Schiit Yggdrasil Less is More (and MIL and OG) Impressions + Measurements

See below. As always, subject to change as I listen more, going back and forth with a variety of recordings. You guys know what a tough exercise - think ballpark. I've also changed a few things around after further consideration. A few extra DACs thrown in for entertainment purposes and relative comparisons. And yes, I do hear the DAC2541 has being slightly more engaging (however it does have less plankton to work with). As usual, I've put some thought into scale.

Engagement / microdynamics / immediacy
Plankton / low-level information / resolution (not macro-detail)
Blackground / clarity
BAL OUTPUTS / AES INPUT and variety of headphones and amps


Richness (entire range, not necessarily warmth, think chocolate cakeness)
Treble F'kery (digititus, sandiness, unnatural edge, prickliness)
BAL OUTPUTS / AES INPUT and variety of headphones and amps


Thought: the prior generation AKM DACs were less resolving, but more engaging and much less F'd sounding. Hence my crapping the new AKM stuff here. I mean, if you are poor, get something like Modius or JNOG**. TBH, I use Modius as cheap instrumentation for headphone measurements. It's technically superb in some respects, but I have zero desire to listen to if I can avoid it. Of course it doesn't matter now with AKM. Cosmic forces beyond our control saw to this.

Obviously these are specific technical aspects that do not really paint the entire picture nor attest to how they may synergize with your own specific components.

**better yet, ditch delta-sigma garbage and get Modi MB and swap AD86xx opamp to OPA2156 (toob) or OPA1656 (clean) -or- wait for entry level stuff from Soekris. I wish we had other choices like RDAC. @CEE TEE? Sigh... if only those Airist Audio guys were the real deal and didn't turn out to be posers. Think what could have been. Then again, their profiles of Haavaad and Princeton should have been red flags.

 

Yes and no. We can still note microdetail through a sheen or on a less than perfectly black canvas. A sheen isn't necessarily opaque nor a less than perfectly dark canvas preventing the imprinting of fine detail. It's like noise. The nature of noise is random - this is why we can hear through noise unless the noise is just really bad. The noise floor we see as a nice line on AmirNADS graphs isn't static. It only looks static because it's averaged two times: first through an FFT processing window of so many tens of thousands of samples, and then another average of those FFTs.

Also, I really dislike the word detail. If you want "detail", get a super crispy sounding Chi-Fi ESS DAC or the latest Chi-Fi IEM claiming to use the latest "advanced" e-stat tech or <insert random periodic table element> driver. Now that's "detail". That being said, the Yggdrasil A2/OG is both more macrodetailed as well as microdetailed to LIM. A2/OG is has a slight application of an over sharpening filter. LIM is like me doing this to every 4k TV I meet: immediately turn down picture sharpening from default (usually 20) to 0. LIM is 4k. A2/OG is 6K. None of this matters if you listen to Walküre Attack! However it most certainly will if you listen to classical recordings, even ancient stuff from the 60s if they were good recordings, e.g., LSO stuff during Decca/Stanley Black.

Many years back, there was a Yggdrasil (A1) super fanboy on HF who argued that the grayness of the Schiit DACs was because of their ability to dig up the finest plankton and have an amazing ability to reproduce granularity to the finest levels. I think he had a point - but only so much because the Yggdrasil A1 and Gungnir A1 are in fact greyish sounding. (I'm not that much of a fanboy). On this thought however, I feel that DACs with overly black backgrounds are this way because there is a simplification going on. This blackness is fake because I don't hear it in real-life. In fact RL tends sound muddy and dirty as F sometimes. It's post tricks from sound engineers that clean up this shit.*

* Which is why the Steve Albini master of In Utero F'ing rules. Boy, that shit got all fucked up between the band and the big studio. Glad 20 years later people realized that Albini actually knew WTF he was doing and the studio guys (and Nirvana for going along with it) were dipshits. BTW, there were bootlegs of those Albini masters floating around 10 years ago. I had them - FLAC/WAV, not mp3. Nirvana is supposed to be heard in a crowded bar with cigarette smoke everywhere, piss and beer on the floor, and the house PA overdriven to massive distortion and broken tweeters. (And I'm not ashamed to say that I liked Pixies and STP more, even though people said STP were posers).

 

Indeed. It's not just like this in the lows, but also in the mids, and highs. Very cohesive this way and probably underappreciated by me for this quality. There's something to be said to a coherent consistent presentation from lows to highs. As I get older, I'm placing even higher priorities on this. It's also often a primary reason for the disconnect between my IEM reviews to others. Any designer of audio products who thinks a BA, DD, and piezo driver can play together should be shot. Their bodies should be defiled, drawn, and quartered - if they developed it, listened to it, and actually thought it was good. These kinds of IEMs are like bad "Asian fusion" where some white chick makes a dish that mixes pine nuts and ginger. (I've so glad the Asian fusion craze has died out).

Still same ratings as above because I cannot account for specific synergies. The ratings above are relative not absolute. I may want to pick up another chassis so I have can two Yggdrasils, one A2, the other LIM.

Big JBLs with Fisher KX200 (PP 7591 fixed bias). Amp has inherent warmth, not outright warmth - great grip on drivers - belying the typical loosey goosey PP Chi-Fi garbage. Big 18" woofers have huge magnet and fairly stiff surrounds for power handling. Super low Qts. As such, despite being 18" bigass woofers, they are dry sounding, especially since cabinet is like 10cuft. JBL butt-cheek tweeter is laid-back and rolls off a bit on the highs, like any compression driver used more than 24 hours. With 10" mids, sound is smooth. Attack is fast, but decay is slow. Yggdrasil A2/OG is the correct choice here - better edge definition, organic bloom to dry lows, and some extra sparkle.

