Yaggy Measurements - Deconstructing ASR / Amir's Hack Job

Discussion in 'Source Measurements' started by purr1n, Jun 7, 2018.

  1. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    OTHER RESOURCES:

    12kHz J-Test. I'll run a straight 12kHz tone, close enough as the J-test is the same thing but with the LSB toggled.
    Single-ended out was used.

    Schiit Yggdrasil DAC 2 Khz span measurement .png
    Yaggy 12k zoom low-1.jpg
    1. mir is not totally wrong on the 60Hz issue, but I am getting different results. The 60Hz at -90dbFS on Amir's measurements is much higher than mine, which is a mere blip. I suspect a ground loop or contamination on Amir's setup. This kind of crap happens all the time and could be triggered by a neighbor using their AC.
    2. For the 2nd, 3rd, and 4th harmonics, my results are more or less consistent with Amir's.
    3. I'd point to point out a trick that Amir did with the visualization. He changed to Y-axis scale to make things appear to be much worse than it is. There actually is an art to visualization. The point is to make things relevant show up more and less relevant things show up less.
    4. Considering how far down (105db+) the spikes are, I really wouldn't worry about it. Could it be better, yes. Does this throw a caution flag? No.
    Here is the full spectrum (for reference):
    Yaggy 12k-1.jpg
     
    Last edited: Jun 8, 2018
    skullet, FutureFrenzy, A1Gear and 5 others like this.
  2. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    Now let's look at the same measurement using the balanced outputs of the Yaggy.

    BAL Out
    Yaggy 12k BAL-1.jpg

    Scroo you Amir! The reason I say this is because I am almost certain that Amir, given how thorough he usually is, also did a balanced output measurement, but because of his hate, he neglected to show the balanced output results on his website. This is not bias on Amir's part. This is hate. To put it bluntly, Amir is a morally bankrupt person!
     
    Last edited: Jun 8, 2018
    thoth, skullet, Mban and 16 others like this.
  3. Ice-man

    Ice-man Friend

    Friend
    Joined:
    Oct 12, 2015
    Likes Received:
    1,839
    Trophy Points:
    113
    How would you know he did a balanced measurement? Just speculating? I mean, I get it...he's a dick, plain and simple. I just wondered if you had some evidence or clue that he'd done those measurements.
     
  4. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    The Force.

    Amir is thorough. If he didn't take balanced measurement, then he is lazy at best, or opportunistic at worst (stopping the moment when bad results arrive). This is a DAC with a balanced option. Everybody who owns it knows that the SE out is not as good sounding as the balanced outs.
     
    Last edited: Jun 7, 2018
    nachocheese70 likes this.
  5. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    5,697
    Trophy Points:
    113
    Location:
    Irvine CA
    Capture.PNG
     
    MWSVette, Koth Ganesh, Kunlun and 7 others like this.
  6. Vtory

    Vtory Illogical Spock

    Friend
    Joined:
    Sep 12, 2016
    Likes Received:
    991
    Trophy Points:
    93
    Location:
    Fort worth, TX
    Marv's measurement set must be moved to a seperate and dedicated thread of Yaggy measurements..
    [​IMG]

    He had balanced results for this test. So, infer he might have other balanced results..
     
    FutureFrenzy and ultrabike like this.
  7. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    I rest my case. Can't you smell that these dude is super shady?

    Do you guys know how many instances I run into weird or oddball measurements that don't make sense, which require me to redo them under better environmental circumstances. Unless you have a faraday shield around your office like Jude, haha, measurements are hard to do.
     
    Last edited: Jun 7, 2018
  8. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    Next up, measurement of linearity.

    Schiit Yggdrasil DAC linearity Measurement.png

    My measurements are presented slightly differently, but just tilt mine clockwise and you will get the idea. Amir's scale (right side) does emphasize the level non-linearity deltas much more than mine.

