Schiit Yggdrasil V2 upgrade Technical Measurements

If you are unfamiliar with audio measurements please use a search engine with the query:
"audio measurements" or "audio measurement handbook"
Look for publications by Richard C. Cabot and also by Bob Metzler, both from Audio Precision. There are other useful publications as well. These will provide basic knowledge.
Interpretation of the following measurements is beyond the scope of this post.

The data presented were acquired as follows:

1. PrismSound dScope III, picoscope 5243B
2. DAC balanced (Bal) output XLR and unbalanced (SE) output RCA
3. 100 Kohm load used for measurements
4. dScope analyzer sample rate 48 KHz unless otherwise noted
5. DAC 44.1 KHz sample rate, 24 bit depth unless otherwise noted
6. USB cables – Schiit Pyst and Audioquest Forest
7. SPDIF input - Tecnec 75 ohm SMPTE 259M/292M Digital Broadcast Cable
8. Unbalance RCA cable - Worlds Best Cable Gotham GAC-2 or DH Labs Silver Sonic Air Matrix
9. Balanced XLR cable - Canare L-4E6S starquad with Neutrik XLR connectors
10. Vaunix Lab Brick USB hub
11. Shielded 14AWG and 16AWG power cables

Schiit yggdrasil V2 (updated from V1)
Measurements commenced after 2 hours of warmup post installing upgrade PCBs.
Measurements were performed over a period of 8 hours.

Notable highlights:
The glitch in V1 previously visible at -90 dBFS is gone in V2.
With V2 Schiit has set a new reference for multiplying multibit DAC Gain Linearity even though V1 Gain Linearity was previously quite good for a multiplying multibit architecture DAC.

@Torq impressions found here

WARNING - PERSONAL OPINION FOLLOWS:
Whether either the glitch at -90 dBFS or gain linearity deviation below -80 dBFS in V1 was audible to most of the listening population is highly debatable in my opinion.

Index
Post 1 - measurement setup description, highlights, comparison graphs V1 - V2
Post 2 - SPDIF input - Balanced outputs part A
Post 3 - SPDIF input - Balanced outputs part B
Post 4 - SPDIF input - SE outputs part A
Post 5 - SPDIF input - SE outputs part B
Post 6 - SPDIF input - Balanced outputs 1 KHz resolution series
Post 7 - SPDIF input - SE outputs 1 KHz resolutions series

tl;dr

Here are the measurements of the notable highlights for SPDIF input Balanced output

1kHz @ -90dBFS


1 kHz gain linearity plotted as a transfer function


1 KHz gain linearity ploted as deviation from the transfer function


yggdrasil V1 before the operation


yggdrasil V2 after update


yggdrasil V2 closeup of analog PCB


Updates included Analog V2, G5 USB, DSP firmware

 

SPDIF input balanced output measurements part A

THD and THD+N vs Frequency


THD and THD+N vs amplitude at 1 KHz


IMD


Frequency Response


Frequency Response (y-axis zoom)


Inferred Jitter


1 KHz gain linearity


Imaging


Dynamic Range


Crosstalk

 

SPDIF input balanced output measurements part B

50 Hz + 7000 Hz dual tone


600 Hz + 1700 Hz dual tone


19 + 20 KHz IMD


1kHz @ -90dBFS


1kHz @ -70dBFS


1 KHz 0 dBFS spectrum


20 Hz square wave 0dBFS 6Vpp 10mS/div


20 Hz square wave 0dBFS 6Vpp 100uS/div

 

SPDIF input SE output measurements part A

THD and THD+N vs Frequency


THD and THD+N vs amplitude at 1 KHz


IMD


Frequency Response


Frequency Response (y-axis zoom)


Inferred Jitter


1 KHz gain linearity


Imaging


Dynamic Range


Crosstalk

 

SPDIF input SE output measurements part B

50 Hz + 7000 Hz dual tone


600 Hz + 1700 Hz dual tone


19 + 20 KHz IMD


1kHz @ -90dBFS


1kHz @ -70dBFS


1 KHz 0 dBFS spectrum


20 Hz square wave 0dBFS 6Vpp 10mS/div


20 Hz square wave 0dBFS 6Vpp 500uS/div

 

