Schiit (MultiBit) Bifrost

I'm a very skeptical person in general. But I believe there is definitely differences in some way between DACs. I'm just not sure at what price point that difference emerges. It's good to see some hard evidence of the Bifrost Multibit being noticeably better, nonetheless the more tests that occur the more solid the evidence gets.

 

The sample size is definitely below something I would personally consider statistically significant. I merely upped the threshold to make them "more" significant, mostly for my own sense of data gathering, without it becoming a choir, audio should never become a choir. Also with this platform of testing, especially when done according to the books (aiming for 0.3dB to 0.1dB), the differences are pretty clear. To actually make it almost inaudible and very hard to hear, you need to go below 0.1dB, which is already a thing of note. Besides that and anyone who does these tests will be able to nod, it's hard because of so many more factors involved while doing these tests.

Plus you need to reference yourself with the material you're testing to be able to actually find what you're looking for. That is something to invest time in, really finding out what these nuances mean in the sighted listening experience and then see if you can find the reference back in blind testing. Plus the chain is equally important, so far only highly resolving transducers were capable enough for me in the blind tests. For example my Shure 1540s, as much as I love them, didn't shine at all in this aspect, it simply lacked the detail retrieval to make this feasible. Then the Amp, O2 is the perfect example of not being able to make enough of a difference in the chain for this level of level matched testing.

Anyhow, I might do one more test if I can find the time in which I will try to shoot for a 50. 10 tracks, 5 loops each, Bifrost Multibit vs. ODAC on my main rig: HD800 + HDVA600. But no promises, for the time being I'm just gonna enjoy the musicality my "analytical" rig has gained. Plus I'm playing around with HD800 mods, including the usage of cork haha.

 

SKiring wrote:

" Prominent songs that we used for the blind and critical tests:
Michael Jackson - Baby Be Mine
Andreas Vollenweider - The Glass Hall
Fabrizio de Andre ft. Mina - La Canzone di Marinella
Dire Straits - Sultans of Swing
Dire Straits - Your Latest Trick
Mahsa Vahdat & Mighty Sam McClain - A Deeper Sense of Longing
Mahsa Vahdat & Mighty Sam McClain - Silent Song
Raul Midon - State of Mind
Marcus Miller - Blast
Tracy Chapman - Fast Car "

I already commented on the fact that I use the Vollenweider track in testing. However, I also wanted to comment that - although I am not familiar with all of them, I suspect that none of the tracks in this list are recorded in real stereo. So, none of the aspects of sound quality that are painstaking preserved by Multibit or the "burrito filter" are present at all in your tests.

You need this sort of recording:

 

Can you help me understand two of the things in the above: 1) What does it mean that a recording isn't made in "real" stereo? Hasn't pretty much everything been recorded in stereo for a very long time now? I suspect there is something I don't know about here. 2) I get that the burrito sauce optimizes signal in the time domain and this is a big part of why instrument positioning is so precise and focused on these DACs, but wouldn't this still work on a mono signal? Cheers and thanks for your insights.

 

Real Stereo....Some (most) recordings are a bunch of single tracks that are manipulated onto a blank canvas. Individual performers are placed on the canvas left-right, up-down, front-back according to the sound tech's desire.

On older (50-60's) recordings, most, if not all of the performers are playing together live although the entire track may or may not be recorded at one time.

I've found myself listening to more and more older recordings just because I prefer the way it sounds (cohesive) to my ears.

 

That looks like a pair of Schoeps modulars probably with MK4 or MK41 capsules in ORTF configuration on what is probably an AEA mic positioner bar, with two out riggers that look like Earthworks QTC-1.

 

I'll attempt some insight into 1. Delivery format is stereo. That delivery comes from a mix created in the studio or on stage. The instruments are often multi-mic'd and brought to the recording console. The mix consists of pan potting the various mics into a position that may or may not match how the musicians were arranged, depending on how well the bleed between mics was controlled. Some mics are just for accent and are mixed in at a level below the main signal just for some presence or other effect. Some mix engineers will go farther and introduce channel delays to the individual mic feeds to further enhance the illusion of a stereo presentation. Then compression, EQ, reverb, etc. are optionally employed to further create the desired main mix sound.

A true stereo recording refers to minimalist stereo recording techniques such as those described in The New Stereo Soundbook, Ron Streicher & F. Alton Everest. This book also details multi-mic setups. The book is a beginning. Apprenticing with an old school recording engineer that really understands mic polar patterns and how to work mic bleed rather than fight it, also helps in the student achieving a more realistic recording. Today many recordings are assembled with musicians recorded either in multiple iso-booths or in tandem, using multi-track multiple take techniques. The recording you hear may have been passed around to many places on the planet before final mix and mastering occur.

