Massdrop x Grace Design SDAC - a budget beast

Discussion in 'Dedicated Source Components' started by project86, Sep 1, 2017.

  1. WilliamLeonhart
    Didn't see that the post before mine was of such fierce speculation, but as someone who has always been proud to have one or two amp/dacs in the budget line - and as someone who knows a tiny bit about making stuffs, I'd like to offer my 2 cents.

    First of all, we listen with our ears. I know I listen with mine. So while most of the budget amp/DACs IMO offer somewhat similar sound quality that many people perceive as better than your laptop/phone's output, it's not always the case.
    In the < $150 line, I've come to regard the ODAC as the golden standard. I've not found anything noticeably better than the ODAC under $200. Hell, I prefer the ODAC than most prominent budget DACs. The Schiit Modi 2 was, let's say 90% identical, but listening very closely, I think it's got less clarity than the ODAC. The fulla was a disappointment. The Dacport line doesn't sound that different from iPods, the Audioengine D1 won't offer anything better than normal headphone jacks with your mp3 and your aa3 (it does sound better with FLAC). The iDSD Nano is ok too, the only plus point is that it can play DSD. The only budget DAC that I consider a true competitor to the ODAC is the Dragonfly, but in the end, we're back at the same place: it's never a noticeable difference in this price range.

    So about years ago I was perfectly OK with my budget DAC/amp. Which explains why I'm in this thread: I got lent a m9xx by a friend and I was blown away. Before that, I had thought most DACs sound the same, I had found that the things that have the same DAC chip as those hundreds-dollar ones would not improve my listening experiences. But no, the m9xx brought a night-and-day improvement over the Modi 2.

    If my memory serves, they both use the same DAC chip. Yet, thanks to my limited knowledge in making stuffs, I'm pretty sure that the DAC chip only play a part in determining how a DAC sound. Take one of the m9xx's top competitor, the iDSD for example: it uses dual Burr-Brown chips, and it's no way warm or dark. The m9xx use AKM, and while it's smooth, it's still more neutral than the other AKM DACs that I know.

    Case in point: the Aune S16, my present DAC because I need a smoother DAC for my cheesy oldie music - which also turns out to be a slightly cheaper with much crappier amp section, the Aune S16. But the m9xx was and has always been special to me. In terms of sound signature, I'd take it over the Hugo (which I owned) and the Woo WA7 any day.

    The best thing is the m9xx costs only $500. It's the "mini, streamlined" version of the m920, which costs the same as the Hugo and the WA7. So on the one hand, the m9xx made me realize money can buy a happy listening experience. On the other hand, it reaffirms my belief that in this audio world, something still costs much less than it should.

    Which I hope would happen again with the SDAC really soon. I hope it can comfortably compete with $200, $300 DACs. I hope it can change my very strong opinion about the typical "budget" DACs.

    Anyway, I'm sorry for this lengthy rant, but it's not every day that I get to burst into proclaiming my love for Grace. (They're even based in Colorado. I f*cking love Colorado).
     
    Last edited: Sep 8, 2017
  2. Zachik
    Can anyone who auditioned the SDAC comment about comparing it to the Schiit Bifrost MB (a.k.a. Bimby)?
    I know very few auditioned the SDAC, but still hoping some auditioned the Bimby as well...
     
  3. project86 Contributor
    Michael - thanks for that very informative post. Lots of good info! It's always great to hear the thoughts behind the design, and why particular choices were made, etc. Can't wait for more folks to start experiencing this little gem for themselves.

    Zachik - it's been a while since I heard a Bifrost. That was back when it first came out, so not the multi-bit edition. Hard to draw any conclusions from 3 or 4 years ago.... but I recall thinking Bifrost was pretty good for its price, while I find the SDAC to be without peer. But they are not in the same price class, so who knows.

    Sorry if that's not more helpful.
     
  4. Zachik
    I understand that Bimby is $600 (compared to SDAC $80), but with all the hype here - I still hope to see some comparison (ignoring the price gap)...
     
  5. project86 Contributor
    Well from memory I'd say it was a fairly close race, but the Bifrost at the time was $350 (without USB).
     
  6. Zachik
    I am looking into the MultiBit version. It is $600. Not $400 or $350...
     
  7. Zachik
    MassDrop just pre-announced the combo CTH+SDAC.
    Did anyone audition a sample / prototype? I am mostly curious about:
    1. How do the SDAC and CTH match? what is the combined sound signature like?
    2. I am a little worried / concerned by interference between the SDAC and CTH. want to make sure there's no noise / hum / hiss introduced due to insufficient isolation between the 2 PCBs...
     
  8. project86 Contributor
    Haven't heard the combo unit but the individual components pair wonderfully together. Usually a sub-$300 combo unit will have inherent issues with treble glare, or loose bass, or fuzzy, indistinct imaging.... this combo has no significant flaws to speak of.
     
  9. Zachik
    What did you think about the sound signature of the combined setup? I auditioned the CTH at SF Meet with ZMF Atticus, and it sounded great together. Very pleasant warm sound (just the way I like it). I suspect the DAC should not change the signature much, but was just wondering...
     
