Musings on audiophile-dom w/ a Topping D90SE Mini Review

Discussion in 'Digital: DACs, USB converters, decrapifiers' started by SoupRKnowva, Aug 29, 2021.

  1. SoupRKnowva

    SoupRKnowva Official SBAF South Korean Ambassador - Friend

    Friend
    Joined:
    Sep 26, 2015
    Likes Received:
    2,956
    Trophy Points:
    93
    Location:
    Austin, TX
    This is going to mostly be background on why I even bothered getting the d90se, a story if you will, that maybe will resonate with others. If you just want to know what I think of the dac, skip to the end, but the tldr is that it sucks.


    I've kind of been in what I would call the "audiophile" end of the pool as far as my equipment goes for quite a few years at this point. I never bought things based on measurements. I bought them with an eye on the measurements, but based on what I had heard at meets and at home, with the impressions of members I trust, and also what I thought of the engineering of the product in question.

    This has led to some excellent purchases on my part. The loaner programs here and at Changstar back in the day have also been invaluable, thats what led me to discover my love for the Gungnir DS. Trying all this gear, and reading impressions here is what allows you to find other members with similar tastes as yours, and even if they have different tastes in gear, you can learn to triangulate what you might think of the gear in question and whether it is worth persuing for your rig.

    This is where we get to why I bought the D90SE. I've always considered myself to be very engineering focused, shit, I got a degree in electrical engineer. I often times don't just read SBAF or head-fi, I do somewhat frequent ASR to lurk. I think reading a broad spectrum of sources gives you the best chance of finding your own biases and coming to the correct conclusions. I also constantly question my own decisions, thought processes, and actions, sometimes drastically.

    So a few weeks back, I was about to make some rather large purchases, a new AVC passive pre and a new DAC. And it was in that moment that I realized I'd never heard any measurement-fi gear. Not at all. And this is when a small nagging part of my brain was like, "Hey idiot, maybe, if the measurements say this shit all sounds the same, maybe it really does and you've just been fooling yourself this entire time, wasting all your money." Now, normally I wouldn't have humored this thought, but going into a rather large purchase, I was in the mood to experiment for a week or two first instead. If anything, I could write about it here and confirm something in the process.

    So what I did was buy a Topping D90SE and a Benchmark AHB2 (this isnt a review of the AHB2, but I will say, I do like the amp so far, it's not an immediate dismissal like the d90se was). These are both the best or one of the best measuring devices in their field. I also want to say that I went into this with an open mind, when I question my own behavior, I'm not half assing it. If anything, I wanted to like the Topping DAC, because its small, much cheaper than what I was looking at, and has a "nice" built in volume control. It would simplify my setup a great deal.


    ~~~~~~~ACTUAL(brief) IMPRESIONS START HERE~~~~~~

    My rig currently consists of a pi4/pi2aes feeding a Schiit Gungnir DS feeding a Coincident Turbo 845 SE powering my Acoustic Zen Adagio speakers.

    The shittiness of the D90SE was obvious immediately upon plugging it in. I had read that it had some trouble with certain digital sources, sometimes having dropouts on optical from poor sources, to include TVs for instance. This isn't just true, it's very true. I get dropouts on the optical input from my TV once or twice every minute. The dropouts are brief, but I have never in my life had a DAC do this, from any source. To include shitty motherboard optical outputs. So already the Topping was in the hole.

    One thing in its favor though was the volume control. I am currently using a Coincident Turbo 845 SE integrated which has a stepped attenuator with some rather large steps in some spots I run into all the time. So I found having perfect 0.5db steps super handy.

    The next ding against the D90SE was with other digital inputs also dropping out. Using I2S from th pi2aes, which is a known low jitter source, I was getting dropouts like once an hour or so, same with RCA SPDIF. At this point I knew the D90SE was going back, but let me briefly give some auditory impressions of this DAC.

    To say that the D90SE can't soundstage, is a diservice to other DACs with poor staging abilities. It isn't just that it stages flat, cause it most definitely does, papersheet flat in fact. But within that 2d plane of "soundstage" there is almost no ability to discern instruments from each other, this is a very messy and congested sounding stage. Instruments and performers have no space or size to them, they are all flattened into the sheet that is the canvas this device draws.

