Review and Measurements of Schiit Yggdrasil V2 DAC

[derp1n]

I already have. It was in the review.

I believe he's asking you to measure it again now, to see if the jitter has changed over the week or so "warm up". But it sounds like you've switched the thing off, so it's too late for that. Unfortunate.

As an aside, the "warm up" discussion on Head-Fi shows how carefully atomicbob choose his battles. The discussion was asking for evidence or explanation of the supposedly very audible Yggdrasil changes over the 1-3 week warm up period (you can easily find people claiming a cold one has "no bass", "thin and grainy", "like garbage out of the box", "congested mess", "warm gooey shit", etc.). Only subjective responses from true believers. Then atomicbob jumps in, "proving" with graphs and references to AD data sheets that the DAC is temperature sensitive... sure, at -155 dB or whatever inaudible level... "confirming" the subjective posts and avoiding the crux of the question without posting anything actively dishonest.

 

[amirm]

I did suggest that on June 26th as part of the regular measurements to assess the warmup effect in our PM thread:

Sorry I misunderstood what you said in the thread. I thought we were done with the experiment as you originally stated once the week-long test was over.

I am actually trying to complete other experiments before I have to send this to you in a couple of days. One of them as you see shortly, required full shut down.

 

[amirm]

As I implied above, there is talk on the head-fi Yggdrasil thread that even a 30 minute warm up helps the sound of this DAC. So I set out to test that theory by shutting down the Yggdrasil after two weeks of running and let it come down to room temp.

I then powered it up and ran it for a little over an hour while take a measurement of THD+N every second. THD+N encompasses any noise from the product and its non-linearities/transfer function. Any change in behavior of the device would change THD+N.

To speed up the temp change and aggravate the situation more, I even applied generous amount of heat to the enclosure on all surfaces starting around 7:00 minutes. The top was so hot I could barely hold my hand over it. An IR scan of the top of the unit showed this:

Due to reflectivity index of the case, it is under-reporting the temperature. It was quite a bit hotter but still, you get the idea.

Here are the results:

Yup. Nothing happened. The little ups and downs are naturally occuring and are due to variations in both the DAC and measurement system.

BTW, the above is at -20 dB signal so don't compare it to my previous measurements (THD+N gets worse with smaller signals).

Here is the other thing: if there were going to be an effect, it would require a DAC with a far better measured performance to matter. The Schiit Yggdrasil has some 30 dB higher noise and distortion than other state-of-the-art DACs. Its distortion and noise would swamp any variability in temperatures.

And oh, electronic noise goes up proportional with temperature. So the notion that "warming up" is good for these things, is exactly the opposite of what you want. The best DAC in the world would be one frozen in space! Look up thermal noise: https://en.wikipedia.org/wiki/Johnson–Nyquist_noise

Summary
We have done long term testing that showed nothing from linearity to SINAD and distortion products changes in any material way. See this post: https://www.audiosciencereview.com/...ts-of-schiit-yggdrasil-v2-dac.3607/post-86400

In this post, I have shown effects of short-term warm-up which is nil.

No way, no how this device changes its sound in the short-term, or long term.

 

[trl]

Amir, if owner will allow you and your willingness, kindness and time will allow that, could you please open up this DAC, unscrew the core-transformer and re-position it with 45 to 90 degrees, so the mains hum to disappear? ARTA or some sort of audio wave analyzer should be used to check in real-time the output sound spectrum while re-arranging the transformer.

From experience, hum might get much lower after re-positioning the transformer or after moving it away from the analog parts (especially from the potentiometer or cables).

 

[amirm]

Amir, if owner will allow you and your willingness, kindness and time will allow that, could you please open up this DAC, unscrew the core-transformer and re-position it with 45 to 90 degrees, so the mains hum to disappear? ARTA or some sort of audio wave analyzer should be used to check in real-time the output sound spectrum while re-arranging the transformer.

The transformers are all soldered to the PCB so cannot be rotated. Here is a picture from my teardown of the last Yggdrasil I test (but have not yet posted):

Notice how the front transformer has two bolts going through it even though it is designed to have four of them. Now look at the one next to it: it has no bolts at all! Someone was pinching pennies....

 

[Alcophone]

To speed up the temp change and aggravate the situation more, I even applied generous amount of heat to the enclosure on all surfaces starting around 7:00 minutes. The top was so hot I could barely hold my hand over it.

And it didn't occur to you to ask me for permission before you do that? Seriously?
Please turn it off now, box everything up the way I packed it and send it on its way, with FedEx, by Friday. I'll PM you the address.
Not cool. Literally.

 

[trl]

Amir, Grain Oriented Sillicon Steel or at least double-sided thick PCB could be used between trafos and electronic components to munumize the EMI field interaction. It would be nice to see an improvement to how this DAC measures.

 

[Blumlein 88]

Amir,

Seeing as how the Yggy is so popular and all, could you redesign or at least re-assemble it in a way that it deserves its accolades and price instead of under-performing inexpensive Chinese gear. Seems like the least you could do as part of testing. I had assumed you understood it was your responsibility as a tester to fix any surprisingly poor gear until it is good. You didn't know that? Haven't you noticed other people who post public test results doing such things? I mean come on, you really owe it to us all.