  10. project86 Contributor
    I don't think the SDAC or the CTH are intrinsically warm. The CTH gives a slight perception of extra low end kick, and both devices are remarkably free of top end glare. So that may come across as sounding "warm" compared to something with leaner bass or a bright, offensive treble presentation. But generally speaking this is what I'd call neutral overall.
     
    Zachik likes this.
  11. feanorean
    And yet my post seems a little prophetic now doesn't it, with the way Massdrop has handled the release of their latest 3 items...

    I was not challenging the coverage by Project86 at all. I read his review with great interest. I apologise if it read that way. I too was astounded by the quality of his response.

    [/QUOTE]
     
    Last edited: Sep 11, 2017
  12. jude Administrator
    We ran measurements of the Massdrop x Grace Design Standard DAC (SDAC) in Head-Fi's audio measurement lab using the Audio Precision APx555 audio analyzer to compare results with Massdrop's published specs. We have so far found our comparable measurements to confirm that Massdrop's published specs are accurate, and leaning toward being conservative.

    Here are the measurements of the SDAC that we have run so far:

    RMS-Level---(20.0000-Hz---20.0000-kHz)---48-kHz-SR---96-kHz---0-dBFS.jpg

    Deviation-(20.0000-Hz---20.0000-kHz)---48-kHz-SR---96-kHz---0-dBFS.jpg

    Signal-to-Noise-Ratio---48-kHz-SR---44.1-kHz---0-dBFS---1-kHz.jpg

    THD+N-Ratio---48-kHz-SR---44.1-kHz---0-dBFS---20-Hz.jpg

    THD+N-Ratio---48-kHz-SR---44.1-kHz---0-dBFS---100-Hz.jpg

    THD+N-Ratio---48-kHz-SR---44.1-kHz---0-dBFS---10-kHz.jpg

    IMD---CCIF-Ratio---48-kHz-SR---96-kHz----6-dBFS.jpg

    IMD---SMPTE-Ratio---48-kHz-SR---96-kHz----6-dBFS.jpg

    Dynamic-Range---AES17---48-kHz-SR---96-kHz---0-dBFS---A-wt.jpg

    Crosstalk---48-kHz-SR---44.1-kHz---0-dBFS---100-points---20-Hz-to-20-kHz.jpg

    Scope---48-kHz-SR---44.1-kHz----90-dBFS---1-kHz.jpg

    FFT---10-kHz-sine-wave---44.1-kHz---48-kHz-SR-bandwidth---192k-points---average-10---LEFT-ONLY.jpg

    FFT---10-kHz-sine-wave---44.1-kHz---48-kHz-SR-bandwidth---192k-points---average-10---RIGHT-ONLY.jpg



    Our audio measurements in this post were made using:
     
    Jimster480, volly, CEE TEE and 2 others like this.
  13. xrk971
    Thanks for the measurements. It would have been nice to see a FFT at the traditional 1kHz so that characteristic harmonic profile of 2nd, 3rd, 4th, 5th, ... can be seen. When done at 10kHz, it's not very informative as all we can see is the second harmonic value. The top of the fundamental should also be visible and not trimmed off since we have no way to reference what the magnitude of the one peak at 20kHz means.

    The main reason FFT and THD specs are made at 1kHz is because we can still measure and see up to 10 or more harmonics within the bandwidth of the devices.

    And the profile is more important than the absolute THD. What we want to see is predominant second harmonic and monotonically decreasing higher orders. Preferably nothing more than 4th or at most 5th order.

    I have never seen a measurement using AP equipment of distortion and an FFT with only 10kHz excitation and not 1kHz.
     
    Last edited: Sep 12, 2017
  14. jude Administrator
    You need only look one page back for just such an FFT on another AP analyzer.

    Also, distortion wasn't shown only at 10 kHz. As they did on the Massdrop page for the drop, I also showed THD+N at 20 Hz and 100 Hz. As I said in my post, I first wanted to compare results with Massdrop's published specs. We can run more measurements later (perhaps next week).

    Again, I can run more SDAC measurements next week if I have the time. (We're shooting the CanJam @ RMAF Preview Video now and next week, too.) When we do run more measurements, I'll post the FFT and distortion product plots with a 1 kHz sine. (The distortion product bar plots go out to the 10th by default.)

    The top of the fundamental is visible and is not trimmed. From Audio Precision's website: "AP audio analyzers are equipped with two dB reference controls, dBrA and dBrB, that can be used to set arbitrary decibel reference values." I set the reference level (dBrA) to 2.15 Vrms to match what Michael Grace had posted on the previous page, 2.15 Vrms being the output level of the 10 kHz sine at -0.15 dBFS. With the top of the fundamental referencing 0 (zero) dBrA, I think the magnitude of the 20 kHz spur (relative to the fundamental) is actually clearer.
     
    Last edited: Sep 13, 2017
  15. xrk971
    Thanks, I just meant from measurements I have seen of almost any other amplifier on published specs or on diyaudio - 1kHz is the usual standard. If you were trying to match what Massdrop did then that is fine. It's just not the most informative point to do a measurement of harmonic distortion as the ear is very sensitive to distortion iin the range from 500Hz to 5kHz. Thanks for the data and for the future measurement at 1kHz.

    That is very nice that Head-Fi has it's own AP analyzer.
     
    Jimster480 likes this.

Share This Page