USB Nervosa Thread Decrapifiers, pro interfaces, and bears oh my

Discussion in 'Digital: DACs, USB converters, decrapifiers' started by zerodeefex, Sep 28, 2015.

  1. nedifer

    nedifer Facebook Friend

    Contributor
    Joined:
    Oct 10, 2015
    Likes Received:
    124
    Trophy Points:
    33
    Location:
    San Francisco Bay Area
    Hi WNovizar,

    The MCLK/WCLK output is for clocking purposes, not music data. You would use it with another device that has a clock input in order to synchronize that device with the Singxer's internal clocks (while relaying the digital music stream via one of the other interfaces).

    Unfortunately, the Gungnir Multibit doesn't support MCLK or WCLK input/syncing (but does have good internal clocks itself).
     
  2. johnjen

    johnjen Doesn’t want to be here but keeps posting anyways

    Pyrate
    Joined:
    Sep 29, 2015
    Likes Received:
    505
    Trophy Points:
    93
    Location:
    Well Grounded
    Since the Gungnir Multibit has no Word Clock input, there is no direct way to connect up the SU-1's WCLK/MCLK output to it.

    -er, ah, what he said ^.

    JJ
     
  3. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Interesting to find the Singxer SU-1 had randomly disappeared from my connected USB devices list today. Disconnecting and reconnecting it's USB cable fixed that, but that's the second time it's done that in a week. It only seems to happen if it's left idle for a while; typically through at least one computer sleep-cycle. Every other connected audio device was still correctly enumerated and recognized, but the SU-1 had vanished.

    This is quite annoying, as it renders it pretty much useless for any kind of distributed or remote DAC connection. I'm not setting up a system that requires periodic, random, device/connection resetting. The last thing I want to do when I sit down to listen to music is to have to go troubleshoot something first, even if it is just cable disconnect/reconnect.

    SU-1 firmware is v2.02

    OS is macOS 10.12.1

    In very stark contrast, even while it has it's own issues, my Aries hasn't been rebooted in months. And my sonicOrbiter SE has been running continuously since I bought it, excepting having to turn it off to get a microRendu configured on the same network.
     
  4. nedifer

    nedifer Facebook Friend

    Contributor
    Joined:
    Oct 10, 2015
    Likes Received:
    124
    Trophy Points:
    33
    Location:
    San Francisco Bay Area
    Are you using a laptop? And, if so, have you tried hibernating it rather than sleeping it?

    Hibernating essentially turns the computer off. It store the system's running configuration to disk before shutting down and restores to it when you power on the computer.

    The advantage, in this case, for hibernation is that it will force your computer to enumerate all of the devices connected to it when it resumes. Computers resuming from a sleep state are more prone to not properly re-enumerate all of the devices connected to them.

    It might be worth a try.
     
  5. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    No, it's not a laptop.

    Since every other USB device I have, including other DACs, other DDCs, drives, cameras, and so on was still present, I'm entirely sure it's something "unique" to either the XMOS interface being used, or the implementation of it. Not the first time I've seen it with XMOS-based devices and not something I care to troubleshoot - just thought it was interesting.

    Also, re-enumerating/re-setting the USB bus in code on the computer side did NOT restore the SU-1 to visibility, so it's clearly a device-side issue.

    Appreciate your trying to help, but I was mostly commenting on it being annoying rather than looking for a solution ;)
     
  6. Madaboutaudio

    Madaboutaudio Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    545
    Trophy Points:
    93
    Location:
    Singapore
    @Torq

    There's a new firmware released for the Singxer:
    http://download.shenzhenaudio.com/drivers/DFU_F1_V212_20161108.rar
    http://www.shenzhenaudio.com/singxe...-xu208-chip-high-end-u8-upgraded-version.html

    Even Audio Electronics like DACs do require reboots* as well. I think there might be some firmware bugs or ageing related electrical issues with the Xmos controller or Singxer's ASIC that require rebooting to restore functionality.

    Just for example I had an old Apple Extreme(N) Router that is so stable that required zero reboots for months or years. It was running some really solid real time kernel or something like that. But the Apple router still performs better with regular rebooting(it gets slower over time if you don't reboot).

    Maybe Aries has solid hardware quality and a solid operating system(i.e. Linux) which doesn't require much rebooting. But I do think you should reboot your Aries once in awhile to so as to get the tip top responsiveness as any kind of long term memory fragmentation is something only a reboot can cure.

    *
    http://www.head-fi.org/t/815368/hap...-introducing-modi-multibit/1380#post_12933242
     
  7. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    While I certainly won't argue that various consumer-electronics devices require (or benefit from) rebooting from time to time, it needing to be done, randomly, between once a day and twice a week, is not something I can consider acceptable if I was to attempt to use the SU-1 with a DAC that wasn't at arm's reach.

    That's fine ... while the Holo Audio Spring DAC sounds best via the SU-1/I2S, my Yggdrasil sounds better via either my Aries or RedNet 3 and AES (vs. the SU-1/AES), and since those three devices haven't required reboots in months (I don't think my Yggdrasil has EVER been rebooted ... thanks to my nice hospital-grade whole-house generator and ATS w/ beefy online UPS), I'll just stick with them - so the SU-1 being minimally flaky isn't a big issue ... I just thought it was interesting.