    The Gungnir DS is no titan of staging/separation/holography, as many of you well know, I like it for other reasons. But when comparing it back to back with the D90SE, it was like going from a shitty mono speaker to surround sound, the stage exploded, giving space and separation to all of the players. The other thing the D90SE lacks completely is emotional involvement. The micro-stuff is just gone, there is no texture to vocals or instruments, the haunting realness of certain female vocals was gone completely, no goosebumps. It also has no drive/intensity/slam at all, and many of you know that is one of the main reasons I love the Gungnir DS.

    I went into this wanting to, almost expecting to, like and keep the Topping D90SE, but left even more dissapointed than I could have imagined I would be. They couldnt even test/design it well enough to get the digital inputs of the device working? (Oh, I should add, using it via USB I never had a dropout) And that is on top of absurdly lackluster musical performance. At 900USD I consider this DAC a ripoff. My Gungnir DS, a DAC that is not lauded for much of anything by anyone, trounces it at pretty much everything, I guess it doesnt have volume control though.


    Sorry @purr1n you asked for a mini review, I guess I did give you one, it just lives at the end of a journey.


    Has anyone else had similar thoughts to question all of their "audiophile" purchases? I feel this is natural, its just my humoring it that was out of the ordinary. But what it did do, was once again confirm my belief that measurements are only telling part of the story in this audio journey we are all on. As the saying goes, if it sounds good and measures bad you're probably measuring the wrong things.

    I didn't listen to this with headphones either, I basically don't use headphones anymore, except my AirPods Max while I'm at work, so YMMV on headphones as far as how much the things I bitched about will stand out there.

    It has also already been put back in the mail to return it and get my money back, so unfortunately further impressions won't be coming.
     
    • Like / Agree Like / Agree x 39
    • Epic Epic x 16
    • List
  2. k4rstar

    k4rstar Britney fan club president

    Friend BWC
    Joined:
    Jun 11, 2016
    Likes Received:
    5,490
    Trophy Points:
    113
    Location:
    Toronto, Canada
    thank you for the reminder to not purchase gear manufactured by Chinese beancounters who select (often fake) components from a spreadsheet for maximum margin control, with circuits designed by "engineers" who have about as much know-how as a high school student who took an electronics 101 elective, marketed by people who treat audio like any other consumer electronics industry, and promoted by people who don't like music.

    https://www.superbestaudiofriends.o...-or-topping-part-deux.9583/page-3#post-309659

    as you can see, the irony is everywhere, but the promises are so convenient that it is only natural to be tempted by them. this is why reminders are needed.
     
    • Like / Agree Like / Agree x 15
    • List
    Last edited: Aug 29, 2021
  3. Cellist88

    Cellist88 Friend

    Friend
    Joined:
    Nov 11, 2015
    Likes Received:
    1,314
    Trophy Points:
    93
    Location:
    NJ
    Think of it like coffee. You dont solely judge a cup based on a reading from a refractometer.
     
    • Like / Agree Like / Agree x 6
    • List
  4. Qildail

    Qildail Facebook Friend

    Joined:
    Aug 6, 2021
    Likes Received:
    241
    Trophy Points:
    33
    Location:
    Iowa
    A long story for another thread sometime, but by the time I had gotten around to shipping the third box of speakers back, I was starting to question my own sanity, in addition to my purchases...

    Thanks for the write-up on the D90SE. Sorry to see you went through a bit of hell for it.
     
    • Like / Agree Like / Agree x 2
    • List
  5. Erroneous

    Erroneous Friend

    Friend
    Joined:
    May 15, 2016
    Likes Received:
    1,202
    Trophy Points:
    93
    Location:
    Texas
    Thankfully you can now trust your ears. Unfortunately that is the expensive way to go. Sorry for your wallet, happy for your ears and future.
     
