Schiit Yggdrasil ‘OG’ Measurements (One month warmup test)

Test Setup:

– Audio Precision APx555 B-Series Analyzer with 200kOhm input impedance
– Audioquest Mackenzie interconnects
– USB Source: Intel PC via ifi iGalvanic 3.0 and iUSB 3.0
– Analyzer galvanically isolated via Intona 7055-D USB 3.0 isolator
– Audioquest Carbon USB and SPDIF cables
– Measurement system and device being tested running on an Audioquest Niagara 5000 power conditioner
– All measurements shown in this post are taken with USB input, XLR output, 44.1khz sample rate unless otherwise specified
– DAC was warmed up for one month before measuring. But measurements from cold, 7hr warmup, and one week warmup are included
– Exact analyzer/filter configurations for each measurement are detailed in the full reports

Content and analysis made possible thanks to support from https://headphones.com, and https://patreon.com/goldensound supporters

ChitChat:

The Schiit yggdrasil is a fairly unique DAC, both in how it sounds, but also in the design and some of the stories about it.
It’s a multibit/R2R dac using an Analog Devices chip, and a discrete output stage, though there are now two other versions available, a ‘More is Less’ version which in Schiit’s view, measures better but sounds worse. Click here for measurements of that.
And a ‘Less is more’ version which apparently measures worse but sounds better, or rather, some people prefer the ‘LIM’ and some people prefer OG.
I’ll be posting measurements and a review of all three once I get my hands on the LIM.

This is the ‘Analog 2’ variant, but with Schiit’s ‘Gen 5’ USB, not Unison.
Unison should not impact any DAC measurements other than Jitter however so these measurements should still be applicable.

This unit was loaned to me by a friend.

One thing to mention is that this DAC has a reputation for having an insanely long warmup time, as in, several days or even weeks some would say. Because of this I’ve included measurements at several points in the warmup stage, reports can be found below, but the measurements displayed in this page are from 1 month of warmup.

Additionally, this DAC has quite a distinctive sound that sets it apart from other DACs in my opinion. It’s hard to say if it sounds ‘better’ or ‘worse’ than various other options, but it is certainly different and very enjoyable! Interestingly there are a few measurable behaviour characteristics that could explain this.

Full Measurement Reports:

Reports available here:

Measurements:

Dynamic Range (AES17): 120.7dB

SNR: 124.2dB

Noise Level RMS (20-20khz): 2.660uVrms

Noise Level RMS (20-90khz): 4.171uVrms

DC Offset: 38.52uV Active, 36.89uV idle

1khz 0dBFS:

Harmonic distortion is fairly high

1khz -3dBFS:

THD+N / Frequency:

THD+N looks fairly flat across the FR, though if the ADC sample rate is raised in order to factor in ultrasonics, there is a rise as the output signal of the DAC gets higher in frequency.

Linearity:

Linearity is quite good!

SINAD/THD+N vs output level:

This is where things start to get very interesting. Most delta sigma DACs have a nearly straight THD+N vs level graph. However many R2R DACs show some ‘sawtoothing’.
On the Yggy however the shape of the graph is quite unusual, and distortion sees some fairly unpredictable and unusual behaviour at higher levels. This could potentially go some way to explaining the ‘intense’ macrodynamics that many find the Yggdrasil to exhibit?

IMD (SMPTE) vs Output Level:

IMD also gets a bit higher once the level gets close to full scale.

Crosstalk:

Crosstalk is quite good, this hump is due to the 50hz PSU leakage which is present in most linear power supplies, not actual crosstalk.

-90.31dBfs 1khz Sine (Recorded with 192khz ADC Sample Rate):

Another bit of interesting behaviour here! There is a fair bit more noise/distortion at the peak of the wave, but not the trough or transitional areas. I’m not sure what effect this could have on sound, but the key thing to mention is that behaviour such as this is not clearly shown on an FFT as an FFT relies on longer capture periods, and will not determine between constant behaviour, or periodic behaviour such as this if it is shorter in duration than the FFT length. It’s important to look at a scope view as well as an FFT.

Filter Ultrasonic Attenuation:

Impulse Response:

Idle Noise Spectrum upto 1.2Mhz:

44.1khz Jitter (USB):

Some jitter products visible, but given the behaviour seen on the Yggdrasil MIL and other DACs with unison USB, it’s likely that a Yggdrasil with unison would have better performance here so don’t pay too much attention to this.

48khz Jitter (USB):

Multitone:

NOTE: I am using a new, higher FFT length method for this test.

But what about the warmup?

Well, I’ve provided reports of the several stages of warmup, and in looking through them, particularly comparing the 7hr to 1 month warmup, I was not able to find any differences at all in any of the tests I did.
Subjectively, I’m not really wanting to comment on it for the simple reason that a month is simply far too long to make a comparison on small details by memory. If I was able to get a ‘warmed up’ Yggy and a ‘cold’ Yggy side by side that’d be very interesting, but I do not feel that comparing things I heard a month apart is in any way reliable at all so I’d prefer not to do so.

If anyone spots anything in the reports that I missed though please do let me know and I’ll update this post.