Mike Moffat of Theta fame and Jason Stoddard of Sumo fame teamed up to create affordable components right here in the good old US of A.

First of a Breed

Schitt Audio recently released the Bifrost, which is the first and least expensive of three planned Digital-to-Analogue Converters (DACs) in the company lineup. The second DAC will have balanced in/output abilities and rumors for the third DAC include words like “Reference” and/or “Statement” DAC.

The main selling points for the Schiit Audio Bifrost DAC includes a modular/upgradable construction of the main DAC and USB input board, no up-sampling such that the Bifrost processes digital input at it’s native rate, an asynchronous USB2 input up to 24/192 input, a completely discrete analogue output section with no IC/opamps; all at an affordable price of $349 for the basic DAC with a USB2 input board available for an additional $100. Schitt also offers a 15-day evaluation period; you can return the DAC no questions asked although you pay for shipping.

Who is Schitt Audio?

Schitt Audio, pronounced just the way you think it should be pronounced, is a fairly new company. Industry heavyweights, Mike Moffat of Theta fame and Jason Stoddard of Sumo fame teamed up to create affordable components right here in the good old US of A. The company has a fairly refreshing corporate persona, which can be experienced at the company website: www.schitt.com.

What is a Schitt Bifrost?

The Schiit Bifrost DAC arrived neatly packed. The construction/appearance appears nicely done with a simple, a well-designed look that could pass for a lifestyle/ipodish type device. It is relatively small at 9 x 6.75 x 2.25” and weighs ~5lbs.

Key features include asynchronous USB 2.0 input as well as Toslink and RCA S/PDIF digital inputs (all up to 24/192 except for the USB2 input which will not accept 176.4 rate) with a single set of standard RCA outputs. The Bifrost will not formally up-sample before the conversion process, however it uses a Delta-Sigma AKM 4399 DAC chip which by its very nature must over-sample as part of its conversion process. The DAC has an upgradable DAC board module and an upgradable USB input module. Setup is as simple as can be, stick on the included rubber feet, plug it in with a cable of your choice (DIY 14AWG solid core woven in my case), connect a digital input (Toslink, USB or RCS S/PDIV), select the input and it makes music. No drivers are required for Macs, and Windows drivers are downloadable for Windows Vista, XP, and 7.

KEY FEATURES

Made in the USA | Fully upgradable DAC | 24/192 BitPerfect | 32 Bit with Discrete Analog Output | Asynchronous USB 2.0 | D/A CONVERSION IC: AKM4399 | ANALOG SUMMING, FILTERING: Fully Discrete, JFET differential topology

SPECIFICATIONS

OUTPUT IMPEDENCE: 10 Ω | FREQUENCY RESPONSE, ANALOG SUMMING STAGE: 2Hz-100KHz, -1dB (actual frequency response limited by sample rate) | MAXIMUM OUTPUT: 2.0VRMS | THD: Less than 0.008%, 20Hz-20KHz | S/N: Greater than 105dB | POWER SUPPLY: 5 stages of regulation, including separate supplies for critical digital and analog sections. | UPGRADABILITY: Separate, modular USB Input Card and DAC/Analog Card are snap-in replaceable. | POWER CONSUMPTION: 12W | SIZE: 9 x 6.75 x 2.25” | WEIGHT: 5 LBS

CONNECTIONS

INPUT: Coaxial SPDIF, Optical SPDIF, USB (optional) | INPUT CAPABILITY: up to 24/192 for all inputs | INPUT RECEIVER SPDIF: Crystal Semiconductor CS8416 | INPUT RECEIVER USB: C-Media CM6631
OUTPUT: Dual RCA (single-ended)

WARRANTY & PRICE

15-Day Satisfaction Guarantee | 5-Year Warranty | PRICE: $349, add $100 for USB

Review setup

A comparison between the Schitt Bifrost USB DAC and my older, slightly modded Cambridge Audio DacMagic proved interesting. Rather than just report on how the Bifrost sounds, I think a comparison between the two DACs gives a better overall picture of how these things sound. The DacMagic is certainly a known quantity and has been widely reviewed, demoed, and/or owned by a large sample of the audiophile community.