SPDIF input - balanced outputs - 1 KHz resolution
DAC equivalent of the limbo - how low can you go

1 KHz sine 0 dBFS 16K FFT


1 KHz sine -70 dBFS 16K FFT


1 KHz sine -90 dBFS 16K FFT


1 KHz sine -100 dBFS 16K FFT


1 KHz sine -110 dBFS 16K FFT


1 KHz sine -120 dBFS 16K FFT


1 KHz sine -130 dBFS 16K FFT


1 KHz sine -140 dBFS 16K FFT


1 KHz sine -144 dBFS 16K FFT


1 KHz sine -145 dBFS 16K FFT


Damned impressive for a multiplying multibit architecture DAC.
Well done Schiit!!

 

SPDIF input - SE outputs - 1 KHz resolution
DAC equivalent of the limbo - how low can you go

1 KHz sine 0 dBFS 16K FFT


1 KHz sine -70 dBFS 16K FFT


1 KHz sine -90 dBFS 16K FFT


1 KHz sine -100 dBFS 16K FFT


1 KHz sine -110 dBFS 16K FFT


1 KHz sine -120 dBFS 16K FFT


1 KHz sine -130 dBFS 16K FFT


1 KHz sine -134 dBFS 16K FFT


1 KHz sine -135 dBFS 16K FFT


While not quite as low as balanced output, this is still impressive for single ended outputs (SE).
Well done Schiit!!

 

reserved 8/8

 

Is that even the same DAC?!

 

This is really boding well for the Gungnir Multibit Analog upgrade (whenever that's going to be announced.)

 

No kidding, inaudible or not, that is an impressive improvement in measurements.

As always, thanks for all the measurements @atomicbob.

 

Schiiiiiiiiit thats good!

Will you be able to do temperature tracking measurements with your automated setup referenced in another post? Warm up time should be considerably shorter with the new output stage and I'd be curious to see what the results are.

 

Mixed bag, but that's no surprise to me.

Channel Cross better, THD+N better
IMD worse, THD a little worse, SNR about the same, slightly worse averaged over both channels,

Interesting how THD+N is lower than THD on Channel A on V2.

 

Huge thanks to @atomicbob for the time and effort involved in doing these properly. Certain nameless NW armchair experts could learn a lot by applying the same rigor you use. These measurements also tell a very different story.

 

Noise of measurements on any given day could account for 1 to 2 dB. Be careful not to spit hairs over minutiae. The measurements from V1 were made in 2015.

 

Thank you @atomicbob

And now I am looking forward to hearing it (when my turn in the queue bubbles to the top).

 

True, with the f/w fixing the glitch, the v1 likely measures better on some things.

 

Great works!

I have a couple of (probably) very silly questions.

Q1. Are these results comparable with your previous post about Yggdrasil v1?

Comparing measurement conditions between both posts, found slight disparities (Y1: Yggdrasil v1, Y2: Yggdrasil v2)

• Y1: picoscope 2205 vs Y2: picoscope 5243B
• Y1: cables to go USB cable vs Y2: Schiit Pyst and AQ Forest
• Y2 post (this one) specified more information such as power cable or unbalanced rca cable (I am assuming you used the same ones for Y1 though)

I have almost no idea whether these differences matter or not. But your another post of usb cables led me to worry about it.


Q2. What is your interpretation between the jitter graph of Y1 and that of Y2? Y2 seems to control jitters better overall, but noise level around 11khz is much higher with Y2, too.

 

@Vtory

All the measurements posted are SPDIF input so I don't know where the USB cable is used but it doesn't seem super relevant. Probably just to connect the scope to a laptop for general operation and to transfer data.

Different measurement hardware is a bigger deal but as long as both as calibrated correctly the data comparison should be relevant.

 

Probably post #8, considering he's upgraded to G5 USB.