    • Like / Agree Like / Agree x 3
    • List
  6. GoodEnoughGear

    GoodEnoughGear Evil Dr. Shultz‎

    Friend
    Joined:
    Oct 25, 2015
    Likes Received:
    2,805
    Trophy Points:
    113
    Location:
    Cape Town, South Africa
    I see this as a healthy exercise. You had the option to return the unit, so it was a pretty safe way to check your biases and your assumptions a bit. If more ASR folks did something similar and actually tested with their own ears, maybe ASR would be less myopic.
     
    • Like / Agree Like / Agree x 9
    • List
  7. purr1n

    purr1n Super Friend

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    74,329
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    This didn't have to suck. The reason it sucks is because the Chinese Topping designers are inexperienced and think SINAD means everything. Or either that, someone paid Amir a good amount of money to assist the Chinese sell the kind of shit that they are willing to do and can only do (copy implementations in datasheets and pick parts via APx555).

    I am surprised but also not surprised this sucks. I've seen another Chinese company (not all are dumb) start to figure out tweaks to change the distortion profile of the ESS chips. Of course the problem is that if one starts to add second harmonic to get nicer pattern (say like a triode), overall SINAD will go up. One can also customize the filter. Topping just ain't gonna do that. I bet the Topping designer also threw in a bunch of OPA1612 again, and picked a default filter purely for the sake of measurements. If the Topping designers don't know shit about how to design discrete, they sure ain't gonna know much about coefficients for a customer filter or harmonic profile.

    The difference between designers of good audio gear and bad inexperienced ones is that the good ones will also listen. They will be willing to sacrifice a few db of SINAD here, and another few db of SINAD there, if it sounds better. Pick the right IV opamp, right output opamp, maybe develop discrete output, custom filter, custom H2/H3, to in essence cook a tasty dish. Audio is ultimately a subjective experience. Just a taste. The best restaurants in the world, their chefs, don't use a SINAD equivalent. Good cooks always taste their food while making it and the best don't use recipes that involve exact cups of that or teaspoons of that.

    --

    Dropouts can happen. I can see this happening from shitty sources or USB even today. However, dropouts from pi2AES: that shouldn't happen. @SoupRKnowva: did you try out different options, PLL, filters, or whatever weirdo 10000 permutations that Chi-Fi DACs usually offer?

    P.S. I suspect the ESS chips have a cheat mode for when benchmark tests are run, e.g. 1kHz tone. Not any different from what the GPUs or Volkswagon diesels did years ago. The fact that H2 and H3 can be tweaked - that makes me super suspicious.
     
    • Like / Agree Like / Agree x 9
    • List
    Last edited: Sep 1, 2021
  8. bboris77

    bboris77 Friend

    Friend
    Joined:
    Dec 12, 2015
    Likes Received:
    723
    Trophy Points:
    93
    At least their ad copy is wildly entertaining, unintentionally.

    upload_2021-9-1_23-2-48.png
    upload_2021-9-1_23-3-19.png
     
    • Like / Agree Like / Agree x 1
    • List
  9. Khronos

    Khronos Rando

    Joined:
    Aug 17, 2020
    Likes Received:
    8
    Trophy Points:
    3
    Location:
    USA
  10. Baten

    Baten Friend

    Friend
    Joined:
    Mar 18, 2018
    Likes Received:
    930
    Trophy Points:
    93
    Location:
    EU
    About the jitter and dropout problems, from what I understand it's because in these mass produced products they can't get any good s/pdif receivers anymore. AKM parts were among the most compatible but since the factory fire manufacturers need to be more inventive with other parts or in-house solutions. And that requires good R&D.
    ((still doesn't explain the I2S drop-outs though))

    Smart smaller brands tend to have stock of these receivers for at least a few more runs or their current offerings. But this might become a bigger problem and not just some SMSL/Topping issue.
     
    Last edited: Sep 1, 2021
  11. Tekker

    Tekker Acquaintance

    Joined:
    Aug 1, 2020
    Likes Received:
    54
    Trophy Points:
    28
    Location:
    Netherlands
    Well those audioscientists over there believe (key word, believe) soundstage and placement comes from an emphasized treble, so no wonder it’s flat as a sheet of paper lol.
     