EC x4 45 Studio. Used Gungnir A2 because Yggdrasil A2/OG does not work. Amp sounds like the OPTs (same as Studio B), so more Studio B sound than 45. Maybe close to same warmth and body (result of interstage transformers) as B but more extended highs. Also loved DAC2541 abd Burl B2 with this amp. SET is gonna sound like SET with rounder sound. Plus primary headphones are not HFM TOTL specials, rather JAR600, ZMF Verite, Grados - none of which have clean lows or perfect highs (small spikes on Verite and Grado), Yggdrasil A2/OG is a disaster waiting to happen, unless I really wanted a THICC sound and overemphasized top octave. Many folks love that extra air to get that sense in RL - I absolutely hate it because it almost always sounds wrong. (Exception of naturally mic'd orchestral or chamber recordings with simple small condenser mic arrangement).

Anyway, LIM will replace Gungnir A2. Brainless choice because LIM SE outs can go direct into SET SE inputs without the need for a XLR->RCA line transformer. No matter how good or carefully selected a line transformer, if this can be avoided, and the intention is to not color the sound, by all means avoid. (Gungnir / Yggdrasil A2 SE outputs are gimped compared to BAL for those who didn't know.)

BTW: All impressions via pi2AES AES output. Unison is overall warmer sound with less incisive transients.

 

Let me send mine to you. PM me info including email and phone (for tracking notification).

Tape and vinyl really different beats from digital - can't really compare.

Yup. the grayness could be those AD57*1 chips. It could also be the output stages. Look at THD. It's a bit higher on A1/A2 than LIM. (The way I look at THD is that it matters... but only up to a certain point where it doesn't matter anymore). There were some tweaks to the output stage that improved the greyness deal on the A2 from A1.

I never really felt Gungnir A1 to be sweet sounding. It is overall warmer and more pleasant - easier to integrate compared to Yggdrasil A1/GS. I did feel the Gungnir A2 upper mids were sweet - bit forward. On the other hand, others felt the A2 sounded beached, probably because of this.

On Soekris DAC2541 vs LIM, hard to say for you. LIM isn't a neutral or dry or accurate as Soekris. LIM sits between Soekris and A2/OG. If GS works for you, then keep it. @ChaChaRealSmooth is using GS with his DNA Stellaris. DNA amps have a wetter bloomier sound. Good fit here. Now @ChaChaRealSmooth just needs to order a LIM for his EC Ultralinear - unless he's perfectly content with the GS for that too.

It's like what Jason says: different flavors. Even the Rockna Wavedream which I feel is the overall best DAC yet, it's got a flavor - one that doesn't happen to work in any of my current setups. This is why 88% of rooms at audio shows and 99% of tables at CanJams suck. Most arrangements to mix and match gear are political or ones of convenience. It's the actual owners who F with this shit who get the best results - they are the best chefs. Unfortunately, the 200% for profit CanJams killed the hobbyist angle. Jude doesn't give a crap about you the little guy as long as he's laughing all the way to the bank.

True story about careful selection of components into a greater whole: https://www.superbestaudiofriends.o...jam-2016-oc-meet-impressions.1638/#post-40897

Except from HF:

And then maybe a year back to the LAX Hilton was this kid, not even in high school (maccat) had this awesome setup with the HD650, a turntable, a Jolida phono-pre, and a Bottlehead Crack. I was like wow, this kid put on a better setup than 90% of the others with 25% of the budget.​

BTW this kid showed up a few years later to a San Diego Meet. FOTM Schiit DAC, FOTM Schiit amp, and FOTM Audeze LCD-3. Sounded like ass (it wasn't that bad, but relative to what he had before, disappointing because everybody by then had a similar setup). This is what happens when you randomly throw TOTL / FOTM stuff together because HF sold you on it.

 

Loaner poll up.

 

OK bad example using GPUs. Have you read the news about Toyota cutting production 40%, or the Walmart CEO saying stuff about supply chain and inventory shortages going well into 2020, or container ships having to anchor off Shenzhen? Heck, a company I am consulting for can't even get the firewalls and network switches we need. Maybe I know something or maybe I don't. The current situation could always change. I'm just trying to helpful here to folks at SBAF. Not trying to be a know it all. It's not just GPUs dude. If you work in any kind of capacity which involves procurement of equipment or material, you would know how the supply chain is totally fucked now, with notices from manufacturers of how it's going to continue to be fucked, with no end in sight. Heck, I can't even buy certain sized inductors for a speaker crossover I'm working on, or a Burl B2 Bomber DAC, or a Sony Alpha 1, unless used. All out of stock.

 

I've warned folks about FOTM and have already made clear this is a sidegrade. The only thing different is that I lowered the fire button from 90% of 70% certainty because of the uncertain global supply chain situation.

Tell that to Toyota, GM, Ford. Do you think they would still NOT release a product because they knew they couldn't meet demand? That would be absolutely asinine. Oh, let's not sell Corvettes this year because we know we can't meet demand. It's better not to make money, not sell a product, because no potential customer will bitch at us for not having an item in stock. I'd murder with my bare hands any CEO who made this decision if I sat on the board.