Cleaning up USB signal for DAC

[ElviaCaprice]

Curiosity @ work. You have a $129 DAC and between the computer and the DAC you have approximately $1100 PCIe USB, power supplies, regeneration, and galvonic isolation.

 

Did you look at any ~ $1200 DAC's?

 

Correct, and your point is? Who says that the DAC component has to be of any certain relative value to be effective?

[plissken]

Correct, and your point is? Who says that the DAC component has to be of any certain relative value to be effective?

 

Just asking if you looked at any $1200 or any other price point then?

 

I don't know if I would call effective a $129 DAC that I had to put $1100 into the USB chain before it to get rid of digital and ground loop gremlins.

[ElviaCaprice]

Just asking if you looked at any $1200 or any other price point then?

 

I don't know if I would call effective a $129 DAC that I put $1100 into the USB chain before it.

 

Far better than $2300 total for effectiveness. Or are you saying that more money means more effective?? Less?? Why look further? Why not look further? Why do any of this? Why even ask?

Plisken, if I thought you we're genuinely interested and not looking to be argumentative, I would gladly be of an enumerative nature.

[plissken]

Far better than $2300 total for effectiveness. Or are you saying that more money means more effective?? Less?? Why look further? Why not look further? Why do any of this? Why even ask?

 

Just suprised to see $1100 thrown at a $129 DAC. I'm more used to seeing something like an $1100 DAC with $129 in the USB cable.

[Jud]

Just suprised to see $1100 thrown at a $129 DAC. I'm more used to seeing something like an $1100 DAC with $129 in the USB cable.

 

Some people might say all the money should have gone to better speakers.

 

We have such variety - that's part of what makes the conversation fascinating.

 

 

Sent from my iPhone using Computer Audiophile

[plissken]

Some people might say all the money should have gone to better speakers.

 

We have such variety - that's part of what makes the conversation fascinating.

 

 

Sent from my iPhone using Computer Audiophile

 

That's usually my recommendation.

[ElviaCaprice]

That's usually my recommendation.

Unfortunately your to quick to make that point often, especially with newbies coming here. It is a valid point, speakers/room treatment, but most of the time those newbies are here to learn more about the digital front end, not the back end. Which you are too dismissive about.

[plissken]

Unfortunately your to quick to make that point often, especially with newbies coming here. It is a valid point, speakers/room treatment, but most of the time those newbies are here to learn more about the digital front end, not the back end. Which you are too dismissive about.

 

I'm not dismissive about the back end. I wouldn't recommend a $129 DAC and $1100 worth of USB chain for starters.

[ElviaCaprice]

I'm not dismissive about the back end. I wouldn't recommend a $129 DAC and $1100 worth of USB chain for starters.

 

FRONT END (dismissive about), point made ty.

[patagent]

FWIW, I haven't seen any improvement in sound by removing USB interface from my primitive audio chain. My setup uses a Portable DAP (Teac HA 90SD) that plays off SD or functions as an external DAC connected via USB. I tried listening with my Sennheiser HD700 and speakers (P363) using all sorts of music files (MP3 to DSD). For now, I'm saving every penny towards nicer speakers.

[Shanepj]

Ignore, entered in error and cannot find how to delete it

[ElviaCaprice]

Hi Shanepj,

I agree about good cables, especially the backend (Analog). But I tend to have found that with proper galvanic isolation, the USB cables/adapters, other than proper impedance, don't make as much of a difference on the front end, other than shorter is better. In fact, it tends to be that if one is to find differences on USB cables on the front end, that this is a sign of ground loop noise, and what those expensive cables are doing is changing that ground loop noise that impacts the sound signature of the DAC conversion. At least this is my current theory, which could change with further findings.

 

I found my own optimization of Windows 10 to be just as effective, as others have found in comparison with AO/Fidelizer or combination.

 

Agree, about the effectiveness of the components not relative to pricing, necessarily, but overall system integration..

[Ralf11]

what is Nagra doing in their $35k DAC?

[Jud]

what is Nagra doing in their $35k DAC?

 

I like to think of DACs as comprised of digital design (the filtering/sample rate conversion), analog design, and parts quality for the first two.

 

The first can now be duplicated or exceeded for virtually all DACs by software running on a computer. Money put into this part of a DAC can be seen as money consumers might easily save.

 

But there is still plenty of room to distinguish DACs in analog design and parts quality.

 

Where the money has gone in the design and parts of the Nagra is something I haven't researched, since I'm so unlikely ever to be in the market for one.

 

 

Sent from my iPhone using Computer Audiophile

[Speed Racer]

I like to think of DACs as comprised of digital design (the filtering/sample rate conversion), analog design, and parts quality for the first two.

