audio-gd line though. Was the di-20 similar in performance to the 24?

No idea, I haven't tested the di24 but from other's impressions, they say its better.

The di20 is excellent, might be some of them on the used market.

 

[Loftprojection]

I get that USB is not good for audio but why are good DDC able to treat that signal with very good results after sending it coax to the DAC? Why aren’t DAC makers able to implement a « DDC USB » in so that their DAC sounds good on coax and usb? Just so they can make more money selling us DDC boxes, like Denafrip, Audio-GD, Gustard…!!!

 

[zukem]

I get that USB is not good for audio but why are good DDC able to treat that signal with very good results after sending it coax to the DAC? Why aren’t DAC makers able to implement a « DDC USB » in so that their DAC sounds good on coax and usb? Just so they can make more money selling us DDC boxes, like Denafrip, Audio-GD, Gustard…!!!

Yeah really... I have a Gustard R26 that I use as my desktop DAC and usb is so noisy while doing anything demanding on my PC. I'm hoping that a DDC solves it but I wish I didn't have to get one

 

[REXNFX]

I get that USB is not good for audio but why are good DDC able to treat that signal with very good results after sending it coax to the DAC? Why aren’t DAC makers able to implement a « DDC USB » in so that their DAC sounds good on coax and usb? Just so they can make more money selling us DDC boxes, like Denafrip, Audio-GD, Gustard…!!!

USB is fine, as per this review you just need a good USB cable.
https://the-ear.net/review-hardware/auralic-vega-g2-2-black-mirror-dac/

 

[Loftprojection]

USB is fine, as per this review you just need a good USB cable.
https://the-ear.net/review-hardware/auralic-vega-g2-2-black-mirror-dac/

I'm not technical but this seems to be an example of exactly what I'm talking about when I say manufactures should design their DAC so we don't need a DDC to get good USB processing like coax.

"What really differentiates the Vega G2.2 from the majority of its competitors is the presence of what Auralic call direct data recording, a fancy name for a large buffer that records up to 4GB of the incoming data prior to outputting it to the DAC using dual high precision femto clocks."

Sure this Auralic is a very expensive DAC but why can't Gustard for example implement some more electronics in their DAC so we don't need to buy their DDC for good USB connectivity!

 

[kingoftown1]

@zukem stay away from Topping (coming from someone who's owned a number of their products over the last few years). If you're having processor/pc activity noise coming through usb, you could always try something like this Elfidelity USB card as a cheap first test. For DDCs, I've had great luck with Singxer, and hear great things about Audio-GD and Denafrips models.

 

[zukem]

@zukem stay away from Topping (coming from someone who's owned a number of their products over the last few years). If you're having processor/pc activity noise coming through usb, you could always try something like this Elfidelity USB card as a cheap first test. For DDCs, I've had great luck with Singxer, and hear great things about Audio-GD and Denafrips models.

Thanks for the suggestions! As much as I liked the form factor of the topping I'm leaning towards the Gustard U18 or the Audio-GD di-20 currently. I unfortunately can't fit a card in my PC with my vertical GPU, hence why I'm looking at DDCs. I'm also just really curious to try I2S.