    As far I can tell, I'm on the latest SU-1 firmware.

    But, again, not a trouble-shooting exercise ... I'll only use the SU-1 in my office ... I have other, more stable, options for the rest of the house.
     
  8. bazelio

    bazelio Friend

    Pyrate
    Joined:
    Nov 16, 2015
    Likes Received:
    3,417
    Trophy Points:
    113
    Devices designed by Kingwa seem to have more quirks (and reboots required) than most, though.... including but not limited to taking care not to touch their volume knobs after walking across the carpet in your socks.
     
  9. Madaboutaudio

    Madaboutaudio Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    545
    Trophy Points:
    93
    Location:
    Singapore
    It's not just Audio-gd. Ibasso, Fiio and almost all chi-fi has some kind of software/hardware/quality issues.

    My old Ibasso DX100 had a software/hardware bug which for no rime or reason will cause the volume to jump to 100% volume instantaneously and also it continuously drains battery rapidly even on standby.
     
  10. Ph0n6

    Ph0n6 New

    Joined:
    Dec 6, 2015
    Likes Received:
    3
    Trophy Points:
    3
    Hi friends. Has anyone tried out the new Ifi iSilencer (it supposed to reduce EMI from USB I think?). I'm not too fond of using a powered hub so I do wonder does this passive solution works?
     
  11. Xecuter

    Xecuter Brush and floss your amp twice a day

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    2,016
    Trophy Points:
    93
    Location:
    Australia
    Hi Guys.
    I was wondering of you guys have heard or had a play with the aria: Warning: 6moons review: http://www.6moons.com/audioreviews2/digibit/1.html
    A guy I know bought 3 of these insanely over priced players and is willing to sell me one at well below 50% RRP.
    I am actually tempted as my PC has long been something I have considered a bottle neck in my system, even with the lynx AES card.
     
  12. Nanekiu

    Nanekiu New

    Joined:
    Jan 3, 2016
    Likes Received:
    15
    Trophy Points:
    8
    Location:
    Trondheim Norway
    You know with UAC3 coming and all ( Looking at you M$) I wonder if bulk mode transfer would be feasible with error correction, as bandwidth is more than guaranteed with usb3.
     
  13. msommers

    msommers High on Epipens

    Pyrate
    Joined:
    Nov 6, 2016
    Likes Received:
    2,750
    Trophy Points:
    113
    Location:
    Calgary, Alberta
    Home Page:
    Since I plan on keeping my Gungnir Multibit for awhile, would it make more sense/sound better to grab an RME with bnc-out vs a lynx aes16e/e22 and using a transformer from aes to bnc?

    If not, using a decent(ish?) aes to bnc transformer from Neutrik is the best avenue?

    Fortunately my motherboard still accepts PCI so might opt for the aes16 anyways if I can find a deal...Why did I find this forum, it's be an expensive learning experience haha...
     
    Last edited: Nov 21, 2016
  14. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    It's possible, but not necessarily that useful; bit-level errors with USB audio are not that much of an issue in practice (unless using out-of-spec cables and long runs). The biggest issue with USB 2.0 Async Audio is noisy GND and Vcc (+5v) lines, which are vastly more likely to affect analog elements in the DAC than the bit-level data.

    Out of interest, an AOIP/NAI box I'm working on at the moment does support a bulk-transfer mode input (it'd need a special player to work usefully that way, but it mounts as a Storage Class device and you can drag/drop files to it). It's there to allow a "guaranteed bit-perfect delivery" option, more for debugging than anything, but in practice it is pretty much unnecessary except, possibly, to address "peace of mind" concerns.

    Given the existing USB 3.0 EMI/RFI issues, I'm not sure I'd want it anywhere near a device that had analog audio circuitry in it. Sure, you can shield your way around it, to a certain extent, but it's still a bit ugly ...
     
  15. Nanekiu

    Nanekiu New

    Joined:
    Jan 3, 2016
    Likes Received:
    15
    Trophy Points:
    8
    Location:
    Trondheim Norway
    The hardest would probably be isolating the ground from the USB, as for 5v ideally the USB input should have a dedicated PSU internally in the DAC and probably a split USB cable with separated power data legs for handshake.

    i believe a excellently implemented USB input is better than a good SPDIF/Toslink,
     
  16. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    I built just such a cable (excuse the ugly, prototypical, aesthetics). You can give that cable an totally isolated PSU that has no connection to the USB source. And it further filters the power lines just for good measure.

    Personally, I've yet to hear a USB input beat the alternatives on any DAC I've auditioned to date. I suppose it could happen, I've just not come across it yet. Which makes me wonder how many DACs are actually implementing their internal USB interface in a manner than then feeds their S/PDIF input rather than going straight to I2S (which is what most DAC chips speak, natively). And if they do that then their ultimate performance is limited by that S/PDIF implementation rather than the USB input feeding it.
     