The first can now be duplicated or exceeded for virtually all DACs by software running on a computer. Money put into this part of a DAC can be seen as money consumers might easily save.

But there is still plenty of room to distinguish DACs in analog design and parts quality.

Where the money has gone in the design and parts of the Nagra is something I haven't researched, since I'm so unlikely ever to be in the market for one.

There is a big difference between "can" and "is". How many DACs out there are run in NOS mode?

 

Are you sure the sample rate conversion software out there is better than what is on the DACs?

 

For example, I am unaware of any software for my MacBook Pro that will do a better job than the "time- and frequency-domain optimized digital filter with a true closed-form solution" that "retains all the original samples" offered in my Schiit Audio Yggdrasil. I make sure my software is setup to send the untouched data to the Yggdrasil.

 

I care nothing about DSD or MCA.....

[Ralf11]

I also think of a DAC as those 2 components.

 

But Nagra's price is way beyooond say ARC or Ayre.....

[Jud]

There is a big difference between "can" and "is". How many DACs out there are run in NOS mode?

 

 

Almost none. All but a handful of DACs first internally oversample to 8x rates (352.8/384KHz), then sigma-delta modulate the result to MHz rates before the final conversion to analog. There are a handful of NOS DACs used without external upsampling that have a following, but this creates higher THD and IMD. Thus to the extent there's an audible difference, it demonstrably results from higher distortion. I have no problem with anyone who likes such a sound - everyone should listen to what they want.

 

Are you sure the sample rate conversion software out there is better than what is on the DACs?

 

For example, I am unaware of any software for my MacBook Pro that will do a better job than the "time- and frequency-domain optimized digital filter with a true closed-form solution" that "retains all the original samples" offered in my Schiit Audio Yggdrasil. I make sure my software is setup to send the untouched data to the Yggdrasil.

 

Have a look at http://src.infinitewav.ca/ , for example at the performance of iZotope 64-bit SRC versus the "ideal" filter performance curves. iZotope (used in Audirvana Plus) is quite close to the ideal.

 

 

I don't know what the performance of the Yggy filters looks like - I'm sure it's good, since Mike Moffat is a very smart man who has been designing digital filters a very long time - but very doubtful the Yggy exceeds performance so close to the ideal.

 

 

Miska offers closed-form filters in HQPlayer for those who like them, though he personally thinks other filters he offers exceed their performance.

 

 

The thing about retaining the original samples is marketing IMHO, unless you wanted binary-identical *re*conversion back to the original, lower digital resolution - closed-form filters uniquely allow you to do that. But that isn't what you're doing; you're taking the result of the Yggy's digital filters and doing a final analog filtering step to convert to analog, not re-converting to digital.

[Jud]

I also think of a DAC as those 2 components.

 

But Nagra's price is way beyooond say ARC or Ayre.....

 

 

 

Yep. I've heard the ARC DAC 8 and did not like it; have heard the Ayre QB-9 and liked it very much. Have never heard the Nagra.

[jabbr]

There is a big difference between "can" and "is". How many DACs out there are run in NOS mode?

 

Are you sure the sample rate conversion software out there is better than what is on the DACs?

 

For example, I am unaware of any software for my MacBook Pro that will do a better job than the "time- and frequency-domain optimized digital filter with a true closed-form solution" that "retains all the original samples" offered in my Schiit Audio Yggdrasil. I make sure my software is setup to send the untouched data to the Yggdrasil.

 

I care nothing about DSD or MCA.....

 

This topic has been discussed ad Infinitum on this forum.

 

Using a general purpose CPU for upsampling has very strong support among members here and is widely done. Both PCM eg see discussions of the Phasure NOS1 as archetypical for PCM as well as HQPlayer, Audirvana, ... well here's a list: http://www.computeraudiophile.com/content/649-20-audio-applications-enhance-your-music-listening-experience/

 

Am I sure? Im sure that there are many DACs which operate in NOS mode using software upsampling.

[richard kimber]

 

Have a look at http://src.infinitewav.ca/ , for example at the performance of iZotope 64-bit SRC versus the "ideal" filter performance curves. iZotope (used in Audirvana Plus) is quite close to the ideal.

 

Hmmnn, I get "src.infinitewav.ca: Name or service not known"

 

- Richard.

[Jud]

Hmmnn, I get "src.infinitewav.ca: Name or service not known"

 

- Richard.

 

 

Very sorry, that should be SRC Comparisons (I always forget the "e" at the end of "infinitewave").

[richard kimber]

Many thanks.

 

- Richard