Field guide to determining Modi MB or Bifrost MB v1 or v2 firmware

Discussion in 'Source Measurements' started by atomicbob, Jul 16, 2018.

  1. Ash1412

    Ash1412 Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    667
    Trophy Points:
    93
    Amplitude should be 0.000316228 if im correct. And USB noise is probably responsible for your results being fuzzier than atomicbob's
     
  2. Elnrik

    Elnrik Super Friendly

    Pyrate
    Joined:
    Jan 10, 2017
    Likes Received:
    8,973
    Trophy Points:
    113
    Location:
    Denver CO
    Home Page:
    I killed a bunch of background processes (Skype... F you Skype...) and moved the cables.

    -90 dBFS
    Scope3.png

    At your -70dBFS setting:
    Scope4Neg70.png

    I don't think my setup is going to be accurate enough to show you what you're looking for.

    I have to do some work today, unfortunately. I'll fiddle around with more this afternoon.
     
  3. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Based on these measurements, I cannot recommend the Modi Multibit "v2". LOL, sorry I couldn't resist invoking the Amir.

    Your measurements are fine. Good enough to see that there is no glitch at -70 and there is dithering at -90.
     
    Last edited: Jul 21, 2018
  4. Elnrik

    Elnrik Super Friendly

    Pyrate
    Joined:
    Jan 10, 2017
    Likes Received:
    8,973
    Trophy Points:
    113
    Location:
    Denver CO
    Home Page:
    IMG_20180719_110853.jpg

    Damnit.

    The Predator narrowly escaped my nasal-java bombardment. The keyboard is not so lucky.
     
  5. bobsherman

    bobsherman Acquaintance

    Joined:
    May 14, 2016
    Likes Received:
    56
    Trophy Points:
    18
    Yes the amir does tend to grow on you...

    ...Like a very disgusting fungus!
     
  6. gixxerwimp

    gixxerwimp Professional tricycle rider

    Pyrate
    Joined:
    Sep 30, 2015
    Likes Received:
    5,773
    Trophy Points:
    113
    Location:
    small island claimed by China
    Thanks for the analysis @atomicbob. I neglected to mention I also used USB input as moving the Eitr from my stack would have been a big PITA. How much difference would this have made?

    I got mine in February. Too bad it seems I just missed the cut. Why does my Modi Multibit sound worse all of a sudden? ... :p
     
  7. Ash1412

    Ash1412 Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    667
    Trophy Points:
    93
    Wait a minute, I'm getting a bit confused. If the signa at 90db is 63.25 uVrms, how are you guys able to set the amplitude to 1.5m (1500 uV) and still see anything? I'm getting these results right now.

    Note: using onboard soundcard

    upload_2018-7-19_22-37-24.png
     
    Last edited: Jul 19, 2018
  8. Hands

    Hands Overzealous Auto Flusher - Measurbator

    Staff Member Pyrate MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    12,285
    Trophy Points:
    113
    Location:
    Colorado
    Home Page:
    Do you have a field guide that can tell me if the DAC sounds good?
     
  9. Ash1412

    Ash1412 Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    667
    Trophy Points:
    93
    Results with amplitude set to 150uV at -90db, looks a lot more like what it should look like

    Note: Using onboard soundcard

    upload_2018-7-19_22-44-27.png
     
    Last edited: Jul 19, 2018
  10. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    Have an appropriate beverage and I'm willing to bet a certain amount of harmony will be restored.

    All audio components have strengths and weaknesses. Listeners have preferences. Matching appropriate mix of component strengths / weakness to listener preference results in auditory happiness. Outside of that, I got nothing. :D

    I'm impressed at how many of you are giving this a try.
     
    Last edited: Jul 19, 2018
  11. gixxerwimp

    gixxerwimp Professional tricycle rider

    Pyrate
    Joined:
    Sep 30, 2015
    Likes Received:
    5,773
    Trophy Points:
    113
    Location:
    small island claimed by China
    The nervosa made me do it ... o_O:confused::eek: ... :drunk:
     
  12. Elnrik

    Elnrik Super Friendly

    Pyrate
    Joined:
    Jan 10, 2017
    Likes Received:
    8,973
    Trophy Points:
    113
    Location:
    Denver CO
    Home Page:
    Yep.

    And I figured I was making a stink about it in the other thread, so I better put up or shut up.

    Since you showed me such an affordable means to figure it out, I couldn't pass it up. I figure I'll have fun fidgeting around with it in other ways too. So, thanks Bob
     
  13. Ash1412

    Ash1412 Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    667
    Trophy Points:
    93
    My results at -70db. Looks like mines v1:
    upload_2018-7-21_5-40-22.png
     
  14. Hekeli

    Hekeli Facebook Friend

    Contributor
    Joined:
    Sep 28, 2015
    Likes Received:
    149
    Trophy Points:
    43
    Location:
    Finland
    I happen to have UCA202 around, so here for reference Gungnir Multibit that had mobo changed in Feb/2018. No glitch, but I guess it's not surprising since it was supposedly fixed ages ago.

    -90dB
    [​IMG]

    -70dB (0.000316228)
    [​IMG]
     
  15. ergopower

    ergopower Friend

    Pyrate
    Joined:
    Mar 21, 2018
    Likes Received:
    815
    Trophy Points:
    93
    Location:
    South Central PA
    Received my UFO202 today (has a MM input for same $30 price). Set it up and got this:

    [​IMG]
    [​IMG]

    So, yay! Pretty happy camper, this was a B-stock bought in April or May.
    Note I originally had trigger set to Normal, but took a couple of 'snapshots' set to Single to get a good clear one.

