Schiit (MultiBit) Bifrost

.

 

I hate when Randos don't read/care "WHAT IS SBAF" before making their posts. It's clear as day when signing up.

You suck, bie.

 

Normal is overrated.

 

I received my upgraded Bifrost Multibit back. I took off the back panel, as promised previously in another thread. The exterior case is the same (same minor cosmetic scratches, same rubber feet). Everything else appears to be brand new:

• The original Bifrost was a delta-sigma "Uber" with an "A" serial number. Now it has a brand new "B" serial number, and the "A" sticker has been peeled off and voided.
• It now has a main board that is marked "Bifrost MB v2.05." It appears to also be marked "2017", which would postdate the production of the Bifrost Uber model (The 4490 replaced the Uber in October 2015.)
  

• It now has a Multibit board that is marked "Schiit BifrostMulti BFM v2.0."
  
  

The "v2.0" marking appears to confirm that there has been at least one major physical update to the Bifrost Multibit card at some point since the introduction of the Multibit model.

I did not fully remove the boards from the case because @rtaylor76 made me a bit worried about breaking off the LEDs, and because it seems to me that the marking text on the cards is enough to confirm the existence of a silent "A2" upgrade to the Bifrost. If people are really clamoring for a full-nude picture of the board, I will reconsider. But for now I'm going to plug it in to start warming up.

 

Nice! Looks like the mainboard got a .05 bump since this:

https://www.superbestaudiofriends.o...ale-of-eight-bifrosts.2643/page-4#post-131540

I'd open it up & take close ups! 1.xx to 2.xx (mainboard) was a cirrus spdif chip to an AK4xxx spdif chip (mostly). Look closely here:

https://www.superbestaudiofriends.o...ale-of-eight-bifrosts.2643/page-4#post-131552

On the new BFM v2.0 is that a white bodge wire?

 

After three days of warmup of my upgraded Bifrost Multibit A2, here are initial impressions vs. a Modi Multibit that has been plugged in for several months continuously. I have no idea if my Modi MB has upgraded firmware or not; I bought it at the right time, but bought used. Both are connected to my PC source via the same Eitr, using an RCA splitter. Outputs from both are run through a SYS to facilitate switching. Amp and headphones used were HeadAmp Gilmore Lite mk2, LCD-2.2 prefazor, HD600.

• Bifrost MB A2 and Modi MB sound at least 99.5% similar. Every single difference noted below is a tiny gradation, just the slightest shade of difference. If wasn't actively listening for differences, I wouldn't have noticed any of these observations. If you're looking for a big upgrade, you really need to look elsewhere.
• Bifrost has a bit more macrodynamic impact & swing, which comes with a slight increase in perceived "air" around notes. Modi sounds slightly "flatter."
• Bifrost has slightly increased clarity of staging.
• Tone: Vocals that sound "breathy" on Modi sound slightly less "breathy" on Bifrost. Acoustic guitar riffs that have prominent atonal "strumming" sound to them on Modi have less "strumming" sound and more tonal content on Bifrost.
• Bass: Bifrost has a bit more heft in the midbass region, especially noticeable in male vocals and in fuller sounding bass guitar lines. This is probably the most noticeable difference between them. Bifrost also has slightly deeper bass extension.
• Bifrost maintains coherence a little better than Modi in complex passages. (My go-to track for testing this is Radiohead's "No Surprises" from the bridge onward, which builds to a big confusing mess.)

I have a theory that the slight increase in midbass response from Bifrost is why some have heard the Bifrost as being "slower" than Modi. Increased midbass affects the whole mix, making atonal elements like breathiness or strumming less prominent overall, and those high-frequency atonal elements are associated with speed. However, having compared Modi and Bifrost directly I don't think it's literally an issue of speed, just tone.

USB-in: I tried feeding the Bifrost MB directly via its USB Gen 5 input. You would think the results would be the same as Eitr/coax, but strangely, Eitr/coax had slightly more bass weight. Not a big difference, but worth noting.

Optical-in: I tried feeding the Bifrost directly from my PC motherboard's ALC1150 codec via a glass Toslink cable. The result differed from Eitr in some ways that are consistent with what I've seen with other DACs: thicker bass texture & warmer tonality, with less volume from the "presence" region upwards but smoother up top. If your motherboard has optical-out, I think it's worth a try. The difference between using optical-in and using Eitr/USB is actually bigger and more noticeable than the differences between Bifrost MB and Modi MB themselves. You might find that you prefer the tonality of optical.

Pricing: A new Bifrost MB is priced at about $170 above the cost of buying both a Modi MB and an Eitr. Minus another $20 for a decent coax cable. I think $150 is a reasonable premium for the marginal sound improvements and one-box convenience of the Bifrost. You can tell something is priced right when it's a tough decision versus possible alternatives.

