Square Wave Hard On

Discussion in 'Measurement Techniques Discussion' started by FidelCastroPS, Sep 19, 2017.

  1. Thad E Ginathom

    Thad E Ginathom Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    14,134
    Trophy Points:
    113
    Location:
    India
    Oh look! A simple way to interpret some of those squiggles. Just what I always wanted.

    Oh wait.

    Nope. Not so simple.

    Well, I guess I'm grateful to @FidelCastroPS for proposing, and to @atomicbob and others for disposing.
     
  2. FidelCastroPS

    FidelCastroPS New

    Banned
    Joined:
    Sep 2, 2017
    Likes Received:
    2
    Trophy Points:
    3
    Location:
    NYC
    Appreciate your input Cspirou. The lower end of Schiit products are made to be as low priced as possible (e.g. USB gen 2 for Modi MB). Adding an opamp ads cost to circuitry and seems contrary to their desired budget optimization. Except maybe, as you state so, to make the Modi technically better, though one may end up with other problems such as distortion. The point remains that Schiit is aware that their square wave is not really square. Are we talking about a trade off between square wave vs THD? If discrete buffer is better than opamp then why Modi is better or equal sounding than other DACs, including Ares?
     
  3. FidelCastroPS

    FidelCastroPS New

    Banned
    Joined:
    Sep 2, 2017
    Likes Received:
    2
    Trophy Points:
    3
    Location:
    NYC
    What is the "normal behaviour"? That is precise point I asked before, whether anybody had a "standard" to compare with. I say this because of my experience developing products and testing them. Compared with what reference?

    The way I see is that Schiit has made changes to the filter for Modi MB, e.g., the math behind it. It may be more accurate than the one used for Iggy (see Schiit own description of Modi MB - a "version" of the same digital filter). It seems to me that all mfg use some kind of filter in addition to hardware, R2R or R2R ladder. It is our business as consumers to ask the right questions. What combination of hardware and software is best for me?

    I am very grateful for the opportunity to examine your data, without which we cannot make serious analysis. On the subject of wasting time, we ask questions from our betters so we can learn more. Isn't why we are here?
     
  4. FidelCastroPS

    FidelCastroPS New

    Banned
    Joined:
    Sep 2, 2017
    Likes Received:
    2
    Trophy Points:
    3
    Location:
    NYC
    True solutions are those that nobody even thought they even needed. To do that you need to ask questions, lots of seemingly stupid questions. What is a stupid question? Something that makes sense to you but not to me. By understanding your error or my misunderstanding, it is possible to achieve great results. Sufficient to say, I've got more interesting answers here than anywhere else.
     
  5. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    @FidelCastroPS:

    The Modi MB I believe uses an opamp for IV from the DAC chip, and then into a discrete output stage. For small signal circuits, opamps are not necessary more expensive than discrete part circuits. The Modi MB likely has a better square wave because there is no servo. From the looks of the square wave, the Yggdrasil might have a servo but I am not sure. The need for one depends upon the circuit and is beyond the scope of this conversation.

    The difference in the digital filter from Yggdrasil and Modi MB is mainly upsampling rate and window size (number of taps). This should not have an effect on the square wave, unless a high-pass was built into it.

    What @atomicbob was saying is that you need to walk before you run, and at this point you need to learn walk. You seem to have some basic knowledge, but you arrive at odd conclusions and associations. Understanding measurements is an extremely complex task and beyond the scope of this forum.

    As a consumer, you are asking all the wrong questions. No one can tell you what combination of hardware, software, or measurements is going to be right for you. 95% of modern DACs measure exceptionally well. The non-linearities are more a matter of nickpicking than anything else, and are likely not to translate into anything subjectively. Because of this, the best and first thing for you to do is to acquire and listen to the gear rather than understand the measurements, or in particular any one single measurement.
     
  6. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    Ok, let me give you one more set of pointers. I am using the square wave response as a quick way to observe system step response. Read about step response here:
    https://en.wikipedia.org/wiki/Step_response

    You might wish to learn how system step response relates to system impulse response.

    In the case of DACs there are recovery filter(s) response, output coupling and impedance, etc. to consider. In the case of Amps there are output capacitors or DC servo systems, output impedance, power supply performance, etc. to consider.
     
    Last edited: Sep 20, 2017
  7. FidelCastroPS

    FidelCastroPS New

    Banned
    Joined:
    Sep 2, 2017
    Likes Received:
    2
    Trophy Points:
    3
    Location:
    NYC

    "are not necessary more expensive . . . the Yggdrasil might have a servo but I am not sure . . . unless a high-pass was built in. "

    Excuse me but you are being very reassuring on your own comments. We can look under the hood of car but if the software is crooked then we are all played for fools (see VW scandal). In this case, it is the filter, the secret recipe!

