This is a review and measurements of Schiit BiFrost Multibit DAC. It retails for $599 plus shipping and is on loan to me from a kind member of the forum. For this review I will be comparing it to the Topping DX7 which retails for $200 less ($399).
Note that this is the special "multi-bit" version of the BiFrost DAC which uses a R2R ladder DAC from analog devices (AD5547). This is an extra cost option.
The Schiit BiFrost Multi-bit comes in a plain looking silver box. It is more substantial than the smaller DACs they offer like the Modi 2 but it still does not give one feeling of high-end or high-quality.
Connectivity is very sparse with USB, Coax and Toslink inputs and only unbalanced RCA output. No AES/EBU digital input or balanced outputs. In contrast the Topping DX7 has all of these other inputs/outputs in addition to headphone output.
A cheezy toggle switch in the back turns the unit on and off. There is only one button in the front which selects the input with cryptic symbols instead of names of the inputs. It is as if they think no one knows what USB means outside of English speaking countries!
Unfortunately no teardown will be coming as the owner wants to preserve the 5 year warranty which comes with this unit. From the bit that I see, build quality leaves something to be desired. A single tiny screw holds the RCA jacks to the back and it was at least two turns loose. When I tried to plug in my RCA cable into it as it was, the jacks started to move backward. Fortunately I stopped and tightened the screw. Still I kept getting the feeling I was going to break those jacks with my rather stiff interconnects.
USB class compliance is good with Roon/Windows recognizing the device without any software installs on Windows 10:
We see that sample rates up to 192 Khz are supported.
Device identity was "Schiit USB Gen 2" which was nice as opposed to some generic Xmos indicator or such.
Let's get into measurements and see how it performs.
Measurement
For comparisons I picked the Topping DX7 as mentioned. Let's get started with our usual J-Test for jitter and noise. First the USB input:
As we see, tons of jitter and distortion products are visible together with higher noise floor. Even the Topping D30 at $129 shipped runs circles around it in how clean its output is relative to Schiit BiFrost Multi-bit.
There are serious design problems here to cause this many distortion products. Samples are just being stepped on. Fortunately levels are low so not a huge audible concern but massive engineering black eye.
Let's test the two units using S/PDIF digital input and see what that shows:
We see a subset of problems but still way, way too many distortion products in the output of Schiit BiFrost Multibit. The Topping DX-7 performance is again squeaky clean compared to it.
For the rest of these tests I continued to use S/PDIF which should give the benefit of doubt to the Schiit DAC.
First our new favorite, the linearity test:
What the heck is the Schiit DAC doing??? Look at that Cyan line. It shows the analog output of the DAC vs digital PCM value. An ideal DAC would have flat line. On Schiit BiFrost Multibit it all of a sudden refuses to obey the input values and follows its own direction! The massive error causes the red line which shows the differential to go wilde and off the chart.
Using my criteria of 0.1 dB error from ideal, the Schiit DAC only manages 58 dB or less than 10 bits of resolution! Did you read that? 10 bits! So even when playing ordinary 16-bit CD, you are sacrificing the low order 6 bits.
In sharp contrast the Topping DX7 shows almost an ideal flat line! From run to run it changes a bit which means it is dominated by tiny amount of noise. Computing its equiv number of bits, we get 19 bits of resolution. That means this DAC does justice to high-resolution audio.
Next let's look at intermodulation test using SMPTE test tones:
That is a stunning difference in performance. The Topping DX7 is some 25 db cleaner than Schiit BiFrost Multibit. That is a massive gulf between them.
Looking at the spectrum of noise and distortion a similar picture shows up:
I changed the vertical scale to percentages to give a more familiar feel to the numbers. As we see, the Topping DX7 is some two orders of magnitude cleaner than the Schiit BiFrost Multibit.
Looking at this measurement with a spectrum analyzer we see this:
Look at the horrible output of the Schiit BiFrost Multibit DAC. Its output has a mix of noise and distortion well above that of DX-7.
Finally a boring frequency response measurement:
0.6 dB of droop is visible by 20 Khz. Not a great sin but would be nice to get a flat response.
Edit: as requested later, here are the measurements of the two DACs when fed a very low level sine wave (-90 db or so):
As we see the Schiit BiFrost Multi-bit heavily glitches and at any rate, doesn't produce anything close to a sine wave. This agrees with earlier measurements that it loses linearity way before we get to such low level signals.
Conclusion
It is no secret that the previous Schiit products have not done well on my workbench. But I was hoping as suggested by many that their high-end products would do better. Alas, what is shown above is just unacceptable. It shows a product that simply has no design hygiene. Either wrong DACs are used or driven incorrectly in the implementation. There are DACs literally at 10% of the cost that do much better.
Yes I know the subjectivists go by what these devices "sound like." But I want one of them with a straight face tell me how their music is better where a spray of distortion and noise is added to any single tone in their music. How can any of this be euphonic???
I have no choice but to give the Schiit BiFrost Multibit a failing grade. NOT Recommended.
As always, I welcome correction, feedback, questions, etc.
If you like this review, please consider donating funds for these types of hardware purchase using Patreon (https://www.patreon.com/audiosciencereview), or upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
Edit: this review has created an explosion of commentary and back and forth across no less than four forums! Member Candlesticks was kind enough to summarize the key posts in this rather long thread:
Edit 2: For explanation of what these measurements mean and the setup, please see: https://www.audiosciencereview.com/forum/index.php?threads/understanding-audio-measurements.2351/
Note that this is the special "multi-bit" version of the BiFrost DAC which uses a R2R ladder DAC from analog devices (AD5547). This is an extra cost option.
The Schiit BiFrost Multi-bit comes in a plain looking silver box. It is more substantial than the smaller DACs they offer like the Modi 2 but it still does not give one feeling of high-end or high-quality.
