Measurements and Review of Schiit Yggdrasil DAC

[DonH56]

32 bits ~ 194.432 dB SNR. The gap between number of bits in and actual measured output performance is one of the reasons that drove the IEEE and other standards bodies to push for ENOB. That said there are good reasons for maintaining much greater number of bits through the filters, like keeping dynamic range maximized through the digital processing since per-stage levels (dynamic range) can vary widely and not always completely predictably (in a practical sense, because it is signal-dependent).

 

[amirm]

@amirm

Is the linearity test performed with or without dither?

TPDF dither is used.

 

[amirm]

How would the results compare between the two choices?

Which two choices?

 

[RayDunzl]

Which two choices?

With or without dither...

Maybe measure one with and without dither as an "educational example" sometime.

Anyway, the more pertinent question is answered.

 

[amirm]

With or without dither...

I don't have the unit anymore so I can't test it anymore. These are 24-bit samples and we are looking way upstream of its LSBs in these measurements.

 

[Jimster480]

I see some people on massdrop still fighting over these measurements... trying to discredit Amir's measurements and conclusions.

 

[Wombat]

I see some people on massdrop still fighting over these measurements... trying to discredit Amir's measurements and conclusions.

I started a new thread on O2 unlicensed changes. Measurements may vary depending on which variant is tested.

 

[Jimster480]

Link me?

 

[VanNeumann]

I have been reading Peter Aczel's old articles in The Audio Critic. Here is what he thought about burning in:

The Burn-in Lie
"This widely reiterated piece of B.S. would have you believe that audio electronics, and even cables, will "sound better" after a burn-in period of days or weeks or months (yes, months). Pure garbage. Capacitors will "form" in a matter of seconds after power-on. Bias will stabilize in a matter of minutes (and shouldn't be all that critical in well-designed equip- ment, to begin with). There is ab- solutely no difference in performance between a correctly designed ampli- fier's (or preamp's or CD player's) first-hour and l000th-hour performance. As for cables, yecch... We're dealing with audiophile voodoo here rather than science. (See also the Duo-Tech review in Issue No. 19, page 36.)

Loudspeakers, however, may re- quire a break-in period of a few hours, perhaps even a day or two, before reaching optimum performance. That's because they are mechanical devices with moving parts under stress that need to settle in. (The same is true of reciprocating engines and firearms.) That doesn't mean a good loudspeaker won't "sound good" right out of the box, any more than a new car with 10 miles on it won't be good to drive. "

https://www.biline.ca/audio_critic/mags/The_Audio_Critic_26_r.pdf (page of the article, page 7 of the pdf)

 

[FunctionalDoc]

they won't respond to emails, I have emailed them about the issues with my USB interface with the Fulla2 before and they full on blamed it on my PC despite whatever emails I sent them.
This is after they replaced my "defective" unit with a "new" unit and the "new" unit had exactly the same problems.

My posting abilities at headfi were revoked after I found the SMSL xUSB and posted about it around the forum. Especially to people asking for a USB->SPDIF converter... because I wasn't recommending the Schiit product my posts kept getting deleted. When people would reply with the Eitr and I told them that the xUSB could do the same thing... my posts were edited or deleted again with warnings from mods that my information cannot be backed up and that I haven't purchased both units to test side by side. Because apparently if you recommend something that isn't a schiit product there needs to be proof that it works with measurements that say that it does the same thing??????? Honestly it made no sense to me and within a few days of recommending it in 5 different threads or so my posting abilities in every thread I had mentioned it in were removed.
There was a guy from ARK Audio who posted in the Magni 3 thread and asked Jason (or whoever was manning that thread) about the voltage numbers in Magni 3 and said that the voltage of their wall wart didn't allow for the voltage they were claiming in output unless they had multiplied the voltage several times over which could cause instability or something along those lines.
They were quick to erase his posts and say that he shouldn't be posting technical information in the thread.

So if you post technical information about schiit products = you are banned from the thread
If you post recommendations to non-schiit products = your posts are deleted and you are later banned from the thread for not providing measurements

They even deleted posts of mine recommending the D30 or M8 in several threads when people were asking about DACs at $250 or less because I said they were better than mimby in my experiences.
My posts were removed for "no conclusive evidence to prove that these DAC's are better than Modi Multibit"... so once again recommending a non-schiit product requires evidence. Recommending a schiit product based on nothing at all but subjectivity = Expected behavior.
As I developed a following on the forum and became more well known they were curating my posts every single day based on me making recommendations for anything people asked me about. In the mrspeakers Aeon thread some people were writing about how the Fulla 2 doesn't work well with the Aeon and the soundstage sounds compressed (I also tested this and found it to sound very poor and sometimes experience a buzzing noise). When I referenced these posts to someone who had an Aeon and was looking at a Fulla 2... my post was deleted and the posts I referenced were also deleted. Someone else responded telling him the Fulla 2 is great... which then sparked a discussion because many people in the thread said specifically not to get it for the Aeon... now I understand that the thread got a bit off topic but within hours they deleted all the posts again except the post telling the guy that the Fulla 2 was great for the reason of "offtopic posts".

