USB audio cracked... finally!

[sandyk]

10 minutes ago, Ralf11 said:

 

eh??

 

 Use the Search facility.

 

[Ralf11]

I am going to assume you were just rounding off, and not invoking the truly terrifying prospect that people were going after you 6 months or so BEFORE you showed up.

 

But I doubt what you say.  If it really IS true, then maybe it is something about the things you say or the way you say them.

 

 

[GeneZ]

1 hour ago, fas42 said:

 

Yes ... people keep reaching out for 'magic', one-size-fits-all solutions ... but, it don't ... work ... that ... way.

 

In my world of dealing with audio systems, I see nearly all of them as being a rat's nest. Ohh, they may look like pretty nifty - but electrically they are as messy as the "rat's nest" of cabling one often finds behind the pretty metal cases - those messes, out of sight, are pretty telling of the fact that little effort has gone into "the devil's in the details" side of things.

  If find that Gen 5 sounds pretty darn good.  But, I also have addressed my AC quality.  

 

Using litz cabling also eliminates a certain nasty sound effect that only solid core can touch, but not as well. 

 

There are just too many variables to contend with besides the USB. 

 

I have had more than a few DACs in my system.  What ever Schiit is doing with theirs, its the first one I listen to making listener fatigue a thing of the past. 

 

Just the other night  I substituted a different power cord to the DAC and my system's sound is once again transformed.  

 

When the system reaches a level of transparency that begins to hone in on the target, anything you change in the system will be heard.

 

Gen 6?   Not sure its what most people need if they do not address everything else that has an effect on what is heard.

[GeneZ]

45 minutes ago, Ralf11 said:

I am going to assume you were just rounding off, and not invoking the truly terrifying prospect that people were going after you 6 months or so BEFORE you showed up.

 

But I doubt what you say.  If it really IS true, then maybe it is something about the things you say or the way you say them.

 

 

 Hillary is looking for a new lawyer.  Pays well.

[fas42]

16 minutes ago, GeneZ said:

 

Just the other night  I substituted a different power cord to the DAC and my system's sound is once again transformed.  

 

When the system reaches a level of transparency that begins to hone in on the target, anything you change in the system will be heard.

 

 

The level of transparency one should be after is that one hears deeper and deeper into the recording - I have 'layers' of CDs, so to speak - there are beginner ones, which any sort of system claiming to be of high end status should do a reasonable job with; right through to highly 'sludgy' ones - say, a very messy, old pop mix, from a poor quality master - the latter will be "drive everyone from the room!" quality from even pretty decently sorted rigs; when they come together then you know your rig's runnin' pretty hot!

 

When 'balancing' the SQ using the latter material, every tiny variation will be obvious, yes - the good news is that a quality hump is there to be crossed, from whence futher subjective improvements are relatively mild ... you've "broken the back" of the system anomalies which get in the way of achieving competent sound.

[GeneZ]

On 12/30/2018 at 10:53 AM, Dutch said:

 

What he said is they’ve developed their own USB.org certified USB interface processor (so hard- and firmware plus software as in drivers) that could replace those from XMOS etc. So it’s still USB but their own, fully controlled and understood implementation of it.

  The socket remains a USB configuration.   Its no longer USB as we have known it. 

[GeneZ]

22 hours ago, fas42 said:

 

The level of transparency one should be after is that one hears deeper and deeper into the recording - I have 'layers' of CDs, so to speak - there are beginner ones, which any sort of system claiming to be of high end status should do a reasonable job with; right through to highly 'sludgy' ones - say, a very messy, old pop mix, from a poor quality master - the latter will be "drive everyone from the room!" quality from even pretty decently sorted rigs; when they come together then you know your rig's runnin' pretty hot!

 

When 'balancing' the SQ using the latter material, every tiny variation will be obvious, yes - the good news is that a quality hump is there to be crossed, from whence futher subjective improvements are relatively mild ... you've "broken the back" of the system anomalies which get in the way of achieving competent sound.

 

I have been considering finding a good CD player that would hook into my Yggdrasil for its audio processing.   I am at a loss as what to look for. Getting the PC out of the way I understand helps for better sound quality.   Even better,  I was thinking of possibly a DVD/Blu-ray player, but they seem to contain their own DACs.  I would surely want the Yggy taking over the audio and not to use a DVD/Blu-ray player's DAC.  

