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

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

  1. Luckbad

    Luckbad Traded in a unicorn for a Corolla

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    3,408
    Trophy Points:
    113
    Location:
    Holly Springs, NC
    If you bug Alex about the UpTone Regen via email, you'll discover he's willing to refund you if you don't like it minus shipping. To me, that makes it worth trying.
     
  2. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
  3. firev1

    firev1 Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    44
    Trophy Points:
    18
    Location:
    Singapore
    Still can help the usb dac if system injects noise via the HDMI link etc etc, it happens

    I tested the the wyrd with the Stoner Acoustics UD120 and though not the ideal dac to test with, I really measured lower spuraei on the output. Its an async dac though not the best performer in jitter. Also I'm surprised to hear that people know Aikatsu :D
     
    Last edited: Oct 7, 2015
  4. Judeus

    Judeus Facebook Friend

    Banned
    Joined:
    Sep 28, 2015
    Likes Received:
    202
    Trophy Points:
    43
    Location:
    summit-fi purgatory
    The thread on what's best forum is pretty hilarious.
     
  5. AustinValentine

    AustinValentine Friend

    Pyrate
    Joined:
    Sep 26, 2015
    Likes Received:
    835
    Trophy Points:
    93
    Location:
    West Chicago Burbs
    Hilarious is one way to put it. I just read through all 53 pages of that shit show.

    It seems like the only person who wants to understand the 8k spike and related spuria is Jkenny. His repeated requests, to my mind, don't seem all that unreasonable...especially in light of Amir & BE718's repeated assertion (to paraphrase) that the Regen either doesn't do anything or makes things worse. For...like nearly 40 pages, Jkenny has been asking for:

    1. Amir to reveal his FFT settings;
    2. For Amir (and, afterward, BE718) to try a 11k test signal instead of a 12k one, to see what the result is.

    He's grabbed onto one detail from Amir and BE718's graphs: that the spurs appear every 125uS. For him, that points to the USB receiver hardware logic and it's effect on the ground noise (post #296), "Hence we see the 8KHz spuria popping up in the FFT - it's an event which happens every 125uS i.e the handling of the USB microframes in the USB receiver hardware logic & it's resultant effect on the ground noise is apparently what's being seen in the FFT plot."

    Jkenny's reason for wanting the remeasure seems pretty sound (post #307):

    "Another question that these measurements have given rise to - why is this 8KHz noise spike appearing with the Regen but not without?

    Is it a case that the Regen's regenerated USB signal's microframe timing is now more accurate & that this now shows on an FFT? Remember an FFT emphasises/magnifies signals which occur in the same frequency - the tighter the frequency the sharper the spike shown on an FFT. If there is a less accurate microframe timing in the USB signal prior to the Regen & it drifts around then this won't be picked up as noticeably in an FFT - could well be not visible at all in an FFT, depending on the drift of the timing."

    From what I see, Jkenny's hypothesis is that the Regen is showing something rather than causing something. (And JohnW's theory about the spurs is that they're caused by the SMPS. As far as I can tell, both Amir and BE718 both used the stock SMPS for the unit measurements.) Jkenny is arguing contra Amir and BE718's assertion that the Regen is causing a problem (BE718, post #353), "Thats the point, they are not ineffective under normal conditions, ie USB straight in. Its the regen that causes a problem, whatever that might be" (and Amir, post #375) "What I am doing is using my knowledge and equipment to capture problems I see as someone verifying performance of the product."

    But instead of addressing what's going on with the Regen and the measurement systems, we get 25 pages of Jkenny arguing and Amir & BE718 acting like dicks to Jkenny. That's an hour of my f'ing life that I'll never get back.
     
  6. Luckbad

    Luckbad Traded in a unicorn for a Corolla

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    3,408
    Trophy Points:
    113
    Location:
    Holly Springs, NC
    Thanks for the summary so I don't have to read it all.
     
  7. wnmnkh

    wnmnkh Friend

    Pyrate
    Joined:
    Sep 29, 2015
    Likes Received:
    80
    Trophy Points:
    28
    Wow, thank you for the summary. I just could not dare to read all of that pages.
     