So with all that being said... I wouldn't expect anyone from Schiit to ever try to talk to someone with measurements because they themselves know that their business is a scam. They are intentionally ripping people off with the idea that they can just force their products to have good reviews by removing all objectivity and poor reviews from all threads. Another tactic they use is when a poor review is posted or someone is dissatisfied there will randomly be a few "regular posters" who will start talking about random songs they like to listen to in the thread and fill up 2 pages of offtopic nonsense (which ofcourse won't be cleaned up by the mods even if reported) so that anyone checking the thread will have to scroll back a few pages to see someones dissatisfaction.

So now I get it about your comments about Jude and Head-Fi on Massdrop. Head-Fi is paid by advertisers like all the other magazines so freedom of speech is still reeled in when your comments may offend advertisers. I understand their running a business and this this is their platform. I now that Jude will never destroy a product and as any good business owner will support his customers the manufacturers that advertise with him. I don't fault Jude for this but it is up to user to figure out the game.

I guess this is what is so nice about this platform is isn't funded by manufacturers and trying to reduce the subjectiveness out of audio. I love Amir pulling back the curtain with measurements to show that a $2K DAc does not perform on the bench as well as $400 DAC. We still have the mystery of if we can hear the difference with our headphones, amp and older ears.

I will say that Schiit are masters of creating a brand of loyal customer following. In the end buyer beware.

 

[watchnerd]

From SBAF: "Bimby does not measure badly. It measures respectably for 16-bit entry level multi-bit DAC. And the fact that Amir paints the opposite picture incorrectly, with folks actually buying that, is only the tip of the iceberg."

That's rich. I believe even their measurements showed it crapped out after 10 bits, and that's not bad?

I'm no expert, but aren't these linearity tests showing the same thing (first is atomicbob's measurement):

and amirm's:

So the Bimby has similar resolution to vinyl playback.

No wonder it sounds so 'analog'....

 

[dallasjustice]

Sorry if this is already posted here. The thread is too long for me to go back. I just noticed a full page ad Schitt took out in 5/18 Stereopile touting their “-144db, 1khz sine wave.” Here’s the link to their measurements.
http://www.superbestaudiofriends.or...rasil-v2-upgrade-technical-measurements.5820/

 

[danadam]

Sorry if this is already posted here. The thread is too long for me to go back. I just noticed a full page ad Schitt took out in 5/18 Stereopile touting their “-144db, 1khz sine wave.” Here’s the link to their measurements.
http://www.superbestaudiofriends.or...rasil-v2-upgrade-technical-measurements.5820/

It even has its own thread
https://www.audiosciencereview.com/...-yggdrasil-ads-on-stereophile-may-issue.2633/

 

[amirm]

Yes, so much for "it is about what you hear and not measurements."

 

[dallasjustice]

Maybe if they made a better product, there wouldn’t be a need to spend so much on advertising.

 

[Sal1950]

Maybe if they made a better product, there wouldn’t be a need to spend so much on advertising.

Then they wouldn't get the rave reviews and build the big fanboy base.

 

[rebbiputzmaker]

Yes, so much for "it is about what you hear and not measurements."

You really can't blame a manufacture to present a counterpoint to what they might feel as a misrepresentation of their products.

 

[amirm]

You really can't blame a manufacture to present a counterpoint to what they might feel as a misrepresentation of their products.

It is a misrepresentation in the ad to imply that any DAC, theirs or others, can remotely reproduce a -144 dB signal. As I explained in the other thread, the measurement they are showing indicates lots of other noise sources that are higher amplitude than the test tone. So whatever they think of my work, is beside the point. They are using measurements that they either don't understand or intentionally hope to mislead the reader.

Let's show this in a graph where I replicate their test conditions but using the superbly measuring RME ADI-2 DAC. For now, focus on the left side only:

We see a very clean 1 kHz tone despite hugely small signal (rightmost bit of 24-bit signal). None of the extra noise products are visible that Schiit shows with their DAC in the same situation.

But now let's look at what that same "sine wave" looks without transformation by FFT on the right. We see that it is nothing but random noise. That 1 kHz tone is somewhere in there but is dwarfed by noise. So our DAC here despite being worlds better than Yggdrasil, still cannot remotely reproduce a 24-bit signal. Laws of physics simply do not allow it.

How does the FFT then show such clean 1 kHz tone? That is because of "signal processing." The FFT is able to provide huge amount of noise reduction by spreading the noise across the full spectrum. The tone however stays put (i.e. not reduced) so we see it clearly.

Conclusions
The measurements shown by Schiit are a) not that good anyway and b) are hugely misleading to readers. No DAC in the world at room temperature and with audio bandwidth can reproduce a signal at the bottom of a 24-bit PCM sample. Various noise sources in electronics including our measurement gear make that a huge impossibility.

This is why my/industry standard linearity tests stop at just -90 dB.

 

[DonH56]

We see a very clean 1 kHz tone despite hugely small signal (rightmost bit of 24-bit signal).

Just struck me as funny. Nice post, Amir.

 

[Jinjuku]

You really can't blame a manufacture to present a counterpoint to what they might feel as a misrepresentation of their products.

You can sure as schiit blame them for not doing it in the first place....