    SE out
    Yaggy linearity SE-1.jpg

    BAL out
    Yaggy linearity BAL-1.jpg

    Either way, I seem to be getting 1-2 more "Amir bits" than Amir.

    These results are actually extremely impressive given that the Yaggy uses discrete parts for output, and not an opamp with like 55db feedback. The SE results are expected given that there is another summer circuit in the signal path.

    6/8/2018 UPDATE: I actually have better results now. I don't know if it was a result of moving cables or the firmware and software upgrade of the AverLAB to 2.0. Yes, not only are humans imperfect, but instrumentation is also not perfect! Will post the results later.
     
    Last edited: Jun 8, 2018
  9. Cspirou

    Cspirou They call me Sparky

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    2,658
    Trophy Points:
    113
    Location:
    Irvine,CA
    This is really why I don't trust him. His data does seem okay, but he plays around with scaling as well as selective omitting and including certain charts and datasets. His invention of Amirbits seems specifically to interpret Schiit DACs as being far worse then they really are.

    Something really odd is that he posted Bifrost Uber jitter measurements only 3 days after the founding of ASR. Only one measurement, that doesn't agree with Stereophile, and NOTHING ELSE.

    He definitely has an anti-Schiit bias that pre-dates the forum. Probably residual anger NWAVGuy has from seeing Schiit succeed despite a relay scandal with the Asgard.
     
    Last edited: Jun 7, 2018
    lm4der, MWSVette, Darren G and 3 others like this.
  10. Vtory

    Vtory Illogical Spock

    Friend
    Joined:
    Sep 12, 2016
    Likes Received:
    991
    Trophy Points:
    93
    Location:
    Fort worth, TX
    Just for comparison purpose:

    @atomicbob 's SE result on gain linearity

    [​IMG]

    BAL result

    [​IMG]

    Look at SE result.. damn close to BAL one... Did atomicbob have a unicorn version of yaggy?
     
    purr1n likes this.
  11. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    1kHz

    Schiit Yggdrasil DAC Harmonic Distortion Measurement.png

    One thing I'd like to point out is how Amir sneakily did not align the peak of the 1kHz signal at 0dbFS (he pushed it up to 7-8dbFS), thereby making the harmonic distortion spikes appear to be 7-8db higher than they normally would on a proper plot. Again, visualization tricks to make the Yaggy appear worse than it really is. Is this more malice on Amir's part?

    SE Out
    Yaggy 1kHz SE-1.jpg

    @atomicbob SE out
    [​IMG]
    BAL Out
    Yaggy 1kHz BAL-1.jpg

    @atomicbob BAL Out
    [​IMG]
     
    Last edited: Jun 8, 2018
  12. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    THD+N

    Schiit Yggdrasil DAC THD+N Measurement.png

    Notwithstanding the Y-axis tweaking, Amir's measurements are simply wrong. If I ever saw anything like the above, I would NOT publish, or if I did, I'd ask for help troubleshooting. Here are my results below. They look very different from Amir's. We are talking about HUGE 25db+ differences.

    Balanced is only slighty better than SE.

    SE Out
    Yaggy THDN SE-1.jpg

    @atomicbob SE Out
    [​IMG]

    BAL Out
    Yaggy THDN BAL-1.jpg

    @atomicbob BAL Out
    [​IMG]

    FYI, 20Hz on the SE THD+N plot is about -85db. 500Hz on the SE THD+N plot is about -93db. Thats 0.005% and 0.002% distortion respectively.

    I have no idea WTF Amir did to get those borked results. Maybe he beam beamed his hate into his measurement gear.
     
    Last edited: Jun 8, 2018
  13. dmckean44

    dmckean44 In a Sherwood S6040CP relationship

    Friend
    Joined:
    Nov 17, 2016
    Likes Received:
    904
    Trophy Points:
    93
    Location:
    San Diego
    Randos and Lurkers,

    You now have five points of measurement data for the Yggdrasil Analog 2. Atomicbob, Amir, Jude, Marv and Schiit themselves. Only Amir's measurements are b0rked as fuck, feel free to draw whatever conclusions you want.
     
    Vatnos, skullet, fianbarr and 16 others like this.
  14. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    Frequency response

    Schiit Yggdrasil DAC Frequency Response Measurement.png

    Amir's plots show a 0.4db rolloff at 20Hz, starting early at around 200Hz. My plots are below. I used 44.1kHz (confirmed by the front panel LEDs on the Yaggy). Note that I am not getting the early cutoff at the highs near 20kHz like Amir.

    SE Out
    Yaggy FR SE-1.jpg

    @atomicbob SE Out
    [​IMG]

    BAL Out
    Yaggy FR BAL-1.jpg

    @atomicbob BAL Out
    [​IMG]

    The SE results are slightly worse than the BAL results with the biggest difference in the bass. Even then, we are seeing 0.1db, maybe at most 0.15db rolloff at 20Hz, with the rolloff starting betwen 50 to 100Hz. Note that I also zoomed in on the Y-axis on my plots so small differences would be more noticeable.

    Again, I don't how what Amir is doing to get his strange results.
     
    Last edited: Jun 8, 2018
    FutureFrenzy likes this.
  15. Vtory

    Vtory Illogical Spock

    Friend
    Joined:
    Sep 12, 2016
    Likes Received:
    991
    Trophy Points:
    93
    Location:
    Fort worth, TX
    [​IMG]
     
    Thad E Ginathom and Cspirou like this.
  16. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    We all have biases. There are brands that I dislike. However, I would NEVER do a HACK job like Amir on a manufacturer that I disliked. I just couldn't because it would be WRONG. I cannot emphasize that during the MANY measurements I've taken, I've run into so much funky stuff. If I saw something off or atypical, I would never in an instant publish and make strong assumptions. Or if I did make a mistake, I'd go back and correct them. Or you guys would call the mistakes out, like in the Changstar days when I was still refining my methodologies.

    Amir is just one sick dude.
     
  17. Cspirou

    Cspirou They call me Sparky

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    2,658
    Trophy Points:
    113
    Location:
    Irvine,CA
    And Stereophile

    Edit: oh, you said Analog 2. Still he took analog 1 measurements which are still borked
     
    dmckean44 likes this.
  18. Senorx12562

    Senorx12562 Friend

    Friend
    Joined:
    Oct 26, 2015
    Likes Received:
    783
    Trophy Points:
    93
    Location:
    Limbo
    Given the re-opening of this thread, I went to the site and read the Yggy/Yaggy review along with some others, and my question is always "So what does the fucking thing sound like, dude." I read audio reviews to find out a few things. Features, ergonomics, potential problems, etc... But add all these things up, and they would maybe add up to 10%. My major interest though is sound, and how it compares to known competitive products. Amir? Crickets.
     
  19. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    33,015
    Trophy Points:
    113
    Location:
    Calabasas, CA
    Last one. -90db 1kHz sine wave.
    Schiit Yggdrasil DAC -90 db linearity Measurement.png

    This is mine, 44.1kHz. I increased gain on the inputs of the Focusrite to get better view.
    -90.33db 24bit squeezed..png
    -90.33db 24bit.png

    @atomicbob SE out
    [​IMG]

    BTW, the 11111111111111111111 to 00000000000000000000 transition glitch is gone. Of course Amir's measurements are going to look like that. He didn't take care of the 60Hz contamination in his measurement setup.
     
    Last edited: Jun 8, 2018
    astromo, Josh83, LetMeBeFrank and 6 others like this.
  20. Vtory

    Vtory Illogical Spock

    Friend
    Joined:
    Sep 12, 2016
    Likes Received:
    991
    Trophy Points:
    93
    Location:
    Fort worth, TX
    @purr1n Is that an unbalanced result?
     

Share This Page