    I learned a long time ago to walk and think with the grey thing between my ears. I've listened to good systems in the late 1980s. Back then, I spent a considerable amount of time choosing a good system to last 20-30 years. I think I did well with KEF 105.2 and B&W Matrix 1 speakers, Threshold S 500 power amp, Krell PAM-1 preamp. I diy recently a Krell KSA5 headphone amp. There were better equipment but my system was within my post college budget.

    I don't know about that "95% of modern DACs measure exceptionally well". I hate this type of generic statement. Measure well against what? Gustard X20 measures exceptionally well but would you rather have it instead of a good R-2R? From what I hear there are just a few good R2R under $1K and a few more between $1-2K and few more over $2K. Right now, I'm trying to figure out Schiit multibit vs R-2R ladder. Getting an open discussion on square wave or the lack of it shouldn't ruffle so many feathers, right?
     
  8. FidelCastroPS

    FidelCastroPS New

    Banned
    Joined:
    Sep 2, 2017
    Likes Received:
    2
    Trophy Points:
    3
    Location:
    NYC
    Thanks atomicbob for your time spent. I once had a 4 variable product to test. I thought I needed to make a very large number of combinations, something like 4x4x4x4 = 256. I thought impractical and un-economical to test it. Then comes this very large company and cut the chase to test just one variable, one that affects all these 4 variables. Maybe that is why some people like square wave.
     
  9. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    GUTB?
     
  10. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    It's obvious that you it all figured out. May I request that you go elsewhere and stop wasting our time then?
     
  11. FidelCastroPS

    FidelCastroPS New

    Banned
    Joined:
    Sep 2, 2017
    Likes Received:
    2
    Trophy Points:
    3
    Location:
    NYC
    Tuberculosis of the Genitourinary System Overview of GUTB
     
  12. Dino

    Dino Friend

    Pyrate
    Joined:
    Jun 21, 2016
    Likes Received:
    3,332
    Trophy Points:
    113
    Dammit. I was looking forward to Fidel's analysis of The Gadget.
     
  13. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Yeah. I sort of got lost with the VW scandal and how that might relate to the mega-combo-burrito-filter causing bad square wave emissions.
     
  14. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    Apparently the filter detects when AtomicBob is doing tests to spit out perfect square waves
     
  15. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    After post 26 I kind of got lost. I don't grok the posts following, and it is probably for the best.
     
  16. Thad E Ginathom

    Thad E Ginathom Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    14,134
    Trophy Points:
    113
    Location:
    India
    Did GUTB even pretend technical knowledge? As far as I can remember his was less than mine. When I short out safety devices, it is usually by accident, and with a hand, not on purpose with a nail. And I generally remember not to use inflammable stuff in places that get hot.

    I'm willing to admit that he might have been less deaf than I am, but hey, that is a matter of fact or not, and it is not a fair world. And I sometimes doubt that some of those guys do hear any more, just they've never been near an audiologist.
    Do you not need to know what the fumes from that filter are doing to your children's health? I think we should be told!

    :pirate07:
     
  17. trl

    trl Acquaintance

    Joined:
    Apr 23, 2017
    Likes Received:
    33
    Trophy Points:
    18
    Location:
    Iasi, Romania
    I've used my Matrix M-Stage HPA-3B for the below test, so same DUT, same PicoScope 2204A, same everything, just the input high-pass filter was "adjusted". A-probe (blue channel) is the input signal, B-probe (red channel) is the 6.3mm output plug of the amplifier.

    Input high-pass filter created with 50kOhms resistors and 2.2uF/250V non-polarized caps:

    [​IMG]

    Input high-pass filter created with 150kOhms resistors and 4.7uF/250V non-polarized caps:
    [​IMG]

    Direct coupled audio signal (input caps bypassed or in short-circuit):
    [​IMG]

    I'm 99% sure that with direct couple low-end bass sounds a bit better, but honestly...without having a direct A/B test with 2 identical devices (one original and the other modified) I can't swear my ears are right.

    Also sine-waves could be affected by a poorly designed input high-pass filter, either in amplitude, either in phase...or perhaps both:
    [​IMG]
     
    Last edited: Mar 23, 2018
  18. trl

    trl Acquaintance

    Joined:
    Apr 23, 2017
    Likes Received:
    33
    Trophy Points:
    18
    Location:
    Iasi, Romania
    I would extend the above affirmations to the low-pass filter as well.
    DUT: the same as above, Matrix M-Stage HPA-3B; also, the same scope.
    A-probe (blue channel) is the input signal, B-probe (red channel) is the 6.3mm output plug of the amplifier.

    Input low-pass filter created with non-polarized SMD caps (can't remember their capacitance, sorry):
    [​IMG]
    [​IMG]

    Input low-pass filter removed (SMD caps desoldered)

    [​IMG]
    [​IMG]
     

Share This Page