In conclusion: Bifrost Multibit A2 sounds marginally better than Modi Multibit. If marginally better isn't enough for you, then you need to look elsewhere.

 

LOL!, no shit! ;D

Post #9134 -- just one example, check out that crusty beast!

 

Really nice writeup... and you may have one of the most interesting Schiit MB's once USB genX is released... it has the purest i2s into the burrito filter (no reclocking)... and genX has some special clocks!

PS. Awesome that they replaced the mainboard...

 

Agreed. The optical jack on the old mainboard was broken when I sent my Bifrost in for the Multibit and USB upgrades. I asked Schiit to fix the optical jack while they had it. So I’m not sure whether replacing the mainboard was necessary for compatibility with the Multibit V2 Card, or just more time-efficient than desoldering, replacing, and testing a new optical jack. Either way, pretty cool to get it back with what appear to be entirely new internals.

 

I have the same codec on my motherboard. It's definitely lack of definition from the jittery optical output of the PC, although I do think the warmness is somewhat pleasurable for crappy sharp/bright recordings if you have to be subjective about it. The Eitr sits somewhere between normal USB and optical, where it's definitely not as etched as Gen 2 USB but also not as warm/lacking in sharpness as mobo optical. Personally hoping the new USB goes a bit more towards optical tonality while maintaining the clarity.

 

I’ve heard jitter described in many ways, but never that it alters tonal balance. I’ve heard of jitter adding grain, harshness, or haze to the highs, but I didn’t notice any of that with Toslink. Just shelved-down volume in the highs and a “smoother” presentation of the highs; all perceivable detail and clarity with Eitr/USB was still there with Toslink. The main difference was tonal balance.

It’s possible that different motherboards produce different quality SPDIF/Toslink outputs even when using the same Realtek audio codec. Or that the quality of the optical cable matters a lot. (Mine is glass fiber.) But this is probably speculation for a different thread. I can only tell you what I hear; it’s up to you how you interpret it.

To bring it back around to the Bifrost: If you think that my notes regarding the Bifrost’s optical-in are attributable to me preferring a lack of definition from jitter, then you probably shouldn’t take the rest of my impressions of the Bifrost very seriously.

 

It's not an attack against your impressions, they're pretty much on the dot subjectively. Jitter is not gonna have a consistent "sound" as long as different DACs use clock signals differently: a delta-sigma DAC would have different non-linearities versus R2R as jitter changes of course, as certain aspects of the jitter quality matters a lot for noise-shaping. I'm also presenting my subjective impression that I think using optical affects transients quite a bit, making them much duller than Eitr to me, but it does get rid of any possible grain via complete electrical isolation (if Eitr was perfect, Schiit wouldn't be releasing a new USB implementation). It doesn't invalidate your good impressions. If liking warmth and smoothness invalidated people's opinions, more than half the site would go down the drain.

 

Any feedback on the latest revision of the Bifrost? I'm assuming you have the latest version running your JBLs...

 

Thanks nice review, but do consider when Schiit comes out with their new USB implementation Eitr will be second best while the Bifrost MB can be upgraded by swapping out the USB board. If you can afford $599 Bifrost MB is the better long term buy right now precisely because it can be upgraded. You would have to buy a new Modi MB to get the new USB. So short term savings results in a long term loss.

Mike Moffat has said that Schitt plans to release their new USB when Microsoft gets to the no new development point on Windows 7 which is January 14, 2020. Personally I hope it is sooner than that but you can appreciate Schiit does not want to have to deal with Windows 7 driver issues.

I have owned both a Bifrost MB and now have a Yggdrasil A2 and the difference in SQ is not as great as you would think using headphones. Powered monitors are in my future which is why I bought the Yggdrasil and to really appreciate what the Yggdrasil can do I think you need a good two channel system. Yggdrasil is the best no doubt and I can appreciate the refined sound on my HD800S but the Bifrost MB is a real bargain at $599.

 

For future reference and such, how would you compare it to Gungnir A2?

I've never had another DAC in its place on my bedroom 2ch, so I'm not sure how it stacks up, though I get the feeling it's not nearly as dark as I remember the old version being.

 

I'm reversing my opinion and eating crow on this one, @Ash1412. You were completely right. Yes, the difference with optical was jitter. And yes, I was enjoying the jitter. Uncomfortable self-realization there.

I just got an iFi SPDIF Purifier (essentially a reclocker), and routing my mobo optical through the Purifier makes it sound nearly identical to Eitr / USB Gen 5 as a source. The differences that remain after reclocking the mobo optical, if any, are tiny and not worth mentioning. The iFi SPDIF Purifier also has the side-benefit of outputting both electrical & optical simultaneously, so I can connect to both types of inputs on the Bifrost to compare input quality. Result: the Bifrost's optical and electrical SPDIF inputs sound exactly the same.

I don't know what I expected, but it wasn't this. Anyway, thanks for getting me to question my prior beliefs.