While vinyl still reigns supreme in my home, I do frequently listen to music from a digital source. My current digital source is a Windows 7, Quad core I7 2.2 laptop with 12GB ram; running FooBar (WASAPI, no dither, 350ms output buffer, 24 bit output) while the rest of the system uses SET 1626 tube pre/300B monoblocks and horn speakers.

My slightly modded DacMagic’s (bypass one of the output stage opamps and removed output caps) sound is big (probably bigger than it should be) with a presentation that stretches across the speakers and fills the soundstage from the floor to the ceiling; however depth behind the speakers is limited. The DacMagic does project into the room, perhaps due to the slight “metallicky reverb” like effect I describe below. The DacMagic has a clearly elevated bass; it is substantial, fat and a bit loose. The somewhat bloated bass response of the DacMagic acts pleasurably like a loudness/Munson-Fletcher curve switch when listening at lower than average levels. The Munson-Fletcher curve is a phenomenon, first published by Harvey Fletcher and Wilden A. Munson, where the human ear is shown to be most sensitive to frequencies between 2k-5kHz. Thus at lower sound levels, frequencies below 2khz will sound too quiet and can be equalized up so they present at a similar level to higher frequencies. This is useful when listening at relatively low levels but not needed at levels typically used by audiophiles.

The DacMagic macrodynamics are good; detail retrieval is excellent although somewhat exaggerated while overall transparency is fairly decent. Tonally, the DacMagic is slightly on the warm side of neutrality. Probably the biggest downside of the DacMagic is a slightly phasey/electronic signature that is ever present. I always attributed this sound in part to the up-sampling nature of the DacMagic; it sounds a little tiny bit like the output of the DacMagic is running through a reverb box. This effect makes everything bigger, instrumental notes have a long, long lingering decay, vocals have a soaring aspect and sibilance is exaggerated with sound stage “air” readily apparent. Almost everything sounds like it was recorded in an under damped, lively, empty church (ok, I exaggerate a bit here but you get the idea).

Sound Quality

"My initial impression was that this DAC is more laid back/recessed, more sharply focused…"

Fresh out of the box and connected with USB, the Bifrost sound was very different from my DacMagic. My initial impression was that this DAC is more laid back/recessed, more sharply focused, smaller sounding, with a smaller bottom end. A couple things really jumped out. The phasey “reverb” like quality of the DacMagic was completely absent. Vocals and instruments appeared smaller but more finely focused/delineated. While soundstage depth and width were reduced, depth was much better in the Bifrost with vocals appearing to come from far behind the speakers. Listening to the Drive By Truckers lead singer Patterson Hood on “A Blessing and a Curse” resulted in a slightly smaller, but very focused and tonally accurate Hood standing several feet behind my speakers while the DacMagic presents Hood as a giant mouth several feet across singing in line with or slightly behind my speakers. While vocals and instruments have great definition through the Bifrost, there was not a lot of cohesion. Each instrument/vocal while cleanly portrayed, suffered from a sense of dislocation from each other i.e. the Bifrost highlighted individual performers at the cost of losing the overall musical synergy in a performance. Listening to Duke Ellington and Johnny Hodges on “Side by Side”, I was initially struck by how distant and isolated the performers sounded and nothing could be further from the truth on this recording or in this performance. I have never been that concerned with PRAT, perhaps because my SET/horn system just naturally boogies but it was very hard to feel the tempo and drive of the performance at this point.