  8. Azteca

    Azteca Friend

    Pyrate
    Joined:
    Sep 26, 2015
    Likes Received:
    1,581
    Trophy Points:
    93
    Location:
    Cincinnati, Ohio
    I posted this in the RMAF thread but:
    Motherboard from scratch + super simple OS + LPS could equal one pure signal.
     
  9. kn19h7

    kn19h7 New

    Joined:
    Sep 27, 2015
    Likes Received:
    8
    Trophy Points:
    3
    Location:
    Singapore
    Oops I missed description for that gif at last..
    Looks like the spikes with UD120 get obviously lesser/lower, but the "skirt" kinda gets a bit dirtier, or maybe its the scaling
    So do I :D
     
  10. Judeus

    Judeus Facebook Friend

    Banned
    Joined:
    Sep 28, 2015
    Likes Received:
    202
    Trophy Points:
    43
    Location:
    summit-fi purgatory
  11. MisterRogers

    MisterRogers Ethernet Nervosa

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    2,063
    Trophy Points:
    93
    Location:
    Aurora, CO
    Home Page:
    Some information on the USB REGEN; Ordered one awhile back, I'm waiting for the October batch to take delivery. In anticipation, I built and tuned a Reflektor-D (Cree diodes, Polymer smoothing caps, Mills resistors, you know - diy bling). At RMAF, Jexby offered to loan me his to check it out. I first said no (patience being a virtue and all), but then thought - 'who am I kidding?' :) It landed Sat, and I've been listening to it with my Yggdrasil, Uber7, HE1K system. It took the place of a Wyrd - which I found to be a very slight improvement with Yggdrasil (but generally a more noticeable improvement with 'lesser' USB receivers).

    Wow. At once - more detailed and more analog. The improvement over Wyrd feeding Yggdrasil is not subtle. When I get to it, I'll try the supplied 'switcher' supply, see if it's still an improvement over Wyrd - and how much. I'm using the adaptor to hang REGEN off the back of Yggdrasil. I'll even try Wyrd 'upstream' of REGEN - for schiits and giggles.

    I wish Wyrd accepted DC in; I suspect we'd see an improvement with it too if we threw an exotic supply at it. Certainly could hack it I guess.

    Bottom line: REGEN with an exotic supply known to excel feeding digital circuits is definitely worthwhile. It's now part of my reference chain.
     
  12. AustinValentine

    AustinValentine Friend

    Pyrate
    Joined:
    Sep 26, 2015
    Likes Received:
    835
    Trophy Points:
    93
    Location:
    West Chicago Burbs
    It's a really substantial difference - more audible than a lot of signal chain shifts. The differences between the Wyrd and Regen, from every measurement I've seen, should be below the threshold of audibility. This is a situation where measurements and subjective listening do not jibe.

    That said: I'm not sure if I prefer the Regen over the Wyrd with the Gungnir Multibit or not. TMoney called it "slightly warmer and punchier than the sound of the DACs running stock." Drfindley said it made the DAC "less incisive and took some bite off the attack, but it seemed to give things more resonance." I agree with both of these sets of impressions, but the biggest difference to my ears is what TomNC and Bill-P commented on,

    "And yeah, TomNC, I've also noticed that about my Gungnir Multibit up to this point. It's smooth, nice, clean and clear, but there is just no well-defined edge or border with some recordings"​

    I feel like the Regen can exacerbate this part of the Gungnir Multibit's character. To me, the Wyrd tightens the edges and the Regen diffuses them. With the HD650, Paradox, and Oppo PM-1, I vastly prefer the Wyrd in chain (Wyrd > Gungnir Multibit > Valhalla 2/Krell Klone). With the Anax-modded HD800, it's the Regen all the way.
     
    Last edited: Aug 26, 2016
  13. MisterRogers

    MisterRogers Ethernet Nervosa

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    2,063
    Trophy Points:
    93
    Location:
    Aurora, CO
    Home Page:
    Yup. It sometimes doesn't bear out in measurements, but my experience has been that digital circuits are more sensitive than analog to PS design / components. I can tell you this; REGEN with a Reflektor-D is less 'digital' than a stock Wyrd - but more detailed; edges seem very 'naturally' defined. It really is Apples to Oranges comparing these devices with stock power vs. dedicated / purpose built.
     