  17. Malabargold

    Malabargold Flipper

    Joined:
    Apr 10, 2016
    Likes Received:
    120
    Trophy Points:
    43
    Here's a post from Baldr himself- It would be interesting to see Schiit's take on a higher end Decrapifier. Maybe they have such a product already in development.


    "On USB processors, reclockers, cleaners, jitter reducers, pre-conditioners, prayer cloths, etc. (Oh yeah, there's also this thing called a Wyrd.)

    Why? What for? What are they supposed to do? Good questions. Does the whole thing center around marrying off your beautiful and virginal D/A converter to some horribly diseased computer USB port? Not that simple, but as my Dad used to say, "If you lie down with a dog, you get fleas."

    Some thirty years ago, as the D/A converter planet was cooling, the only way to get signals on board was S/PDIF or the amazing similar AES-EBU. Period. There were coax and TOS link variants (Which do offer ground isolation but suffer from horrible jitter. Oh, they are cheap and are convenient profit centers.)

    In that far away universe, computers lived far away from audio systems as they spewed rf energy from mutiple clocks which interfered with rf based devices of the time (TVs and tuners). Wisdom proper to the time separated high end audio systems and computers as much as possible. If they were to close together, audiophile guests would point and snicker.

    Fast forward to 6 years ago as I began my first D/A design for Schiit (which utlimately became the Bifrost). In a desktop audio (headphone) system world, USB sockets dominated in a new era when the audio media lived on or was streamed through the computer. Old notions of quality were lost; MP3 was everywhere. The digital source bar was so low it would trip a grasshopper. It sounded like ass.

    I was amazed that more users did not acquire/demand spdif connectivity for their computer based audio. It became obvious to me that the convenience of USB dominated. Although the state of its art has improved, it still has a ways to go to trump the best of S/PDIF.

    So it is no surprise to me that various products have appeared on the market to fix or enhance USB. Some even purport to use USB sockets to fix other parameters of the DAC itself. Some categories are below:

    USB Filters. These tend to be the least expensive USB addition. They can be passive ($) or active (more $$). Such filters are built in to all Schiit USB nputs on all digital products.

    Devices to fix USB power. This is one of the functions of the Schiit Wyrd, the only product I will comment on. The last six years in the evolution of computers (servers) have headed them in an ever smaller direction. Along with this comes less power available from the USB sockets to power the connected USB devices themselves. The Schiit Wyrd has a separate power supply designed to drive the most power hungry USB input circuits. There is also a USB repeater function which separates the input and output USB data wires. Many users report they hear an improvement when using the Wyrd, even if the extra USB power is not needed. The Schiit Wyrd is $99.

    USB Isolators. Several design choices, whether it be optical, transformer, or inductive (cheapest). This is good engineering practice if properly implemented. The idea is that the computer's trash and slop is not transferred in real time to D/A converter. This is tricky to implement as most isolators are jitter prone.

    USB Clocks. This is the least understood and farthest removed from where it really counts enhancement. This is not to say that clocks are not important. At the end of the day, however, they only matter at the input to the DAC chip or array, with the most important by far being the Word Clock or Convert Command. In all Schiit gear, there are at least two and probably three major stages of clocks prior to the DAC chip, where our adaptaclock and clock cleaners live. Fixing bad clocks ahead of all of that is not anywhere as effective as it is at the DAC and therefore money not totally well. That said, good clocks upstream can't hurt, and can provide an incremental improvement.

    These features are what I have seen a variety of products implement in the wild. There are even those which have implemented more than one of them, but they usually sell for prices approaching a kilobuck. A product such as this should cost far less and would probably be fun to design. Who knows? Maybe. Probably."
     
  18. msommers

    msommers High on Epipens

    Pyrate
    Joined:
    Nov 6, 2016
    Likes Received:
    2,750
    Trophy Points:
    113
    Location:
    Calgary, Alberta
    Home Page:
    Found a AES16 PCI on eBay with the 1604 cable also which all added up to a reasonable ~$300 CDN shipped to my door. Picked up a bnc cable from BJC and a Neutrik 110->75 transformer, all going into the bnc of the Gungnir Multibit. The Asus P8P67 motherboard is new-ish but apparently old enough to still take PCI (thank God, the e was too much to justify).

    Will report back what differences I find compared to the Motherboard USB 2.0->Audio Quest Carbon->Gungnir Multibit.
     
  19. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,461
    Trophy Points:
    113
    Location:
    Winnipeg

    Followup from Mike...

     
  20. Madra

    Madra Acquaintance

    Joined:
    Oct 18, 2015
    Likes Received:
    26
    Trophy Points:
    18
    @Torq , given the extensive DAC and transport search you have gone through, did you happen to audition the Berkely USB DDC?
    I currently have a friend's unit on loan (having some difficulty returning it ;)) and compared to my Audiophilleo 2, it is a significant improvement. Higher in resolution and overall realism, It has undoubtedly enhanced my listening enjoyment. I am also interested in seeing how it compares to the much cheaper SU-1.
     

Share This Page