I think that some consumers have a problem with Schiit's approach because you guys simply refuse to fit into a very specific box when it comes to extolling the virtues of your (new) products compared to most other manufacturers. In other words, you do not provide a comfortable reassurance to this type of consumer that they will be guaranteed a 25% improvement to their level of enjoyment from the latest iteration of their favourite product. Some people need a promise from a manufacturer that the new product is absolutely and utterly superior to the now suddenly old/obsolete/redundant piece of Schiit they have. They do not want to discover the potential improvements for themselves - they simply want to be told it's better.
Oh well, you can't please everyone.
I am definitely susceptible to that, though it's not as simple as being told that it's better (marketing) as it is being indicated that it as good as before or better through specifications, listening impressions, measurements, price and, yes, also marketing.
Basically, I really like the Yggdrasil Analog 2, it is still my favorite DAC, and I even prefer it over the Holo Audio Spring 2 KTE, which is more expensive. So if Schiit made a true Yggdrasil 2 (or Analog 3 or whatever you want to call it) that even Mike feels trumps his original (and upgraded) creation, the best DAC he knows how to make, that has better specs, a similar or higher price... I'd buy that in a heartbeat. Because I wouldn't expect to like it less - though I could get disappointed.
Better how? Well, audio is funny in that you don't know what you're missing until you heard it, so: surprise me.
With the two new options that were announced, it's tricky. Objectively, they seem like steps down. For one, they are cheaper. Now that one doesn't mean much to me in the case of Schiit, since they follow a simple cost + standard margin = price approach, which is why their prices don't conveniently fall on the end of their respective price brackets. I like that, and it's very possible that a cheaper part from a different manufacturer performs better in some way, allowing for a better product that is cheaper. Or maybe it requires fewer parts to get to behave. So let's ignore the price here.
But then there's the bits part. 16 bit chips? Objectively less capable than the OG with 20 bit chips. Well, for > 16-bit audio. I always try to look for components that can handle 24/192, for one it has become a sort of standard for high end audio and I do have some 24/192 material.
But Yggdrasil OG already doesn't really "naturally" resolve all 24 bits anyway, so that would be a clear area for improvement - from an objective point. How many of those recordings actually contain 24 bits of signal, I don't know. Whether I could really hear that with my air purifier whizzing in the background, I don't know.
And more importantly I mainly listen to Spotify Premium. So that's 16 bits and lossy compression. Spotify Hifi with just lossless audio, not even high resolution audio, will be a very welcome upgrade for me. So the "Less is more" option is actually interesting. Wouldn't be the first time an improvement is achieved by trying less and doing it better.
Within the same chip family I'd lean towards the one with higher accuracy. 20 bit Yggdrasil OG vs. 18 bit Gungnir, for instance.
But the 16 and 20 bit TI chips that Schiit will use are from different families. The 16 bit equivalent to the 20 bit DAC11001A would be the DAC81001 (there's an 18 bit one as well, the DAC91001).
Instead, Schiit uses the DAC8812. Now that's a stereo DAC while the 20 bit one is single channel - but Schiit uses one per channel per phase again if that Reddit post is accurate. I'm wondering whether they are simply not using the second channel of each DAC chip or whether they somehow benefit from it anyway.
It also has a relative accuracy of 1 LSB Max and differential nonlinearity of 1 LSB Max (LSB = least significant bit) while the DAC81001/DAC91001/DAC11001A family all are specified with "Integral linearity: 4-LSB INL (max)". They also have a "1-LSB DNL (max)" rating, so I'm assuming that relative accuracy = integral linearity, but I'm just a clueless customer looking for objective indications of performance here without being able to put them in context. But 4 LSB looks quite a bit worse, so I suspect the DAC8812 is the better 16 bit chip vs. the DAC81001 and possibly even the DAC91001 (16 - 1 > 18 - 4).
Anyway, the DAC8812 also operates in current mode vs. voltage mode.
That's interesting and unusual. To me that makes it different enough to be be open to the idea that it's actually better than the Yggdrasil OG (though it might be worse), at least for 16 bit content. So I think I'll give that one a shot.
Edit: Apparently the AD5547 in the Modi Multibit is also a current mode DAC, so I suppose this point is moot.
For me to routinely listen to 24 bit material, Amazon would have to essentially clone Spotify Connect, because the Blu OS app integration just doesn't cut it in terms of usability. Or Qobuz would have to do that while heavily expanding their catalog (it's been a while since I tried it). Or Tidal would have to allow a true lossless tier with unadulterated 24/192 FLAC files and none of that MQA business.
I don't see any of that happening, though.
I suspect the "More is less" version is more for people who can't buy anything without Amir's approval, so I'm less curious about that one as someone who has the OG, but impressions might sway me. It does annoy me that a seemingly lower spec part (4-LSB INL vs. 1-LSB INL for the AD5791BRUZ in the OG) results in a better measuring product. But measurements are complicated and not all measures taken to improve then inherently sound better.
Yes, if the the Yggdrasil OG measured as good as it sounded, part of me would sleep better. However, that part of me doesn't make my purchasing decisions.
So I might get the Less is More one. But I'm also curious about the Holo May, the Denafrips Terminator Plus, and Chord Dave + M-Scaler. So I might not. We'll see.
If it weren't for you I'd never known about them!
