yggdrasil technical measurements

[jfoxvol]

Science on headfi. At last. I'm duly impressed with these measurements. Thanks for all the work. Much appreciated.

 

[prot]

Science on headfi. At last. I'm duly impressed with these measurements. Thanks for all the work. Much appreciated.

+100

Congratulations atomicbob and keep em coming please.

 

[DrKC]

I noticed that the dynamic range was limited by what appeared to be a 120Hz power supply spur.  Are the rest of those at higher frequencies harmonically related?  Without those, it seems to approach 128dB.  That's pretty amazing.

 

[jcx]

its expected that a "true 20 bit DAC" that can run from +/- 10 V references should have excellent audio bandwidth S/N
 
for the raw AD5791 numbers you see 7.5 nV/rtHz on the datasheet - for conventional audio 7.5 nV * sqrt(20 kHz) ~= 1 uVrms  noise vs ~7 Vrms max sine output for about 136 dB !
 
of course that's S/N for a static output, before trying to output audio signal, overcome the glitch problem - the dynamic spec numbers on the datasheet for the single DAC, no S/H are awful for audio
 
 
I would like to see better numbers, circuit details - when do we get a "tear down" review? - and would Head-Fi allow it, or even links
 

 

[rickles]

Science on headfi. At last. I'm duly impressed with these measurements. Thanks for all the work. Much appreciated.

^^^^

 

[tekprimehifi]

I live in Florida and the ambient temperature right now is ~~90f.  I think I know a faster way to bring this baby up to operating temps. Hell the ups truck isn;t ac'd it might come warm.

 

[BirdManOfCT]

  I live in Florida and the ambient temperature right now is ~~90f.  I think I know a faster way to bring this baby up to operating temps. Hell the ups truck isn;t ac'd it might come warm.

LOL!

 

[citraian]

Hey, you said that "The THD+N graph measures artificially high due to a driver / OS interaction on my computer."
Are there any news on this? Also, what is the driver / OS interaction that you were talking about?
Thanks

 

[atomicbob]

  Hey, you said that "The THD+N graph measures artificially high due to a driver / OS interaction on my computer."
Are there any news on this? Also, what is the driver / OS interaction that you were talking about?
Thanks

I haven't returned to the Ygg in awhile. Pesky day job absorbed 140% of my time for several months. That is hopefully behind me now and I can return to the measurement activities. I have made some for the other Schiit DACs and will post them soon.

Windows has this very annoying thing called the K-mixer in the audio paths depending on how an application accesses a given audio driver. Some applications are more sophisticated at avoiding the K-mixer than others. The day those measurements were made I had managed to avoid the Windows "helper" for all but the THD+N and THD measurements. With the K-mixer in the path there was an artificial clipping of the digital output at -0.1 dBFS. Those measurements would not be affected if the output is set for -0.2 dBFS.

 

[citraian]

  I haven't returned to the Ygg in awhile. Pesky day job absorbed 140% of my time for several months. That is hopefully behind me now and I can return to the measurement activities. I have made some for the other Schiit DACs and will post them soon.

Windows has this very annoying thing called the K-mixer in the audio paths depending on how an application accesses a given audio driver. Some applications are more sophisticated at avoiding the K-mixer than others. The day those measurements were made I had managed to avoid the Windows "helper" for all but the THD+N and THD measurements. With the K-mixer in the path there was an artificial clipping of the digital output at -0.1 dBFS. Those measurements would not be affected if the output is set for -0.2 dBFS.

Did this driver / OS interaction you encountered affect the THD when listening as well?

 

[atomicbob]

  Did this driver / OS interaction you encountered affect the THD when listening as well?

Depends on the player. JRiver has multiple ways to connect to USB drivers and can avoid such issues completely. So can foobar2000. Other players may or may not be as adept. Thus my tendency to use only JRiver and foobar2000 for Windows music players. As far as I know, this OS / driver interaction is not a problem on Mac or linux. Only MS Windows is "blessed".

 

[ZoNtO]

@atomicbob Any chance you ever did or would consider doing a full slew of single-ended Yggy measurements like you did with the Gumby?

 

[bballas]

its good?

 

[atomicbob]

@atomicbob Any chance you ever did or would consider doing a full slew of single-ended Yggy measurements like you did with the Gumby?

Yes, this is in my queue along with correcting the THD measurements. I've been busy with Gungnir MB and Bifrost MB and a new Cavalli Liquid Crimson. Personally I've become enchanted with the Bifrost MB performance and spend more time listening and less time measuring with my discretionary time.

 

[atomicbob]

  its good?

Yes, it is very very good.