[Superdad]

11 minutes ago, GeneZ said:

  The socket remains a USB configuration.   Its no longer USB as we have known it. 

 

Come again?  If it uses USB protocols, then there is a PHY and MAC processor of some sort handling the half-duplex signal. So still USB pretty much as we know it. As mentioned before, if they decide to write or license a USB IP core to run on an FPGA, that's fine--as long as they take good care of drivers, etc.

 

Short of creating a new electrical interface (certainly the data can be reformatted--still needing a protocol engine--and send over a wire full-duplex/differential if they really work at it), Schitt's new USB will still be USB.  Nothing wrong with that though. 

[Ralf11]

True, but it may not be "Your Parents' USB"

[GeneZ]

1 hour ago, Superdad said:

 

Come again?  If it uses USB protocols, then there is a PHY and MAC processor of some sort handling the half-duplex signal.  So still USB pretty much as we know it. As mentioned before, if they decide to write or license a USB IP core to run on an FPGA, that's fine--as long as they take good care of drivers, etc.

But short of creating a new electrical interface (certainly the data can be reformatted--still needing a protocol engine--and send over a wire full-duplex/differential if they really work at it), Schitt's new USB will still be USB.  Nothing wrong with that though. 

 

Not sure if we are mincing words here.  

 

  I'm a little dense here. So? The USB signal that the DAC will see with the Schiitized USB?   It will be the same USB signal as before?   Or, simply the same roadway as before, but better paved?

 

After listening to what you said? That is what I read. 

 

Unless you are saying that Schiit will create software, designed to bypass and eliminate the negatives that USB at present attempts have failed to deal with.  Is that what we are talking about?   Like an ASIO on steroids?

[Hugo9000]

17 minutes ago, GeneZ said:

 

I have been considering finding a good CD player that would hook into my Yggdrasil for its audio processing.   I am at a loss as what to look for. Getting the PC out of the way I understand helps for better sound quality.   Even better,  I was thinking of possibly a DVD/Blu-ray player, but they seem to contain their own DACs.  I would surely want the Yggy taking over the audio and not to use a DVD/Blu-ray player's DAC.  

You could take a look at the Yamaha Aventage BD-A1060.  It supports Blu-ray, DVD, CD, SACD, etc., and has digital outputs if you wish to avoid the built-in DAC.  It also has balanced analog outputs, if that's of any interest.  It looks pretty well-made for the price.  They have some internal shots on the features page for it.

 

https://usa.yamaha.com/products/audio_visual/blu-ray_players/bd-a1060/features.html#product-tabs

 

If/when my Denon disc player kicks the bucket, that's probably what I'll get to replace it, unless something better for the money comes along by that time haha!

 

 

Or, you could get a CD transport.  Cambridge Audio makes an affordable one.  I have never owned anything by that brand, so I have no idea about their quality control or longevity.

 

https://www.cambridgeaudio.com/usa/en/products/cx/cxc

 

 

[fas42]

49 minutes ago, GeneZ said:

 

I have been considering finding a good CD player that would hook into my Yggdrasil for its audio processing.   I am at a loss as what to look for. Getting the PC out of the way I understand helps for better sound quality.   Even better,  I was thinking of possibly a DVD/Blu-ray player, but they seem to contain their own DACs.  I would surely want the Yggy taking over the audio and not to use a DVD/Blu-ray player's DAC.  

 

So, you would be getting the CDP to try and achieve a higher standard of replay; rather than just have an alternative to computer input, when you happen to have CDs on hand? If the former, I would be looking at what experiences people out there have had with their Yggdrasils with various CD devices - there might be a happy combo that have, yes, good synergy; which simply means that they don't tread all that much on each other's toes, electrically . Or at least less so than other pairings.

 

Personally, I would continue to try and squeeze the best sound out of that DAC from computer source. As a way of finding out more about what one has to do to make it happen. I just happened to stay with CDs for historical reasons - I've largely tuned out of fiddling with computers these days; they're a means to an end, and absolutely nothing more.

[Ralf11]

the Lexicon CD player is... legendary

[Hugo9000]

35 minutes ago, Ralf11 said:

the Lexicon CD player is... legendary

Wouldn't notorious be a better word? 

[pkane2001]

On 12/27/2018 at 11:13 AM, manisandher said:

 

I look forward to your findings. (If there's anything you'd like me to repeat, let me know.)

 

Mani.

 

Let me share some of the preliminary findings, as I had a few minutes to play with A/B data comparisons.

 

The new (experimental) tool I added to my DeltaWave software is the ability compare average phase for both captures. Phase represents a timing offset of individual frequencies. As such, it should be a good test for how close timing matches between the two waveforms. Phase is expressed in degrees, the bottom axis is the frequency.

 

Here's the phase difference between two A captures, the smaller the vertical excursions, the smaller the phase error, the better is the match between the two captures:

 

You'll notice that while there are occasional jumps, up to about 13KHz, the phase differences remain fairly controlled and small.  

 

Here is the phase difference between A and B captures:

 

Here, the error starts increasing much earlier, and just above 1KHz is already well above the level of the A/A phase difference at the same frequency.

 

A B/B comparison produced something interesting. It turned out to be much more like the A/B comparison than the A/A comparison:

 

Here are the RMS phase errors up to 10KHz for each of the comparisons:

	A vs. A	A vs. B	B vs. B
20Hz - 10KHz	1.41°	8.62°	7.70°

 

I repeated the same test with a couple of other captures from the same set of tests and these results were consistent. The phase error was much larger whenever a B capture was used in the comparison, even when comparing two Bs. One possible explanation is that the timing error was much larger and possibly more variable in the B captures. 'A' captures match up to each other much better in phase, at least up to about 13KHz, while B captures don't match well to As or other Bs. Any other explanations?

 

Whether these differences are audible or not, I don't know. Based on Mani's ABX results, probably yes. Regardless, there is a difference in timing (phase) of various frequency components between A captures and B captures, with B being more random and with larger variations. Maybe Peter can explain why this might happen with whatever was the SFS setting used for B captures, as that was the main difference between A and B.

 

[PeterSt]

@pkane2001, it could be an idea to install XXHighEnd on some random Windows PC you have hanging around. It would allow you to compare on my guidance. I don't think this is about (the proof of) listening tests any more. How to capture (record) yourself the differences is a subject on its own, and probably fairly undoable (that was my premise to the red/blue pill experience and that didn't change). So key sub-project here is to rule out the recording anomalies. Right ? This requires easy change of settings of which I unconditionally know that they will make a difference to the output. With this practice under control (that leading to the better versions of DeltaWave), we will be up to the real task.

 

I will start thinking of what you just came up with, already knowing that all we perceive for differences will be about phase ("changes").

 

Having said this, might you have two PC's, one with a later Windows 10 and an other with Windows 8, you might be able to squeeze out differences even more easily, but now with the same XXHighEnd settings on both instances. N.b.: Windows 8.1 might be more logical to have, but is not suitable (has never been supported because too bad IME for audio). Windows 7 SP1 (not RTM) would also be a best contender (next to W10).

 

Peter

[pkane2001]

16 minutes ago, PeterSt said:

@pkane2001, it could be an idea to install XXHighEnd on some random Windows PC you have hanging around. It would allow you to compare on my guidance. I don't think this is about (the proof of) listening tests any more. How to capture (record) yourself the differences is a subject on its own, and probably fairly undoable (that was my premise to the red/blue pill experience and that didn't change). So key sub-project here is to rule out the recording anomalies. Right ? This requires easy change of settings of which I unconditionally know that they will make a difference to the output. With this practice under control (that leading to the better versions of DeltaWave), we will be up to the real task.

 

I will start thinking of what you just came up with, already knowing that all we perceive for differences will be about phase ("changes").

 

Having said this, might you have two PC's, one with a later Windows 10 and an other with Windows 8, you might be able to squeeze out differences even more easily, but now with the same XXHighEnd settings on both instances. N.b.: Windows 8.1 might be more logical to have, but is not suitable (has never been supported because too bad IME for audio). Windows 7 SP1 (not RTM) would also be a best contender (next to W10).