  14. 7Ryder

    7Ryder Friend

    Pyrate
    Joined:
    Sep 30, 2015
    Likes Received:
    4
    Trophy Points:
    3
    I also bought Uptone's JS-2 linear power supply and, yes, the REGEN scales nicely with an improved power supply. I bought the JS-2 to use with my my dedicated audio PC...otherwise it's really overkill to spend that much on a power supply (IMO) to use on a $175 tweak. It will be interesting to see how the new power supply they are working on will improve the REGEN...it will definitely be much cheaper than the JS-2
     
  15. MisterRogers

    MisterRogers Ethernet Nervosa

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    2,063
    Trophy Points:
    93
    Location:
    Aurora, CO
    Home Page:
    Having logged quite a bit more time with REGEN, Wyrd->REGEN, REGEN-> WYRD...

    I've settled on REGEN hanging off the back of Yggdrasil, decent USB cable, and Reflekor->D 7V PS. In listening and considering the observations above...

    It seems that with Wyrd in the chain, the depth of the sound stage is flattened a bit; background voices / instruments are brought forward some. With REGEN only there's more depth. This doesn't (IMO) affect the delineation of instruments / voices at all. I have a 'K2 HD' of Corazon. In 'Mal Bicho', there's a off-beat vocal between the lead. With Wyrd, it's tougher to isolate. With REGEN, it's much easier.

    I built a second PS for my inbound 2nd REGEN yesterday. The R-D PS seems extremely well suited to powering this thing. I've offered to build Jexby one of these... If Uptone doesn't come out with their dedicated REGEN PS, we could probably put together a small group buy - if we have enough members interested.
     
    Last edited: Oct 18, 2015
  16. JewBear

    JewBear Almost "Made"

    Joined:
    Sep 28, 2015
    Likes Received:
    270
    Trophy Points:
    63
    Can someone actually explain how the USB cleaner is supposed to improve sound? I can't think of one technical reason it should as all the output stages on fancier dacs are powered by a separate power supply anyways. Jitter is already a non issue with decent DACs, so I'm really struggling to see how this isn't all placebo.
     
  17. velvetx

    velvetx Gear Master West/Vendor Spotlight Moderator

    Staff Member Pyrate
    Joined:
    Sep 26, 2015
    Likes Received:
    1,067
    Trophy Points:
    113
    Location:
    Bay Area, CA
    I think these are just one of those things you are going to have to listen to to see if it's right for you. Personally I tried the iUSB 3.0 and the unfiltered USB with a dac/amp sounded better. I don't want to deter anyone from getting a usb filter but to me it didn't make the sound impact I was expecting or even a positive one. Again like cables I suggest you listen subjectively then make a decision on whether or not you need one.

    Just a little of what I heard from the iUSB 3.0:

    It constricted the sound stage and to me really ruined the music. It was almost as polarizing as going from an open backed headphone and making it closed back. That is the difference I heard and it wasn't good at all to me. It wasn't a "cleaner" sound it was less refined and actually made the tracks I listened to worse.
     
  18. MisterRogers

    MisterRogers Ethernet Nervosa

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    2,063
    Trophy Points:
    93
    Location:
    Aurora, CO
    Home Page:
    Hmmm, haven't read up on the iUSB 3.0 - is it a filter? Or by 'filter' do you mean the business of providing cleaner 5V, along with re-clocked or regenerated packets?
     
  19. Judeus

    Judeus Facebook Friend

    Banned
    Joined:
    Sep 28, 2015
    Likes Received:
    202
    Trophy Points:
    43
    Location:
    summit-fi purgatory
    I really don't see how that is possible at all unless the iusb 3.0 really fucks with something
     
  20. velvetx

    velvetx Gear Master West/Vendor Spotlight Moderator

    Staff Member Pyrate
    Joined:
    Sep 26, 2015
    Likes Received:
    1,067
    Trophy Points:
    113
    Location:
    Bay Area, CA
    Yes, I call it filtering. Sorry should have been more specific
     

Share This Page