    • Like / Agree Like / Agree x 1
    • List
  12. SoupRKnowva

    SoupRKnowva Official SBAF South Korean Ambassador - Friend

    Friend
    Joined:
    Sep 26, 2015
    Likes Received:
    2,956
    Trophy Points:
    93
    Location:
    Austin, TX
    I tried it out in the default configuration which I believe is the minimum phase fast roll off filter with xlr and rca outputs, and changed to the linear phase fast roll off filter with just xlr outputs and had the dropout issues using both filters. I did not try all the filters though.

    I wouldn't be against hearing a soekris DAC, but I have already bought another DAC that will almost certainly be my endgame.

    This is true, but at some point, it doesn't really matter to me. The buck has to stop somewhere, and for me that is with the product I am buying, Topping, or whoever, because like you said this may end up effecting more companies, needs to figure out how to make it work. Not just point their finger to a part supplier and say its their fault, that supplier isn't selling me a DAC that is supposed to do the thing.

    And yeah, it basically blew my mind when I heard my first dropout via I2S from the pi2aes.

    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    A little more context. I did these comparisons with the Gungnir using the Topping both DAC direct into the ahb2, as well as into the Coincident both with the DAC volume control with the coincident in bypass and using the Coincident volume control. Because of the spur of the moment decision to get this thing, I only intended to use it DAC direct, and do not have a preamp since the Coincident is an integrated, the comparisons with the Gungnir on the ahb2 used my Jotunheim as the pre for the Gungnir, which couldn't have been doing it any favors either.


    I still can't really wrap my head around just how bad the D90SE was. The worst part though, is that there is a group of people out there, that would say anything that sounds different than it, is less accurate/distorted.

    Another thing I find hilarious, is that I have seen other people say the reason the soundstage sucks is because the crosstalk on the D90SE isn't as good as the rest of its measurements. Which is funny for several reasons. These came not from ASR, but from other audiophiles, who arguably aren't measurement-fi types, so why are we theory mongering about this particular measurement? Not to mention, it still has ~ -110db crosstalk at 22KHz and is -140 at 1Khz, so its not like they're "bad". Also, the only Gungnir DS measurements I could find were from Hands on Changstar, but they show it having -96db at the best case and like -70db at the worst case and its soundstage destroys the d90se
     
    • Like / Agree Like / Agree x 2
    • List
    Last edited: Sep 2, 2021
  13. purr1n

    purr1n Super Friend

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    74,329
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Crosstalk has nothing to do with soundstage. Feedback and how its applied can have more to do with soundstage than crosstalk, i.e. high gain vs low gain, active RIAA vs passive RIAA, EQ with LC in active or LC in passive. In the case of RIAA, TT's are maybe able to get 33db crosstalk on a very good day - so go figure - I've heard phonostages that push the soundstage beyond the wall and others that stage upfront and flat.

    70db crosstalk is the difference between ambient in your living room and a chainsaw three yards away. The truth of the matter is ESS chips tends to suck with respect to soundstage. No ESS based DAC has ever made me say wow to its soundstage. This is not to say that it can't. It's just that I don't see manufacturers making similar efforts as they have with say the AKM, e.g. Burl and Crane Song, but these two companies do some customization instead of taking a cookie-cutter approach.

    In theory, there should be some ability to tweak the soundstage via the filter coefficients.

    With dropouts, there could be something broken. A bad part, bad solder. I didn't have any dropout issues with the Topping D70 DAC. Although the D70 was utterly unimpressive and really a poor value.
     
    • Like / Agree Like / Agree x 2
    • List
    Last edited: Sep 2, 2021
  14. whitefang

    whitefang Rando

    Joined:
    Aug 20, 2021
    Likes Received:
    0
    Trophy Points:
    1
    Location:
    Bay Area, California
    I have almost exactly the same experience with you but on the A90 Amp. It completely flatten the soundstage and is the most inorganic, lifeless sound I have ever heard from any devices, the sound is worse than the 3.5mm of my iPhone SE.
    I am afraid I am biased so I have my wife to do a blind test with me and I can correctly pick it up 95% of time(I failed one time in 20 tries and is probably due to fatigue of my brain/auditory memory)
    And I am actually a fan of so-called "bright/cold/analytical/sterile/boring/flat/dry sound" like the Benchmark DAC1. HPA4, GSX mk2, THX789 etc.
    The unit works fine though, nothing broken, and it does have a black background with absolutely no hiss on IEM but that's it.