Tonally, the Schiit Audio Bifrost DAC appeared essentially neutral although vocals lost some of the “chestiness” presented by the DacMagic. Vocals initially sounded comparatively threadbare and very slightly “nasal” through the Bifrost when compared to the DacMagic. While I would not describe the Bifrost as cold or clinical, it certainly was less romantic and warm then the DacMagic. Although the overall picture at this point was somewhat smaller and recessed, within that presentation the high frequencies were far more prominent in the Bifrost with excellent focus, detail and control. Somewhat surprising in light of the excellent hi-freq reproduction, was that the sense of “air” or the reproduction of the recording space surrounding the performers/instrument was diminished in the Bifrost compared to the almost artificially enhanced sense of space and air produced by the DacMagic. While the DacMagic had more of this quality, more is not always better. I felt that the DacMagic was providing a little too much of a good thing. Combined, these vocal/hi-freq/”air” differences between DACs could easily be heard while listening to Robert Plant and Alison Krause’s album “Rising Sand” or Don Chamber’s recent album “Zebulon”.

The bottom end/bass as reproduced by the Bifrost was leaner, tighter and more detailed, however this lean bass tilted the overall sound slightly towards its rather exceptional reproduction of high frequencies. I prefer the full-bodied sound provided by the DacMagic but clearly the Bifrost was presenting textures/subtleties in the lower frequencies that simply did not exist through the DacMagic.

In terms of background blackness, the Bifrost has greater “emptiness” when things get quiet and/or between notes, which I suppose is desirable. On the other hand, the DacMagic did such a terrific job of portraying the ambiance of the recording event in space/time that even between notes you often got a sense of where people were standing/sitting etc. There were no hum/noise issues noted with either DAC. I was unable to determine if either DAC was turned on/off with my ear about 6” away from a 100db efficient horn speaker. The outboard AC transformer of the DacMagic runs very hot, in fact the plastic case surrounding the AC step-down transformer is now brittle due to the heat and one of the screws just fell out while the Bifrost case runs fairly cool.

So, overall the Schiit Audio Bifrost DAC sounded very focused, a bit lean, leaning towards the analytical, very controlled but with a loss of PRAT and musical cohesion. A disappointment, both my wife (whom I rely on as an “impartial” observer) and I preferred my old DacMagic at this point. This was somewhat of a surprise as the Schitt DAC has many aspects that I consider desirable including made in USA by a small company with eccentric personalities and principles and a host of neat features such as non up-sampling, async USB 2.0, discrete JFET output stage etc.

This initial comparison was with a USB connection to the BF and a Toslink connection to the DacMagic. I tried reversing the connections and found that the overall results remained the same. The Toslink input on both DACs appeared to provide a little more bottom end weight/bloom when compared to the USB input as well as a slightly warmer sounding presentation. I preferred the Toslink connection with both DACs although the difference was not as great as I expected. All comparisons were done with both USB and Toslink for each DAC. I finally settled on the toslink connection as the one I would use going forward with the Bifrost.

Sound Quality a little later on……. aka Holy mother of all burn-ins!

I believe in and have experienced the sonic nature of a new component changing after it has some hours of use under its belt. Typically, the difference is fairly small and I have always attributed at least some of that perceived difference after burn-in as neurophysiological in nature (i.e. in fact it is my auditory cortex that is burning-in). I had my old DacMagic running as a reference to provide a stable baseline for comparison. Throughout this comparison, the modded DacMagic sounded roughly the same while the Bifrost appeared to change its sonic presentation as the burn-in process continued.

At 48 hours, the subtle nasal quality with vocals had disappeared and what remained was the Bifrost’s characteristic smooth, slightly recessed, controlled and very neutral vocal balance. Singers started to sing more from their chest and less from their mouths. Vocals still present as smaller, more focused and with much greater depth when compared to the DacMagic. Initially, the Bifrost heavily favored the leading edge of notes, transitioning to smooth control of the body of that same note and minimizing the trailing edge. As the break-in process continued, the portrayal of the beginning/main body/trailing edges started to bloom/fill in somewhat such that the middle/body of the note became more prominent while trailing edges lingered longer resulting in a more attractive and balanced sonic presentation. Both macro/micro dynamics continued to improve somewhat during this period from what I would describe as good to startlingly good on recordings where this prerequisite level of dynamics reside.

"My head was nodding, my foot started tapping and my interest in the music increased significantly."