Connectivity is very sparse with USB, Coax and Toslink inputs and only unbalanced RCA output. No AES/EBU digital input or balanced outputs. In contrast the Topping DX7 has all of these other inputs/outputs in addition to headphone output.
A cheezy toggle switch in the back turns the unit on and off. There is only one button in the front which selects the input with cryptic symbols instead of names of the inputs. It is as if they think no one knows what USB means outside of English speaking countries!
Unfortunately no teardown will be coming as the owner wants to preserve the 5 year warranty which comes with this unit. From the bit that I see, build quality leaves something to be desired. A single tiny screw holds the RCA jacks to the back and it was at least two turns loose. When I tried to plug in my RCA cable into it as it was, the jacks started to move backward. Fortunately I stopped and tightened the screw. Still I kept getting the feeling I was going to break those jacks with my rather stiff interconnects.
USB class compliance is good with Roon/Windows recognizing the device without any software installs on Windows 10:
We see that sample rates up to 192 Khz are supported.
Device identity was "Schiit USB Gen 2" which was nice as opposed to some generic Xmos indicator or such.
Let's get into measurements and see how it performs.
Measurement
For comparisons I picked the Topping DX7 as mentioned. Let's get started with our usual J-Test for jitter and noise. First the USB input:
As we see, tons of jitter and distortion products are visible together with higher noise floor. Even the Topping D30 at $129 shipped runs circles around it in how clean its output is relative to Schiit BiFrost Multi-bit.
There are serious design problems here to cause this many distortion products. Samples are just being stepped on. Fortunately levels are low so not a huge audible concern but massive engineering black eye.
Let's test the two units using S/PDIF digital input and see what that shows:
We see a subset of problems but still way, way too many distortion products in the output of Schiit BiFrost Multibit. The Topping DX-7 performance is again squeaky clean compared to it.
For the rest of these tests I continued to use S/PDIF which should give the benefit of doubt to the Schiit DAC.
First our new favorite, the linearity test:
What the heck is the Schiit DAC doing??? Look at that Cyan line. It shows the analog output of the DAC vs digital PCM value. An ideal DAC would have flat line. On Schiit BiFrost Multibit it all of a sudden refuses to obey the input values and follows its own direction! The massive error causes the red line which shows the differential to go wilde and off the chart.
Using my criteria of 0.1 dB error from ideal, the Schiit DAC only manages 58 dB or less than 10 bits of resolution! Did you read that? 10 bits! So even when playing ordinary 16-bit CD, you are sacrificing the low order 6 bits.
In sharp contrast the Topping DX7 shows almost an ideal flat line! From run to run it changes a bit which means it is dominated by tiny amount of noise. Computing its equiv number of bits, we get 19 bits of resolution. That means this DAC does justice to high-resolution audio.
Next let's look at intermodulation test using SMPTE test tones:
That is a stunning difference in performance. The Topping DX7 is some 25 db cleaner than Schiit BiFrost Multibit. That is a massive gulf between them.
Looking at the spectrum of noise and distortion a similar picture shows up:
I changed the vertical scale to percentages to give a more familiar feel to the numbers. As we see, the Topping DX7 is some two orders of magnitude cleaner than the Schiit BiFrost Multibit.
Looking at this measurement with a spectrum analyzer we see this:
Look at the horrible output of the Schiit BiFrost Multibit DAC. Its output has a mix of noise and distortion well above that of DX-7.
Finally a boring frequency response measurement:
0.6 dB of droop is visible by 20 Khz. Not a great sin but would be nice to get a flat response.
Edit: as requested later, here are the measurements of the two DACs when fed a very low level sine wave (-90 db or so):
As we see the Schiit BiFrost Multi-bit heavily glitches and at any rate, doesn't produce anything close to a sine wave. This agrees with earlier measurements that it loses linearity way before we get to such low level signals.
Conclusion
It is no secret that the previous Schiit products have not done well on my workbench. But I was hoping as suggested by many that their high-end products would do better. Alas, what is shown above is just unacceptable. It shows a product that simply has no design hygiene. Either wrong DACs are used or driven incorrectly in the implementation. There are DACs literally at 10% of the cost that do much better.
Yes I know the subjectivists go by what these devices "sound like." But I want one of them with a straight face tell me how their music is better where a spray of distortion and noise is added to any single tone in their music. How can any of this be euphonic???
I have no choice but to give the Schiit BiFrost Multibit a failing grade. NOT Recommended.
As always, I welcome correction, feedback, questions, etc.
If you like this review, please consider donating funds for these types of hardware purchase using Patreon (https://www.patreon.com/audiosciencereview), or upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
Edit: this review has created an explosion of commentary and back and forth across no less than four forums! Member Candlesticks was kind enough to summarize the key posts in this rather long thread:
Edit 2: For explanation of what these measurements mean and the setup, please see: https://www.audiosciencereview.com/forum/index.php?threads/understanding-audio-measurements.2351/
@amirm as this thread is now 20 pages I've compiled a set of links so that anyone new entering this thread. If you want could you edit them into your original post please.
Below is an abridged series of events for people new to the thread.
Measurements posted by SBAF user Atomicbob:
Measurements of Schiit Bifrost Multibit after warm up overnight by amirm.
Response by amirm after seeing Atomicbob's measurements as linked above. The second part. The third part.
First post by amirm in the Mike Moffat (Schiit engineer) blog on Head-Fi. Further down are more posts by amirm in that thread. Response from Mike Moffat. Reply by amirm.
Post written by Atomicbob on measurement procedure. Response by amirm to that post.
Post by Atomicbob on non-linearity of Schiit Bifrost Multibit. Response by amirm.
Last edited: