USB Bridge

Discussion in 'DACs' started by stoutblock, Dec 9, 2017.

  1. gvl

    gvl Well-Known Member

    Messages:
    973
    I mean not generations, but say Gustard U12 is $170 on Amazon with Prime shipping and I believe it comes with proper drivers.
     

     

    Please register to disable this ad.

  2. enginedr

    enginedr Well-Known Member

    Messages:
    837
    Location:
    nyc
    There is another issue of computer noise that has not been addressed in this thread . To isolate the DAC from the computer you need to use a AOIP or a optical Toslink .
    One good solution for USB is the PS audio LANRover .
     
    nyhifihead likes this.
  3. for_p1

    for_p1 Addicted Member

    Messages:
    5,469
    Location:
    North TX
    If you use SPDIF, you need to do clock recovery all over again. Thus it almost does not matter how good is clock in USB interface. The same thing is about HDMI audio de-embedders. Both only work well if DAC has very good clock recovery circuit or ASRC along with independent clock. It is somewhat easier to achieve if all components use a single word clock. In this case there is no need for buffers, just good PLL to reduce signal front incertaintly (commonly called jitter). My DAC uses ASRC and I saw huge difference in distortion between it being on or off. I can imagine that this can be heard too. But if DAC circuit is bad, nothing really helps it. Do not waste your money and get a DAC that has no design error. They are affordable enough today.
     
  4. stoutblock

    stoutblock If it sounds good, it is good... Subscriber

    Messages:
    9,255
    Location:
    Northwest USA
    The U12 is an established and proven product but has the previous generation XMOS (U8) which is fine but has about half the processing capability of the newer XU208. I’m also not sure it has the latest clock but I know they are totally acceptable. Depending on what USB interface you have now, I think you can’t go wrong with the Gustard U12.
     
    nyhifihead likes this.
  5. stoutblock

    stoutblock If it sounds good, it is good... Subscriber

    Messages:
    9,255
    Location:
    Northwest USA
    Absolutly agree. In my case my main DAC is a Modwright Elyse which I absolutly love the sound it provides for my digital files in my system. Maybe it is partly due to the great anolog output design, maybe it is the PS design but I’m sure it is also due to the digital circuit design utilizing the PCM1794. The PCM1794 is strange choice as it is H/W controlled which is not the normal choice these days but does give the designer many controls not common with other chips. It also means it can’t process I2s or DSD but I really don’t care as I really like this DAC and 24/192 is fine with me (actually 24/96 is fine with me). So even though I am still relying on the internal clock because I’m interfacing with SPDIF I still hear a substantial improvement with this new USB interface. The internal USB in the Elyse is an Optoma NuForce which is not real common, and uses unique drivers but has a decent reputation. The new Breeze sounds better as I described at the beginning of this thread. It is not subtle.

    Now if I can just get Dan Wright to design and build an Elyse II with all the latest developments. I’m sure he will once the dust settles just a little. One thing I hope he does not touch is his analog output circuit of the Elyse as it makes beautiful music!
     
  6. gvl

    gvl Well-Known Member

    Messages:
    973
    I have none, using Toslink out of a, vintage now, Creative SB notebook card which I'm not very confident in, so any more or less decent USB DDC out to coax should be an improvement. The DAC in question has no USB. I don't do hires or DSD, so even previous XMOS should be adequate I would think.
     

     

    Please register to disable this ad.

  7. stoutblock

    stoutblock If it sounds good, it is good... Subscriber

    Messages:
    9,255
    Location:
    Northwest USA
    What external DAC do you have? The reason I ask is the money may be better spent on the DAC itself rather than the USB interface.
     
  8. gvl

    gvl Well-Known Member

    Messages:
    973
    It's a Parasound D/AC-1000, just bought it to experiment with so just need an interface. Not to mention I would probably have to spend in 4 digits to get to the same level of performance. I have other more modern DACs with USB, but I've been curious about these vintage multibit DACs, and it does not disappoint.
     
    Last edited: Dec 12, 2017
  9. botrytis

    botrytis Trying not to be a Small Speaker Hoarder Subscriber

    Messages:
    29,302
    Location:
    In FLUX currently......
    Bull. A good computer PS can takes care of this issue. I have seen people talk about this but no one has ever measured or even shown graphs were this exists. TOSLINK has it's own flaws to deal with. So don't say it is a good solution. You are switching on problem for another. Also, TOSLINK is useless above 96/24 resolution.

    The same noise problems happen with any system. I mean people are buying streamers which ARE PC's. They don't use any exotic parts and many use spinning HDs (Oh my gosh, the motor noise but since it is an audiophile system it should different - NOT). They just cost a shit ton of money because you are paying for the term 'audiophile'. I can give an example from the sciences. I wanted to buy a timer, for my lab and went the Fisher catalog and saw a 3 channel timer for $125 USD. I went to the local cooking store and bought the exact same model timer (w/o the Fisher logo) for $25 USD.
     
    Last edited: Dec 12, 2017
  10. botrytis

    botrytis Trying not to be a Small Speaker Hoarder Subscriber

    Messages:
    29,302
    Location:
    In FLUX currently......
    People think digital music is different than a digital signal. If it affects one it affects all. Let's not put digital audio as some magic fairy that is affected by things the regular digital signals are not. There is no difference. If you minimize noise in your computer by buying the best parts possible, it should also work for digital audio.

    Most people don't realize that FLAC is based on the *.jpg system for pictures. It is a similar type of lossless compression system.
     
    Last edited: Dec 12, 2017
  11. gvl

    gvl Well-Known Member

    Messages:
    973
    Wrong. Classic JPEG is lossy, FLAC uses a completely different compression algorithm.
     
    ben_ and E.Auer like this.

     

    Please register to disable this ad.

  12. botrytis

    botrytis Trying not to be a Small Speaker Hoarder Subscriber

    Messages:
    29,302
    Location:
    In FLUX currently......
    If you actually read the FLAC papers, the first ones to come out on it, that is where the idea for FAC came from but instead of losing the data - one could make a 'Container' type system just like *.zip, etc. use.
     
  13. gvl

    gvl Well-Known Member

    Messages:
    973
    I'm sorry, but I don't see your point. Container is just that - a container, think of a cardboard box with Christmas decorations kept in a closet with a "Christmas decorations" label on it. Same idea.
     
  14. for_p1

    for_p1 Addicted Member

    Messages:
    5,469
    Location:
    North TX
    FLAC is just like ZIP. You can actually take text file and compress it with FLAC. After you decompress it - it will be the same text file. In that regard it works like ZIP, RAR or any other lossless compression tool. The only difference is that algorithm is optimized for audio data allowing a bit better compression ratio than plain ZIP. That is why when you use FLAC to compress DoP encoded DSD file, after expanding it is successfully played by DSD DAC. No bits are lost in the process.
     
    ben_ likes this.
  15. gvl

    gvl Well-Known Member

    Messages:
    973
    Thanks, I'm well aware of how compression works. It is just using a JPEG analogy was wrong and then that container comment just seemed random.
     
    olderroust likes this.
  16. nyhifihead

    nyhifihead Well-Known Member

    Messages:
    748
    Location:
    long island NY

    the new XMOS u208 is superior. I use a gustard u12 and I recently had a singxer su-1 u208 based DDC in my system it improved the sonics of both my RDC 7.1 and schiit yggdrisil

    gvl I have heard great thing about that parasound dac- I believed its a pcm63k which is an impressive chip. using a DDC out to digital coax spdif will give you a great increase in fidelity compared to optical out of an on board or PCI sound card.

    many flock to the many new USB DACs because of the ease, but there is something to be said for the value of using older spdif/aes based DACs with an external DDC, dollar for dollar I think a used parasound/theta etc 90s DAC paired with a u208 DDC would be killer.

    a friend of mine uses a CAL tube DAC with a gustard u12. the performance was much better with the more precise gustard than spdif out of onboard.



    I too believe that USB interfaces only increase in quality with new chipsets & technology

    the actually DAC chips themselves don't follow this trend. newer is not always better. its all in the design, you can still get superior sound out of near 20 year old chips pcm63/tda15xx etc
     

     

    Please register to disable this ad.

  17. gvl

    gvl Well-Known Member

    Messages:
    973
    I had it for a couple of days and I think it is fantastic. And even if you consider the cost of the 96kHz/24bit upgrade kit and a decent DDC it is still less expensive than say a comparable Schiit multibit DAC, even used. I did try it with a coax out of my CDP SPDIF, and while not a night and day difference it sounded better than optical out from the sound card when listening to a FLAC rip of the same CD, but probably that will be the extent of improvement with a good DDC.
     
    nyhifihead likes this.
  18. for_p1

    for_p1 Addicted Member

    Messages:
    5,469
    Location:
    North TX
    Do you know how XMOS became a standard in USB audio? When XMOS made their new multicore DSP, they decided that it is necessary to give an example of its use for interface data processing. For that purpose their developers wrote USB client code running on that DSP. And they make it open source and available for free. That is why everyone jumped in and started using this software as base of their own design running on XMOS processor. Before that manufacturers were just too lazy to write USB client code. There is nothing in XMOS chips that is not available in other processors for embedded systems. And some DAC makers successfully use TI or other chips to process audio data.
     
  19. gvl

    gvl Well-Known Member

    Messages:
    973
    I admit I've never looked into the details of XMOS designs, but a more powerful processing engine/more memory can't be a bad thing as designers face fewer constraints when writing processing code and can deal with some situations that can arise without or making fewer compromises. That is probably where the improvements come from, if there are any.
     
  20. nyhifihead

    nyhifihead Well-Known Member

    Messages:
    748
    Location:
    long island NY
    for_p1 thanks for the info, it makes a lot of sense since development r&d is the major barrier to entry

    the Chinese copy & pirate everything- until I read XMOS stereo was free I had assumed the supplied drivers were cracked versions of xmos commercial software

    I don't believe XMOS is truly open source- the driver they supply is actually for "evaluation" and only supports stereo. multichannel is definitely proprietary.

    some of the Chinese mfg gustard, breeze, etc actually send modified XMOS unsigned drivers with their products, possibly to circumvent some built in limitations, or to get around the blatant reselling of "evaluation" software commercially


    as far as the sonic performance I can't tell you whether its the topology/component/PS change between the u8 and u208 based DDC or its the actual chip itself

    I know that the u208 based DDCs sounded better.
     
    stoutblock likes this.

Share This Page