Modi 2U and D30 DAC: Deconstructing Amir's Hack Job Part II

Discussion in 'Source Measurements' started by purr1n, Jun 8, 2018.

  1. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,623
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I couldn't but help notice that Amir decided to do a Modi 2U vs D30 comparison after I posted my D30 measurements, said it sounded like shit (level matched with other DACs), and threw in the Modi 2U as a reference. Going to do a reality check of Amir's measurements.

    Let's start with linearity. Amir measurements here.
    Topping D30 versus Schiit Modi 2 Uber Linearity Measurement.png

    Here are my results. Note this time time around, I did a zoom in from -140 to -90db to better see the differences. I also drew a straight line to denote linear levels behavior. Well, first thing's first. It would appear that Amir has not calibrated his AP correctly since I am getting vastly superior results to him with my AverLAB (who no one has ever heard of, so it can't be good).

    Here is the Schiit Modi 2U
    Modi 2U linearity-1.jpg

    Here is the Topping D30
    D30 linearity-1.jpg

    Looking at the above results, it would appear that the Modi 2U is exhibits linear behavior down to -104db whereas the Topping D30 starts to deviate from being linear at -99db. This is a almost 1-bit difference! The Topping D30 advertises itself as a 24 bit DAC, but it is linear to only 99db (a little bit better than 16 bits). Because of this I CANNOT RECOMMEND the D30.
     
    Last edited: Jun 11, 2018
  2. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,198
    Trophy Points:
    113
    Location:
    Northwest France
    He also started applying linearity bits (amirbits) to headphone amps. I think the Lyr 3 has 16 amirbits.
     
  3. dmckean44

    dmckean44 In a Sherwood S6040CP relationship

    Pyrate
    Joined:
    Nov 17, 2016
    Likes Received:
    1,425
    Trophy Points:
    93
    Location:
    Peoria, IL
    HAHAHAHA!!
     
  4. LetMeBeFrank

    LetMeBeFrank Won't tell anyone my name is actually Francis

    Pyrate
    Joined:
    Aug 4, 2017
    Likes Received:
    3,758
    Trophy Points:
    93
    Location:
    Jackson, Mi
    Objectively speaking, according to his graphs, wouldn't a 24 bit DAC that only does 15 "bits" (37.5% less than advertised) be worse than a 16 bit DAC that only does 12 "bits"(25% less than advertised)?

    But since they are amirbits I guess it's all bullshit anyway.
     
  5. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,623
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Here is Amir's jitter test.
    Topping D30 versus Schiit Modi 2 Uber Jitter and Noise Measurement over USB.png

    Here are my JTEST results at 12kHz. Those sidebands are normal. They are embedded in the JTEST file. These are USB reports

    Modi 2U USB
    Modi 2U JTEST-1.jpg

    D30 USB
    D30  JTEST-1.jpg

    The D30 floor is absolutely flat with the JTEST file not introducing any jitter. The Modi 2U has that noise skirt, probably a result of jitter. This is terrible. Just terrible. Because of this, I CANNOT RECOMMEND the Modi 2U.
     
  6. Brooko

    Brooko Facebook Friend

    Joined:
    Oct 16, 2017
    Likes Received:
    132
    Trophy Points:
    33
    Location:
    New Zealand
    I see he and Jude are looking into swapping information on parameters, and possibly going back to measure the same Yggdrasil. That would be a huge step forward, and hopefully clear things up as far as the differences go. I hope it happens - as I think its going to expose quite a bit of difference in set-up and method.
     
  7. Vtory

    Vtory Audiophile™

    Pyrate MZR
    Joined:
    Sep 12, 2016
    Likes Received:
    10,826
    Trophy Points:
    113
    Location:
    East Coast
    [​IMG]

    Wow now he is trying to piss off ap555 lol.

    Amir uses +-0.1db as thresholds, implying results are valid no lower than -110db. That means his (secretly tuned?) older ap is only able to examine up to 18 amir-bits. But there were a number of his previous reviews arguing 19-20 bits of excellent gears. He contradicts himself. Cool.

    Hope Audio-precision guys to make a tough call. Like Uptone, AP should not sell their products to Amir. May lose short-term revenue.. but in the long run it will be beneficial to everybody but Amir.
     
  8. Brooko

    Brooko Facebook Friend

    Joined:
    Oct 16, 2017
    Likes Received:
    132
    Trophy Points:
    33
    Location:
    New Zealand
    BTW Marv - thanks for all your comparative posts. i'm finding it all very informative. My knowledge in this area of measurement is not very strong - so the straight forward comparisons are great for continued research into where the differences are and why they matter.
     
  9. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,623
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    You are welcome. I'm sure you guys know that I am using a little bit of satire. Thought I needed to say this just in case.

    DAC measurements are just like headphone measurements. Except much harder.

    Different rigs, different results. Inexperience or lack of care, like measuring 8 DACs in 8 hours may lead to subpar results. Care needs to be taken with cable placement, equipment stacking, the lights around the house, grounding, etc.
     