The sense that the musicians were all playing separate solos also disappeared; the overall musical gestalt gelled such that the performance now overshadows the individual performers except during solo efforts and where appropriate to the recording. PRAT made its appearance; the timing and the momentum generated by the artists performance was much easier to hear and feel. My head was nodding, my foot started tapping and my interest in the music increased significantly. Ellington and Hodges now rocked! Much better. At this point, my long suffering and somewhat reluctant audio partner (wife) changed her earlier preference towards my DacMagic to the Bifrost. I played several different tracks for her blinded as to which DAC was in use and she now preferred the Bifrost. I love it when she agrees with me.

This improvement continued through the ~80 hour period. Bass became a little fuller but continues to remain a little lighter in weight than I would prefer. However, the Bifrost bass that exists is very tight and tuneful with detail that simply did not exist when played through the DacMagic. The Bifrost bass does go deep but is not bloated or pumped up in any way. The best of both worlds would be the quantity of bass of the DacMagic with the tunefulness and quality of bass supplied by the Bifrost.

At this point in time, the Bifrost strikes me as a very neutral and detailed DAC with accurate tonal qualities, much better than average micro and macrodynamics and strong soundstage qualities (especially depth, perhaps slightly less so with height and width). Detail retrieval and focus are excellent. The overall sound is somewhat biased towards the mids and up and this probably remains my main criticism as I like a big fat musical ass. In some ways, this DAC appeals more to my mind than my heart. Poor recordings sound in general pretty bad through the Bifrost; conversely, the Bifrost allows great recordings to sound fabulous. In general, I prefer all my recordings to sound good and the DacMagic does a better job of helping to spruce up dead, overly compressed or just plain poorly recorded music. But in this case, the heights reached by the Bifrost on good or better recordings just outweighs my reservations regarding how poorly recorded material sounds.

As part of the Bifrost review for HionHiFi (originally posted in an online audio forum as an initial impression/review), the editor asked me to expand upon my initial review after several months of ownership. I have always felt that the best way to judge a component is to use it in a system for an extended period of time. This often allows the reviewer to get a handle on the subtleties that differentiate one component from another.

In general, my appreciation of the Bifrost has grown since I wrote down my initial impressions above. The strengths of the Bifrost, including excellent detail retrieval, good tonal neutrality and balance, high levels of transparency, as well as a solid sense of momentum and timing remain. With several hundred hours of use, the sound has become slightly warmer, deviating just slightly from what I consider to be neutral which is a plus in my book when dealing with digital sound. The very slightly recessed nature of the midrange and bass frequencies has largely disappeared leaving a better balanced presentation.

I notice that I can listen at significantly louder sound levels with the Bifrost compared to my old DacMagic. Once a certain volume level was reached, the DacMagic will start to sound congested as instruments and vocals merge and lose some of their definition. This did not appear to be a function of my monoblock amplifiers running out of steam since louder with the Bifrost just gets you more volume and scale. I think this is a result of the extraordinary transparency and clarity of presentation displayed by the Bifrost. I recently listened to Bartok’s Concerto for Orchestra (Living Stereo, Fritz Reiner and the CSO) at a ridiculously high volume and was amazed at how tight and controlled the crescendos sounded. I wish more recordings would use similar lower levels of compression. I felt like the guy in the old Maxell commercial with the guy sitting in a chair listening to Wagner while the sound blows back his hair and a glass of wine. Wonderful!

I have made a few changes to my system during this period. I switched from Foobar to J Rivers Media Center 17 (WASAPI Event and no up-sampling) and although I think I detected a very slight improvement to the sound including a slightly fuller and more natural sound, my main reason was the easy and fast ripping capabilities as well as the easy to use and pretty looking interface.