 

[amirm]

And it didn't occur to you to ask me for permission before you do that? Seriously?
Please turn it off now, box everything up the way I packed it and send it on its way, with FedEx, by Friday. I'll PM you the address.
Not cool. Literally.

What is not cool? Warming the unit up? This is routine testing. The unit better have been tested in temperature chamber to far higher temps than I took it. Just sun shining on it will cook it to much hotter temps. And here, I just heated the outside for a couple of minutes. Not the internals. The picture I showed was from a different unit, not yours.

I always take full responsibility for anything I test and if i damage it, I will buy it. So please don't assume that I subjected your box to any danger. I would gladly buy it from you if you think that has occured. Just let me know.

But anyway, I will box it up and send it to you tomorrow....

 

[amirm]

Amir, Grain Oriented Sillicon Steel or at least double-sided thick PCB could be used between trafos and electronic components to munumize the EMI field interaction. It would be nice to see an improvement to how this DAC measures.

Definitely not my choice of design. It has a large motherboard with other boards plugging into it. I don't like to see any connectors like that. Worse yet signals route around the motherboard to relays and such. This allows for more noise pick up.

 

[rebbiputzmaker]

Definitely not my choice of design. It has a large motherboard with other boards plugging into it. I don't like to see any connectors like that. Worse yet signals route around the motherboard to relays and such. This allows for more noise pick up.

Do you have any better examples of how it should be done? How would you design it?

 

[Wombat]

Do you have any better examples of how it should be done? How would you design it?

What is not clear in Amir' statement. Sheesh.

TEXAS INSTRUMENTS website is a mine of information. Just Google. Tell Schitt about it, too.

 

[rebbiputzmaker]

What is not clear in Amir' statement. Sheesh.

TEXAS INSTRUMENTS website is a mine of information. Just Google. Tell Schitt about it, too.

Maybe it would be interesting for some members to see some better audio designs. If you are not interested just don't look, instead of trying to control and stifle conversation.

 

[Wombat]

Maybe it would be interesting for some members to see some better audio designs. If you are not interested just don't look, instead of trying to control and stifle conversation.

Sorry. I thought you were Schiit defending again, as usual, in your obverse style.

Somewhere on this forum there is a very detailed TI guide for DAC designers that I posted. It would answer your question and more. I can't find it, unfortunately.
Asking someone how they would design a DAC, given the complexity, is not a reasonable question and you should know that.

I would love to see you add to discussion but you are generally dismissive of factual comment.

 

[Liberal Democracy]

Amir,

Seeing as how the Yggy is so popular and all, could you redesign or at least re-assemble it in a way that it deserves its accolades and price instead of under-performing inexpensive Chinese gear. Seems like the least you could do as part of testing. I had assumed you understood it was your responsibility as a tester to fix any surprisingly poor gear until it is good. You didn't know that? Haven't you noticed other people who post public test results doing such things? I mean come on, you really owe it to us all.

View attachment 14002

I wonder if amir could ‘test’ my mums washing machine, it’s making bad noises ..

Damn no good deed goes unpunished.

 

[Wombat]

I wonder if amir could ‘test’ my mums washing machine, it’s making bad noises ..

Damn no good deed goes unpunished.

Does she still wash your undies?

 

[Blumlein 88]

I wonder if amir could ‘test’ my mums washing machine, it’s making bad noises ..

Damn no good deed goes unpunished.

Imagine that. My sister's dryer is having troubles too. It is all USA made being 32 years old, but is not a Schiit dryer.

 

[DonH56]

I've gotten to really hate these threads and deplore the amount of time wasted that could be better spent testing other gear.

The commercial product temperature range is 0 to 70 deg C, or 32 to 158 deg F. Amir's measurements are well under that. Too hot to touch is more like 120-140 degF or so IIRC.

Since we're on the subject, a quick search indicates the water in a washer on the hot cycle runs around 140 deg F, and the inside of a dryer runs about 120 - 150 deg F.

 

[Liberal Democracy]

I've gotten to really hate these threads and deplore the amount of time wasted that could be better spent testing other gear.

The commercial product temperature range is 0 to 70 deg C, or 32 to 158 deg F. Amir's measurements are well under that. Too hot to touch is more like 120-140 degF or so IIRC.

Since we're on the subject, a quick search indicates the water in a washer on the hot cycle runs around 140 deg F, and the inside of a dryer runs about 120 - 150 deg F.

Difficult to separate popular with populism, I would like to see us give schiit a rest as it seems to take discussion back to the Middle Ages and really it all gets rather tedious not to mention outright rude wrt how many seem to treat amir. I have at times felt rather sad about that as though stubborn ( on balance a very successful character trait for him lol) as hell hes truly decent and is doing the whole Audio industry/ community a huge service in what he’s trying to do here and how he’s going about it.

Still I guess in a way that would be giving in, the Audiophile world is more religion than anything else. If we want to help change that we are going to have to put our tin hats on and dig in.

Possibly the kind of dynamic we are encountering can be explained by this ..,
http://www.iflscience.com/brain/con...ience-down-warring-brain-networks-study-says/

 

[amirm]

Do you have any better examples of how it should be done? How would you design it?

By far the most common method is a single board with separate power supply. Here is the Benchmark DAC3 for example: