Schiit Yggdrasil -- 21 bit?


Schiit says that Yggdrasil is a 21 bit DAC. But the DAC chips that they put in the device ( Analog Devices AD5791BRUZ, 2 per channel) are 20 bit with the error of plus-minus 0.5 LSB.

How can the DAC be 21 bit if the chips are 20 bit? Using two chips per channel does reduce the RMS voltage of the noise by  a square root of 2. But how can you get to 21 bit from there?

Can someone please explain.
defiantboomerang
But fuses DO work, if used correctly.

Is there a way of using a correctly rated AC mains fuse the wrong way??
Or are you saying that they are directional??

Cheers George
What I’m saying is that, if turned in one direction, the dimensionality (and general sonics) sound "off," or, as others put it, "flattened," as though you folded an accordion in to its smallest size. If you turn the fuse the other way (whichever is the "right" way), it will be as though you have the accordion fully extended (with your arms fully out to either side of you). I don’t know the mechanics, but I can hear the effect easily, and so can anyone who has their system set up optimally. (Not as easy to do as it sounds, given that a local dealer here has a $100,000+ setup and it sounds very mediocre). No dynamic range, individual instruments don’t jump out at you as they do in an orchestra in real life, and very little inflections [ dotted 1/8 notes, etc.] apparent in the presentation. And I’ve owned many of the components he has, (Wilson speakers, ARC, Transparent and Nordost cabling) in earlier incarnations, and so I know what the earlier incarnations sounded like, and his should sound FAR better than what I had 20 years ago. They don’t.)
Anyway, it’s not that the fuses can cause damage: they just don’t sound "great" in the same way that if you put a component on a kitchen table, it doesn’t sound as good as if you have an audio rack, built specifically to inhibit vibration. It’ll sound "okay," but "okay" is not "optimal."

Incidentally, my Yggy has hit its 8th day. Last night, I walked OUT of the room, dissatisfied with Marianne Faithfull’s voice on Kurt Weill’s "Seven Deadly Sins." It sounded "fractionalized," as though some of the notes in her singing (particularly when she was singing "throaty" notes) had disappeared. Now, I HAD just moved around power cords on the Yggy (put a Shunyata Python ZiTron on it and moved the Yggy’s stock power cord to my Arcam FMJ) - and disliked the results. So I unplugged them, put the ZiTron back on the Arcam and an ESP power cord (20 years old, but hey, it’s better than the Yggy’s stock cord) back on the Yggy (I’d removed the stock cord around 5 days, just to see if it made a difference. It did.) BACK onto the Yggy. It sounded okay, but frankly, if you move a power cord - or speaker cable, for that matter - and expect it to sound great 5 minutes later, you’re in for a disappointment. Again, it’ll "sound" okay, but if you check it two hours later (yes, I know there are doubting Thomases here: I don’t care. I did setup when I was the Equipment Manager for Fi Magazine, and some of that equipment was pretty exalted stuff and anyone who was a careful listener could hear it), it will sound much superior.
Anyway, didn’t hear it again until this afternoon - an hour ago. And was impressed. REALLY impressed. And the general consensus is that it needs longer than the two week burn-in to reach optimal sonics, although Thomas (whoever he is), at Schiit, wrote me an email saying they didn’t subscribe to "burn-in." Amusing that his name is Thomas (the doubting one), because whether its burn-in or something else, that bass drum on the JVC XRCD version of Holsts’ "The Planets" on cut 6, "Uranus," was downright gut-punching (and I have Nola Thunderbolt subs, so if there’s no improvement, it just ain’t there). I don’t know what they’re doing that others aren’t, but the sound is akin to a Mercury Living Presence sound: close up and personal. And the instruments are (finally) taking on a very "physical" presentation (which means the midbass is filling in.) Anytime you have a really, REALLY 3-D presentation, it’s the midbass. (Ask any musician who plays in an orchestra.) That’s why earlier version of the Wilson WATT/Puppies (I had versions 1,2,3,4) sounded so "holographic": Wilson’s sound - back then - was weighted towards the midbass, which made instruments sound extremely "there" as though there was an invisible Thelma Houston singing in front of you. The Yggy has - so far - a VERY good midbass, although I THINK the drum in Uranus is more upper midbass than lower (anyone who knows more about this, [plenty of people] please correct me if I’m wrong here). In any case, I can hardly wait for day 10. Some people are saying a month (notably, Robert Harley, TAS’ Executive director), whose review of the Yggdrasil is there for anyone to read on the TAS website.
This is clearly a reference-quality DAC. I haven’t discovered the shortcomings, because I’m agog at the music that comes out of the Nola speakers, and I don’t agree with Thomas at Schitt that there is no "burn-in." (Neither does anyone at Head-Fi.org, for that matter.)

if turned in one direction, the dimensionality (and general sonics) sound "off," or, as others put it, "flattened," I don’t know the mechanics

There are no mechanics just AC mains theory, which needs brushing up on.
Mains in US changes it’s flow direction 60 x a second, 50 here in Au.

To be able to hear if a mains fuse "is" directional, you would have to pull it out and change it's direction  60 x's in one second and! to stay in sync with the mains doing it!!!   

Cheers George
and I don’t agree with Thomas at Schitt that there is no "burn-in." (Neither does anyone at Head-Fi.org, for that matter.)
That is not true as written. *Most* on head-fi (Yggy impression thread) agree burn-in is required (to optimize sound quality). Some do not.