 

Peter

Hi Peter,

 

I do have a Windows 10 PC I use with my audio system, so can use that for testing.

 

Windows 7 or 8.1 I'll have to install on a Macbook Pro using Boot Camp, but that's also doable. Not sure if I can still find Win7 SP1 ISO to install, but I'll take a look.

[PeterSt]

1 hour ago, pkane2001 said:

I do have a Windows 10 PC I use with my audio system, so can use that for testing.

 

Paul, please stick to that then. Anyway, I came to the idea because of my own "Phase" remark and the small year it took to get Windows 8 right on that once, where everybody (of the XXHighEnd community) had the same Phase problems with it (so me too). This was relative to its predecessor W7 and with W10 the general problem disappeared (or could not be noticed any more because of how XXHighEnd was set up by that time).

 

XXHighEnd within itself will do just the same, but might look less trustworthy because of (for you unknown) changed settings. A difference between two OSes with the same settings, could be more "speaking". But may also not be really worth while. All takes time as well.

[manisandher]

8 hours ago, pkane2001 said:

 

Let me share some of the preliminary findings, as I had a few minutes to play with A/B data comparisons.

 

The new (experimental) tool I added to my DeltaWave software is the ability compare average phase for both captures. Phase represents a timing offset of individual frequencies. As such, it should be a good test for how close timing matches between the two waveforms. Phase is expressed in degrees, the bottom axis is the frequency.

 

Here's the phase difference between two A captures, the smaller the vertical excursions, the smaller the phase error, the better is the match between the two captures:

 

You'll notice that while there are occasional jumps, up to about 13KHz, the phase differences remain fairly controlled and small.  

 

Here is the phase difference between A and B captures:

 

Here, the error starts increasing much earlier, and just above 1KHz is already well above the level of the A/A phase difference at the same frequency.

 

A B/B comparison produced something interesting. It turned out to be much more like the A/B comparison than the A/A comparison:

 

Here are the RMS phase errors up to 10KHz for each of the comparisons:

	A vs. A	A vs. B	B vs. B
20Hz - 10KHz	1.41°	8.62°	7.70°

 

I repeated the same test with a couple of other captures from the same set of tests and these results were consistent. The phase error was much larger whenever a B capture was used in the comparison, even when comparing two Bs. One possible explanation is that the timing error was much larger and possibly more variable in the B captures. 'A' captures match up to each other much better in phase, at least up to about 13KHz, while B captures don't match well to As or other Bs. Any other explanations?

 

Whether these differences are audible or not, I don't know. Based on Mani's ABX results, probably yes. Regardless, there is a difference in timing (phase) of various frequency components between A captures and B captures, with B being more random and with larger variations. Maybe Peter can explain why this might happen with whatever was the SFS setting used for B captures, as that was the main difference between A and B.

 

 

Paul, this looks really encouraging. Perhaps it'd make more sense to post further findings in the red/blue pill thread itself, or indeed start a new thread?

 

Mani.

[manisandher]

6 hours ago, PeterSt said:

I don't think this is about (the proof of) listening tests any more.

 

I think the correlation between measurements and audibility remains of paramount importance. If Paul's measurements bring up anything interesting, I'd consider redoing some of the blind listening tests.

 

Mani.

[sandyk]

18 minutes ago, manisandher said:

 

I think the correlation between measurements and audibility remains of paramount importance. If Paul's measurements bring up anything interesting, I'd consider redoing some of the blind listening tests.

 

Mani.

  Mani

Perhaps Shadders (Richard) may be interested this time around, depending on where he lives in the U.K.  ?

 He could bring a different perspective.

 

Regards

Alex

[CuteStudio]

On 12/31/2018 at 8:07 PM, 4est said:

Perhaps I didn't make it clear when I have said that I have done this? That would include Volumino too btw. I am sorry that I sound too emotive for your tastes. I do not care to be misquoted or chided. You have done both. Unless it has changed, using the i2s out of a Pi requires ASRC, which I find inferior both technically and sonically to synchronous clocking on the DACs I have tried. These were DIY affairs where one can get by using super short ( under 100mm/4") i2s lines. I imagine it is possible to do if things are in cases, but the leads are best as short as possible. Even still, this is DIY and it is impractical to have a micro computer dangling off the back of your DAC.

