Schiit Eitr Preview Thread

Would you go so far to say that a stock RPi + Eitr sounds the same like microRendu + Eitr?

 

@Kattefjaes Yea no worries. It is a usability issue for sure. Would definitely be better if the USB modes that that cannot be supported by SPDIF were removed, or at least truncate to 24 bits.

 

Yes.

But see the subsequent post (which I'm making separately so as to make it clear it's a general statement, not just a response to your question).

 

There are way too many questions in this thread that amount to highly-specific nervosa and, to be honest, I think most of them are completely unnecessary. Hopefully this will put some of that to rest as the situation, from my perspective, is currently very simple:

As far as I can tell, the audible results are the same regardless of the source if you're using Eitr*.

I've found no, reliably discernible, audible change in having ANY other de-crapifiers, filters, isolators or interfaces in front of Eitr. This includes RPi, Wyrd, Sonicorbiter SE**, microRendu**, Intona and JCat isolators, Jitterbug, Singxer SU-1, Auralic Aries or RedNet D16 AES.​

That's vs. using the direct USB output of a current MacBook Pro, a current Mac Pro, a Surface 4, a Dell 7600 workstation (possibly the worst USB I've ever heard), and a bunch of iOS, Android and other portable devices, into Eitr.​

And just for good measure I borrowed a Mutec MC3+ and put it after the Eitr in the chain - and that not only didn't improve things, but it made them worse.​

I've tried this both with Schiit's re-clocking DACs, the Holo Audio Spring DAC, and half a dozen other non-reclocking DACs. As long as Eitr is the last part of the chain, what comes before it does not seem to matter. This is true of some other devices too - if you have effective isolation and good enough clocks HOW you get them doesn't matter. Eitr just happens to be the most approachable and practical way to get them from USB that I'm aware of right now.

This was all part of testing I did with the prototype Eitr over several months.

If I didn't run my DACs remotely from my computers (via Roon), I personally would already have put my Auralic Aries units up for sale.

--

*For Yggdrasil and Bifrost this seems to also be true for the USB Gen 5 board.
**Even with PSUs better than anything you're going to find in the audio add-on "LPS" world.

 

I don't want to muddy the waters but if @baldr read this: I am curious if a similar device would make any sense for buffering/optimizing coax signals from cheapo CD players.

 

Huh... It's almost as if the Eitr is doing a really great job of isolating and re-clocking the digital audio signal!?!? And further... it's almost as if good isolation and re-clocking is all these devices really need to do. So strange.... </sarcasm>

 

OK, @Torq - since it seems to be open season for asking you stupid questions- is it an audible improvement to use the Eitr to feed a reclocking Schiit DAC (e.g. a Gungnir Multibit) compared to J. Random scuzzy motherboard TOSlink which it's reclocking? If so, what do you think the factors are that achieve this?

(Not asking just to be a tosser- genuinely want to know.)

 

Great, thanks a lot! That should make things so much easier.
If now there was someone who has the opportunity to compare the outcome of Eitr vs. RPi/Allo DigiOne...

 

It shouldn't - at least as long as the clock-mode LED isn't illuminating.

But "shouldn't" isn't "no".

I used to run my first Yggdrasil via TOSLINK from my Mac Pro using a Lifatec cable. At some point I compared inputs on the Yggdrasil and wind up deciding that I liked the results better via AES and any one of a number of boxes to get that output. Whether that difference is real or imagined it'd be hard to say. But I switched away from TOSLINK and ultimately settled on the Aries via AES. That's been the fundamental basis for comparison since that point.

I can only think of one, which is highly speculative, and makes some very wobbly assumptions about how certain things work with Adapticlock, S/PDIF, and physics in general. These are not things I believe, just a postulate of the might-just-be-possible if we suspend accepted reality to a sufficient degree*.

Clock issues, i.e. jitter, should be moot unless the clock-mode LED comes on. Let's assume that's an absolute.

That pretty much limits the potential for a difference in sound to the internal implementation of the TOSLINK interface and highly-improbable but perhaps-possible-at-the-limits-of-physics and crazy-theories on the effect on the decoding data and/or noise in the rest of the system. In essence, it would need to be the case that there was enough smearing/jitter in the signal, in such a form that the bi-phase S/PDIF signals weren't screwed up in a manner that affected the clock recovery but somehow resulted in sample data being read incorrectly**.

I would put the odds of that somewhere around the same levels as nervosa-afflicted randos stopping asking clearly redundant questions.

Which is to say as close to zero as makes no difference.

Otherwise, we're down to there being additional noise imparted into the system via the TOSLINK input from the DAC-side implementation itself. I'd be staggered if that was really happening. Photo-diodes are not known for radiating spurious crap, and I would assume, and expect, that the recovered electrical pulse-train was simply passed to the very same circuit that handles the electrical signals coming in from the AES and S/DPIF connections - so that'd all be common.

So, see prior emboldened statement ...