  10. huynhhungvnr

    huynhhungvnr New

    Joined:
    Mar 19, 2018
    Likes Received:
    20
    Trophy Points:
    8
    Location:
    VN
  11. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,623
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Now here is a Modi 2U vs D30 result that Amir did not cover.

    Modi 2U
    Modi 2U LF-1.jpg

    D30
    D30 LF-1.jpg

    What this at 60Hz,180Hz, 300Hz, and 420Hz on the Topping D30? Power mains noise. Obviously the people who designed the D30 did just a terrible, terrible job with the engineering of this DAC. This is totally unacceptable. Because of this, I CANNOT RECOMMEND the D30.
     
  12. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,623
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Might not be a bad thing. The more expensive analyzers come with a lot of bells and whistles, make life much easier, and are more dummy proof. Take this from a guy who doesn't get gifted high dollar instrumentation from the likes of AP, GRAS, etc. and has had to do with lesser gear.
     
  13. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,198
    Trophy Points:
    113
    Location:
    Northwest France
    Amir might have been giving AP a bad name and they are doing damage control.
     
  14. Hands

    Hands Overzealous Auto Flusher - Measurbator

    Staff Member Pyrate MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    12,284
    Trophy Points:
    113
    Location:
    Colorado
    Home Page:
    But her solder fluxes...
     
  15. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,623
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    One more. 1kHz.

    We'll start with Amir's result:
    Topping D30 versus Schiit Modi 2 DAC 1 kHz Residual Noise and Distortion Measurement.png

    Here are mine, which look quite a bit different.

    Modi 2U
    Modi 2U 1k-1.jpg

    D30
    D30 1k-1.jpg


    Modi 2U has higher 3rd, 4th, and 8th order.
    D30 has higher 2nd, 5th, 7th, 9th, and 11th.

    Since higher order odd harmonics are the most detrimental to sound quality, the D30 loses. I CANNOT RECOMMEND THE D30.


    FINAL SUMMARY:

    Topping D30: x3 I CANNOT RECOMMENDS
    Schiit Modi 2U: x1 I CANNOT RECOMMENDS

    CONCLUSION:

    Do not buy the Topping D30. It received x3 more I CANNOT RECOMMENDS in the objective tests than Modi 2U. In level matched subjective tests using the Massdrop THX AAA amp (uber measurements according to Massdrop site - to be verified) and modified HD800 headphones, the Modi 2U sounded OK, but the Topping D30 sounded like shit. Yes, I ALWAYS level match. If I didn't level match, than the Topping D30 would sound better right? Because it has 3db higher output.
     
    Last edited: Jun 8, 2018
  16. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,623
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I don't read HF, but I don't have a problem with HF denizens. Many HF'ers ultimately make it here. I am sad that the quality of content at HF has declined so much though. Sometimes I want to jump in (my wife has an account), but ultimately before I finish, I don't hit submit. It's not worth it. Most of us were HF'ers at one point, and if Jude hadn't silently banned many of us, we would still be HF'ers. I do dislike it when Jude uses his measurement gear to e-peen Tyll.

    Reddit: I like Reddit. There are a few wack jobs there. Some of those wack jobs even signed up here, but we ended up banning them because SBAF just wasn't a good fit. For this reason, there are a few that dislike me. Sometimes I post on Reddit, like when I have new headphone measurements. I should contribute there more often. I think the reason I don't is because the Reddit folks who would fit in here are already here.

    Hydrogen Audio. No idea what's going on there right now. Years ago, they used to freak out over NOS R2R implementations based on the PCM1704 because of the early roll-off. Just don't know what to say. Difficult to deal with that kind of thinking.

    ASR. Haven't kept track of them or read them until yesterday. I didn't even know that Amir had Yggdrasil A2 measurements, or that he followed up with a Modi 2U vs Topping D30 test a few weeks after I announced the D30 sounded like shit, which it does. I'll be putting an entry in my DNS hosts file so I can never reach them again after today.
     
  17. Darren G

    Darren G Friend

    Pyrate
    Joined:
    Apr 4, 2017
    Likes Received:
    589
    Trophy Points:
    93
    Location:
    Pennsylvania, PA
    I think he just listens to noise levels, and apparently he can hear the sound of a gnat farting in the wind.
     
  18. Merrick

    Merrick A lidless ear

    Pyrate
    Joined:
    Jan 6, 2016
    Likes Received:
    12,556
    Trophy Points:
    113
    Location:
    Portland, OR
    Funny, that’s the sound I hear whenever I read anything Amir writes!
     
  19. Darren G

    Darren G Friend

    Pyrate
    Joined:
    Apr 4, 2017
    Likes Received:
    589
    Trophy Points:
    93
    Location:
    Pennsylvania, PA
    I can NOT recommend listening to noise as a replacement for listening to music.
     
  20. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,623
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    The thing I find interesting is how Amir finds faults in Schiit gear that no one else (Jude, @atomicbob, JA, me, etc.) seems to be able to (with some of his measurements WAY off), but conveniently ignores faults on comparison gear (D30 in this case) by not showing them, as if he had an ulterior motive.
     

Share This Page