I continue to use the Toslink optical connection and wondered if it was a good idea to have the modular USB board installed. A couple weeks back, I decided to remove the USB board. Unscrew about ~12 screws and you have the Bifrost circuit board in your hands. The USB board is held in place to the main board by two screws. Remove the screws and it pops out. My impression is that the sound I get from the Bifrost with the USB board removed is very slightly smoother. This is not to suggest that the Bifrost has any grain or grit but recordings just seem to become more “analogue” sounding or easier to listen to. Because it takes a fair amount of time to remove the Bifrost, open it up, and then reverse all this; I was not able to really do any listening comparisons but the differences if any were slight. Perhaps the power supply works better without having to power up the USB board or perhaps the USB board emits some EMI? Or perhaps I am biased because I prefer the concept of a Toslink connected Schiit Audio Bifrost DAC without an unused USB board flapping in the wind?

I have a nice, old PlayStation 1 sitting around in my basement that has been lightly modded (removed muting relays and an opamp) and a quick comparison revealed that the PS1 sound is solidly geared towards the middle with a nice, velvety smooth midrange but is severely lacking at both the bottom and top ends in comparison to the Bifrost. Details reveled with the Bifrost simply did not exist with the PS1. The PS1’s strength is that is focuses on reproducing the core essence of a performance and leaves everything else out. This is attractive in comparison to digital sound in the 80s/90s, which often portrayed a performance in such a way as to make music detached, artificial, clinical and fatiguing sounding. No such compromise is needed with the Bifrost; it reproduces the full spectrum from top to bottom on the recording in a way that makes music exciting, realistic and pleasurable.

What We Liked:

• High frequencies: Extended, detailed, and very revealing
• Middle frequencies: Focused, transparent, clearly delineated and tonal accurate (very slightly on the warm side of neutral) with superb depth and generally excellent imaging.
• Bass frequencies: Tuneful, focused and never bloated or bumped up sounding

What We Didn’t Like:

• Bass frequencies: While the bass goes deep and is very tight, tuneful, it is not as full, room shaking as one might find it a more expensive product

Overall Sonic presentation:

• This is not a DAC that will make your crappy, early 80s digital recordings sound (much) better. This DAC will reveal the source in all its glory (or horror). The Bifrost usually finds some nice qualities from even poor recordings but you will be acutely aware of the source quality.

Conclusion

I am keeping the Schiit Bifrost mainly because once the Bifrost highlighted the slightly phasy/metallicky character that overlies the DacMagic’s portrayal of music, it simply became impossible to ignore; just like the first time you spot a small door ding on your new car, that small ding now jumps out at you EVERY time you walk up to the vehicle. The Bifrost has amazing overall transparency and reveals details that simply are not found with other DACs and CD players I have owned. The bottom line is the more I listen to the Bifrost, the more I appreciate its ability to untangle the messy bits coming out of my laptop. Is it a giant killer? Hard to say really, at this point in time we expect both excellent measured performance and superior sonic portrayals from even modestly priced DACs. A more expensive DAC would probably provide a little more of everything.

The $350 Schiit Audio Bifrost DAC is holding it’s own in a system that for the most part is made up of considerably more expensive components. What other DAC should anyone consider? I do have a certain fondness for the chunky solidness and tonal qualities of the older NOS R2R type DACs, they put an emphasis on a type of musical reproduction that I favor. Some of the Bifrosts competitors from companies like Cambridge Audio, Arcam, and Peachtree have been well reviewed but lack some of the features of the Bifrost (Made in USA, upgradable, discrete output stage etc). For these reasons along with the excellent sound, reasonable price tag, and a 15-day trial period, the Schitt Bifrost is an easy recommendation to make.

Author and Associated Equipment

Author-
Pete Alexandre
Speakers- JBL 4320 Studio monitor horns
Analog Sources

• Well Tempered Super (Reference arm, Platter, motor and clamp on a Super base)
• Nuded Brass Plate Denon DL-103 or Zu Audio DL-103
• Artemis Labs PL-1 phono stage

Digital Sources

• Asus i7 2.2 laptop, 12GB Ram
• J Rivers Media Center 17
• Schitt Bifrost DAC

Amplifier- Modded 2 stage Fi 300B Monoblocks
Power Management- Adcom ILS (brick wall type) |
Dedicated 20A outlet for amplifiers | Dedicated 15A outlet for source components
Audio Racks- StandDesign stands

Manufacturer Contact