 

yes! My recently departed colleague, Frank Laico was one of the old school recording engineers at Columbia 30th St Studio. Before he passed, I was privileged to learn how he placed mics, worked with bleed, how they arranged the various groups in the studio. 30th St studio was huge. Frank often said 100x100x100 ft, but we think it was something closer to 100x80x60. Still, a very large and amazing place. Now gone.

 

Ahh, perfect. Thanks @skank and @atomicbob. This actually explains a lot. So that seems to imply that most classical recordings are true stereo, because they don't multitrack those (or most of those), right? That is, they just mic the entire ensemble?

 

Not necessarily. I've sat in on sessions where there are multiple accent mics positioned in each of the orchestral sections, and some soloists will be spot mic'd. These must be time delayed to the main pair (and outriggers if any) to avoid comb filtering and other detracting effects. To find an example of true stereo recording, and oriented to headphone people, look for binaural recordings such as those from Chesky records. David uses a Neumann KU-100.

 

I have some of the Chesky binaural recordings (those are very cool), but... does true stereo _have_ to be a binaural recording?

EDIT: nm, stupid question. i'm so headphone-centric that I didn't really think about it.

 

Yea 0.1dB to 0.3dB I feel is easily detectable with any amount of critical listening. If you're really going to torture yourself with more AB, then you might as well mix things up every 10 tries. Not sure how your setup is, but switch any attenuators in the chain. Redo volume matching again from scratch, etc. I was really happy to finally identify the DACs until I redid the volume matching and that messed up my next round of 10.

 

No worries, I've done that too. Have the sloped forehead to prove it.
The book I mentioned details a number of stereo recording techniques. You can see the table of contents here: http://www.stereosoundbook.com/pages/contents.html

 

Multitrack tape machines came into use in the very end of the 1960s. So, in the 1970s, a lot of classical recordings were multiple mike performances. Deutsche Grammophon is a good example. They would place one mike in the woodwinds, one mike in the brass, etc (or sometimes more than one in each section). Then, like a rock recording, they would mix the orchestra.

But, classical recordings from the late 50s and 60s were often true stereo - good examples being the RCA Living Stereo and Decca's Mercury Living Presence series.

Anyway, the original point was that if one is questioning whether there are actual differences between DACs, then one needs to do tests with the most "revealing" of recordings. This is where the "Audiophile Record Labels" come into play, e.g. Opus3, 2L, Sheffield Labs, Refernce Recordings and others. There are some Opus3 tracks that I use for testing that I would never listen to for enjoyment.

Conversely, if one is going to do an entirely subjective comparison with no A-B or blind aspects, then just use your favorite music that is very familiar to you.

 

I don't want to turn this into the statistical epistemology thread, but that level is what's appropriate for (many) scientific papers, because the goal there is to prove that an effect exists beyond a reasonable doubt. When the question is "do DACs make a different on the sound," if you're relatively agnostic about it, then it's reasonable to use a "preponderance of the evidence" standard, where even an above 50% likelihood of a real result should lean you in that direction.

Mostly, though, I feel kinda bad for Schiit. Because they make reasonably-priced equipment and don't make grandiose claims for it, skeptical-minded people end up buying their stuff and insist on actually proving that it's good, whereas if they made $4,000 DACs and described them as "the sound of luxurious listening" or whatever, they could pretty much guarantee that hard-headed buyers would skip right over them and only the most committed, hardcore-believer audiophile would even consider their stuff.

 

Yeah if I give it another go it's gonna be a full test like the last one, from start to finish making sure my friend measures it up for me. And I'll be aiming for 50 sets. Still gotta see all that, I mean I have myself convinced at least. It's definitely there in my opinion and I value my sighted listening tests more haha.

 

I have a great idea for some – they could blindfold themselves, add earplugs and experiment on making love to various consenting women (or women audiophiles to consenting men – like it would be a problem to find them). They could switch midstroke, to see if they could tell them apart. Those who are indifferent to, never been properly laid, or dislike the opposite sex could start getting laid science forums to prove that it is impossible to tell any difference between them. They, of course, would end up with the worst of the bunch. Those of us who love, appreciate, enjoy, and treat our lovers well would end up with the best. To say nothing of enjoying life.



Audio is to be enjoyed, with or without my gear. Please!

 

I want to pin this up somewhere.

 

Welp, no longer feeling bad for Schiit.

 

Yes, we have the secret weapon: Mike.

Seriously, if you aren't having fun with your gear, your gear is having fun with you.