    This is the reason I joined this site. Actual, usable science-y stuff.
    This was pretty simple to do, @atomicbob, you rock
     
  16. ergopower

    ergopower Friend

    Pyrate
    Joined:
    Mar 21, 2018
    Likes Received:
    815
    Trophy Points:
    93
    Location:
    South Central PA
    It's a slow Saturday evening. I decided to repeat the measurement putting a budget decrapifier into the chain. For this measurement, USB out is to Brookstone Big Blue amplifier, which can take USB input and output SPDIF; to SPDIF in of Modi Multibit. I think it looks better, but waiting for smarter folks to tell me what's going on here.

    [​IMG]

    Whisky glass is empty, off to bed shortly.
     
  17. gixxerwimp

    gixxerwimp Professional tricycle rider

    Pyrate
    Joined:
    Sep 30, 2015
    Likes Received:
    5,773
    Trophy Points:
    113
    Location:
    small island claimed by China
    I also am not getting the same amplitudes as Bob and Elnrik, but I think they're a bit higher than yours.

    Glitch measurements spurred me to try this again, with the hope that it would be visible above the noise in my system. And I figured I'd stick the Eitr in as well, which seems to have helped. I also unplugged my Ethernet cable, but that didn't make any difference. It looks like I'm getting some DC fluctuation as the curve "offsets" up and down a lot. I grabbed some single triggers and it looks like the 440Hz signal is coming through at -90dB, though there's a lot of noise that seems to be correlated with the top of the sine wave.

    Edit: Signal generator settings labels fixed.

    Cleanest 0.0000316 I could capture (-90dB)
    Eitr-Mimby2.jpg

    Worst looking 0.0000316 (-90dB)
    Eitr-Mimby4.jpg

    Two 0.000316288 captures (-70dB)
    Glitch1.jpg
    Glitch2.jpg

    It doesn't look like there are consistent spikes at the zero crossing.

    @atomicbob would you mind looking at these again?
     
    Last edited: Jul 22, 2018
  18. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    Ok, time for a few words about A/D and D/A converter input line-ups. There will be some math but nothing complicated.

    We will begin with DACs.
    Consider various manufacturers DAC specifications for 0 dBFS output. We find nominal values of
    Vrms: 1.1, 1.23, 2.00, 3.46, 4.36, 6.91, 12.28, etc.
    corresponding to
    dBu: +3, +4, +8.23, +13, +15, +19, +24, etc.

    There are more, but you get the idea. With all those DAC output line-ups the amount of voltage per bit varies over quite a range. Consider the difference for 16 bit representation between a DAC with 0 dBFS = 1.1 Vrms and a DAC with 0 dBFS = 3.46 Vrms.

    Conversion from Vrms to Vpp (for sinusoids only)
    Vpp = Vrms * sqrt(2) * 2
    We often approximately sqrt(2) as 1.414.

    Voltage Peak to Peak for the first is approximately 3.11 Vpp and given a 16 bit system is 3.11 / 65536 = 47.45 uV / bit
    In a similar calculation for the second DAC with 0 DBFS = 3.46 Vrms
    Vpp = 9.79
    9.79 / 65536 = 149.4 uV / bit

    The first DAC changes voltage by approximately 47.5 uV / bit while the second changes voltage by approximately 149 uV / bit or roughly a factor of 3x.

    The Modi MB is nominally 2 Vrms for 0 dBFS
    2 Vrms = 5.656 Vpp
    5.656 / 65536 = 86.3 uV / bit

    Now consider the same applies to A/D converters for reference as to what voltage is necessary to produce 0 dBFS.

    In the case of the UCA202 with Windows record gain set for 100, it only takes 40 mVrms to produce 0 dBFS. 40 mVrms = 113.1 mVpp.
    Vpp = 113.1 mV
    113.1 / 65563 = 1.726 uV / bit

    Now for the Realtek ALC888 codec the specification lists 1.1 Vrms typical to produce 0 dBFS.
    1.1 Vrms = 3.11 Vpp
    3.11 / 65536 = 47.45 uV / bit

    Now with voltage / bits known we have the following
    Modi MB 86.3 uV / bit
    UCA202 1.726 uV / bit
    ALC888 47.45 uV / bit

    Sending -90 dBFS sine to Modi MB only changes the output by ± 1 bit producing 172.6 uV which is sent to the input of the A/D.
    UCA202 will produce ±50 bits of change while
    ALC888 will only produce ±5 bits of change.

    The Soundcard Scope program works with bits. You have to determine what those bits represent based on the A/D uV / bit as discussed above. Thus ALC888 has 10x fewer bits to represent the Modi MB output than the UCA202.

    Finally, both A/D converters have roughly 90 dB SNR. They are both at their performance limits at -90 dBFS for their respective systems. Slight amounts of mains noise or EMI / RFI may easily contaminate the measurements.

    Any math errors above are mine, possibly due enjoying a glass of wine. Please point out any errors and I will correct.
     
  19. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    @gixxerwimp - I think you have the signal generator settings labels reversed on your screenshots. As far as the results, I still believe you have v1 firmware, though it may be possible your A/D is not up to the task as described in the post above.

    *edit* looks better now with signal generator settings labels corrected, thank-you.
     
    Last edited: Jul 23, 2018
  20. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Looks like v1 to me too.
     

Share This Page