 

As to SandyK, you are completely mistaken. He and I are familiar with each other. I was suggesting he watch how he speaks to people. It had nothing to do with audio, but forum etiquette.

 

Mostly I didn't like the way you suggested that everyone is foolish to not do what you suggest when there are alternatives that technically exceed that suggestion. I'll not clog this thread about this further and I am sorry if I went off the handle.

 

Misquoting and chiding? I cannot be responsible for how you interpret my comments, although I will try to avoid the triggers as I learn them of course.

 

You were the one who said my comment was 'steeped in BS':

I asked politely for you to clarify, can you please clarify that statement?

 

I have to state MY views here. I cannot state yours or anyone else's or we descend into Groupthink. The entire improving USB debate is to me an exercise in 'perfecting mediocrity' and is a problem that has already been solved. You chose to interpret that as me calling you foolish, that was not intended, the intended purpose was to say 'hey guys, if you want a perfect digital link to your DAC look at I2S because it already does it and has for ages". Again, I cannot be responsible for your interpretations of my words, only of the words themselves.

 

My observation that people spent their Christmas on here bitterly debating cables also appears to be true if you look back at the posts. It's odd that the content that's supposed to be emotive (MUSIC) evokes far less emotion that cables which for some reason always get everyone fired up. It's bizarre, and also the reason the man on the street laughs as 'Hifi people'. I've read entire reviews about how a cable 'sounds' that only mention in passing the music, the speakers, the amp or the DAC etc. To me that is laughable, but I realise I need to change as a person to gain reverence for cables, and that this heretical attitude makes me a bad person to be scorned in forums such as this. For the man on the street however this cable worship has killed the HiFi industry. It's done wonders for the cable industry though LOL, volumes are down but money is up. Much easier to sell a $200 cable than a decent $200 amplifier and far far less work.

 

Now back to I2S:

A) If one runs the Pi I2S with the internal Master there is of course serious jitter and an ASRC and or reclocker will help.

B) If one runs the Pi with an external clock (Slave) then the timing is as accurate as the external clock.

 

Are you saying that in scenario B you still need an ASRC? If so - why?

 

BTW Apologies in advance for using 'master' and 'slave', these words have been deemed offensive in today's PC minefield and companies like DJango are redefining them as 'primary'/'replica' instead, which means master/slave.

[marce]

I2S was developed as an on board interface, it was never designed as a a digital interface to go over cables any distance and apart from the 1986 Phillips spec, there is not a lot else... Not the best interface for distance between boxes, OK for on PCB chip to chip comms. Transmit it over a distance and you get the same problems as any other digital interface.

 

https://www.sparkfun.com/datasheets/BreakoutBoards/I2SBUS.pdf

[Summit]

One of the great thing (in the developed part of the world at least) is that we can chose between a lots of things. Some of our choices can have some negative external effects, while others only the person that uses the stuff will notice. Which type of digital protocol or cables someone want to use is a good example of then a personal choice is just a personal choice that won’t inflict harmfully on other people or their possibility to make their own choices.

 

I have some experience with I2S, both as an I2S signal and as I2S over LVDS. With my former DAC the I2S over LVDS was better sounding than: I2S over LAN, USB, S/PDIF and AES/EBU. The point I like to make is that no matter which digital protocol you/we chose the implementation of the design is of very big importance for the outcome, and what’s best with one DAC is not necessarily the best with another DAC or audio system.

 

I don’t have the possibility to choose I2S on the DAC I now use so I can’t say if it would sounding better or worse than USB or SPDIF. But even if I had the option I wouldn’t really know if it was because I2S is truly better than USB (direct) per se, or if it’s the extra filtering stage which a good DDC with its own LPS provides, that make it sound better to me. To clarify the last sentence it is maybe good to know that I now use a LAN renderer with USB out that feed a DDC with S/PDIF out to my DAC that has GI on all its inputs.  

[PeterSt]

2 hours ago, Summit said:

was better sounding than: I2S over LAN, USB, S/PDIF and AES/EBU.

 

... which all does not exist.