    I guess the D90 does have good synergy with the A90 because you cant make a dead sound more dead. thus they together suck 50% less.

    I always wonder what in particular is the problem with the A90 that makes it sound so wrong, and if there is any kind of measurement could show the problem.
     
    Last edited: Sep 2, 2021
  15. Qildail

    Qildail Facebook Friend

    Joined:
    Aug 6, 2021
    Likes Received:
    241
    Trophy Points:
    33
    Location:
    Iowa
    Between the Tubes-Solid State Wars, the Loudness Wars, the Vinyl-Reel Conflict (nasty one that… medics were respooling the tapes for weeks after the first battle), the Analog-Digital War, the CD-Streaming War, and now the SINAD War, I urge diligence. One poorly interpreted sentence near someone with control over the op-amps, and we'll soon end up with the first component named Sound Death Doubler Combo DAC, with graphs proving outputs sound 200% deader than the competition.

    </sarcasm>
    </martini>
     
    • Epic Epic x 3
    • Like / Agree Like / Agree x 1
    • List
  16. purr1n

    purr1n Super Friend

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    74,329
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Yes, there is a measurement. It's called global negative feedback. Transformer tube amps will have 0db of feedback, with maybe having 2db, 3db, or 6db. Solid-state amps will have between 20-33db of feedback. Chip amps will have > 100db of feedback.

    Topping's favorite opamp, the OPA1612 has 130db of open loop gain. That's 130db of feedback if used in unity gain.

    Another method is THD. If THD gets too good, it's matter of tons of open loop gain and tons of feedback. The best sounding gear has between 70-90db THD.
     
    • Like / Agree Like / Agree x 3
    • List
    Last edited: Sep 3, 2021
  17. purr1n

    purr1n Super Friend

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    74,329
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Soups will hate the Soekris. I already know this. Long time members are fully aware of my powahs of precognition when it comes to audio gear and certain people I know. There is a reason that Soups stuck it out with the Gungnir DS despite it not being exceptional at most things.
     
    • Like / Agree Like / Agree x 1
    • List
    Last edited: Sep 3, 2021
  18. BarnBurner

    BarnBurner Acquaintance

    Joined:
    May 24, 2021
    Likes Received:
    40
    Trophy Points:
    18
    Location:
    New England, USA

    Thank you for this great review. I had a somewhat similar experience with the Benchmark DAC3 HGC--it was technically flawless and definitely studio - grade, but left me really empty. I did a bunch of stuff, swapping streamers, transports, amps. When I finally hooked up the 7-yr old Krell streamer everything was there, all the dynamics, soundstage, and energy. And it can't really do more than 16/44.1--at least mine couldn't, no matter what I tried. Constant dropouts.

    Anyway, even at Redbook quality the implementation was so fantastic that I promptly sold the Benchmark. I would happily own another piece of their gear, and others could do the same comparison and think in completely mental, but for these ears in my specific system it wasn't even close.
     
  19. dericchan1

    dericchan1 Acquaintance

    Joined:
    Sep 14, 2021
    Likes Received:
    24
    Trophy Points:
    18
    Location:
    Canada
    I am for one actually against buying any chi-fi regardless of subjective or objective reviews, hence the decision to purchase my ifi neo idsd, which I am absolutely happy with.

    however, based on what you have described with the drop outs and the flat sound, could this be a defective unit? There can’t be so much hypes and praises out there by many if it messes up like that?
     
  20. Tekker

    Tekker Acquaintance

    Joined:
    Aug 1, 2020
    Likes Received:
    54
    Trophy Points:
    28
    Location:
    Netherlands
    That’s why ASR is a cult. There really are that many people who consider those units “audibly transparent” just because of their SINAD and Amir praising it.
     
    • Like / Agree Like / Agree x 1
    • List

Share This Page