--

*For a non-reclocking DAC jitter is, of course, the primary culprit and is relatively established.
*This is, as far as I can fathom, actually impossible due to the way the bi-phase clock and data encoding actually work and if this was somehow occurring it would almost certainly result in the clock-mode LED illuminating anyway, in which case all bets are off.

 

Thanks, @Torq - that's exactly the sort of thing that I was getting at. Your discomfort is duly noted, and I have a completely straight face now.

 

Yea seriously running Win7 I just plugged it in, windows figured out how to use it, FooBar gave me a KS option [ASIO pisses my pc off] and upon first listen with the LCD XC via the iFi Micro iCanSE... NOISE WHY GAWD IS THERE MORE NOISE

Not the Etir's fault, I have a really Nice Vinyl rip of Steely Dan and I always notice the noise at the end during the quieter parts, but never during the song... well now I can hear the noise from the Vinyl rip throughout the whole damn song.... I checked a CD rip of the same album [an it wasn't there]

Also new, listened to The Shortest Straw now... there's a touch of Echo on the BIG drum hit in the intro... never noticed that before! So... blacker/Airer is how my rig sounds after putting the Etir in the mix YAY let's switch over to the ZMF Eikon x Project Ember II [which is by the way running my new BJC Rca's]

So with the Eikon the vinyl noise is there... as always it's not AS obvious, so good on the LCD XC brought that detail a little more forward, but again happy I own the Eikon, it's there it's just a little subtle'r as it ought to be, what I DO appreciate is on Steely Dan's Dirty Work, his vocals are SUPER thick, pre etir BJCrca his voice didn't echo as much, nor did I get as much of the texture and individuality of the strings, I can say the same for the bass, more texture better resolve, everything's like a smidge air'er there's a smidge more texture and resolve after dropping the Etir and the BJC cables in, I imagine though it's more the Etir, I get a lot of the same improvements out of the iCAN SE as I do on my Ember II, and the iCan SE is using basic Monoprice cables still [I'm not ready to get another set of BJC Rca's in 8 feet]

Echo is there on Metallica's Shortest Straw as well, there's also a more texture between the channels, the Guitar's always sounded... exactly in sync on the two channels, but their not, the strumming is different on each of the two channels, so oooooooh! Whoa yea the imaging opens up too,

I have to say @schiit nicely done! Never realized how shitty my NFB10ES2 was via USB, a real shame, either way so far the Etir's doing an excellent improving my existing dac!! It's a shame how much little stuff I've been missing over the years, not a huge amount maybe like an extra 3-4% of texture or plankton that wasn't there before, not including the improvements to the overall imaging!

 

This feels like the objectivist wet dream come true: DDC, done right, shouldn't matter. As it's all about passing bits from point A to point B. If Eitr manages this, and obsoletes myriad isolators, de-whatefiers and 'audiophile' USB cables, that would be absolutely awesome.

 

To sum everything up (all other opinions are wrong):

Is (decrapifier magical USB thing) better than Eitr? No.

What is on par with the Eitr: Aurlic Aries, Lynx cards.

Should I worry myself to death like I have for years with all sorts of alternate USB cables and decrapifiers in front of or behind the Eitr? No. This made my $200 USB cable sound the same as my $10 Amazon basics cable. Yes, I could hear a difference before, now I can't now.

Does it sometimes act funny and not play? Yes, make sure you are in 24-bit or 16-bit mode and not 32-bit mode and you'll be happy.

So you're saying I should just put the Eitr between my DAC and my computer, stop worrying about USB and just enjoy music? YES. That's the whole point of this device.

 

Eitr will be a sure purchase for me, just waiting for one in black

What I want now is a A/B/C comparo:

PC > Eitr > Modi Multibit vs. Gungnir Multibit vs. Yiggy

 

That's easy ... in fact I'll go one further and include Bifrost MB. The more ">" between each DAC, the bigger the difference:

Eitr* -> Yggdrasil >>> Gungnir MB >>>>>> Bifrost MB >> Modi MB​

Shocking, huh?**

Depending on your current source, Eitr can make a bigger difference with Bifrost MB and Modi MB than it does with Gungnir MB or Yggdrasil, which is presumably down to the two more basic units not having Adapticlock and the Eitr, consequently, "fixing" more, respectively, for those units. Differences may, or may not, be easier to hear on the higher-end units however ... depending on the rest of your chain.

I have a minimum fee for stating, and re-stating, the bleedin' obvious. So that'll be $49.95; payable to the SBAF fund for the terminally repetitive.

--

*And nobody who's tried it and/or is paying attention here cares as to what's before the Eitr as it doesn't make any difference - so I left out the "PC".
**I actually have a proper four-way Schiit multi-bit DAC comparison pending, though that'll go in the "Life after Yggdrasil" thread.

 

The same because Yggdrasil and Gungnir Multibit coming out now are both Gen 5.

Regarding older models, I know this is obvious but for the sake of simplicity:

5>3>2