[malocadi]

I have a Schiit Bifrost 2 on the way and I'm trying to read up on what makes its new Unison USB interface so popular. Apart from Schiit's own description on what it actually is I'm having a hard time seeing what makes people so excited for it.

So per their description it's an in-house implementation tailored specifically for delivering PCM and provides electromagnetic isolation. The creators claim it's better than spdif. I guess my question really is, how can a different USB interface provide better quality sound? In my mind it's an interface for delivering bits, 1s and 0s, which may or may not have error correction to a source that then translates it to sound. Regardless of how the interface is implemented the bits coming in should always be the same right? Especially given that it's the same interface used for delivering data in computers where a single bit of difference corrupts the whole file.

 

[bigshot]

I haven't found that USB has any particular need for error correction, particularly when you're passing as small a file as audio. I don't know anything about this particular model, but it sounds to me like a solution to a problem that doesn't really exist. Not uncommon in audiophilia.

 

[Davesrose]

I have a Schiit Bifrost 2 on the way and I'm trying to read up on what makes its new Unison USB interface so popular. Apart from Schiit's own description on what it actually is I'm having a hard time seeing what makes people so excited for it.

So per their description it's an in-house implementation tailored specifically for delivering PCM and provides electromagnetic isolation. The creators claim it's better than spdif. I guess my question really is, how can a different USB interface provide better quality sound? In my mind it's an interface for delivering bits, 1s and 0s, which may or may not have error correction to a source that then translates it to sound. Regardless of how the interface is implemented the bits coming in should always be the same right? Especially given that it's the same interface used for delivering data in computers where a single bit of difference corrupts the whole file.

Seems like marketing hype. I've noticed most all DACs made in the last decade or so....the USB connection is USB 2. That's more than enough for audio (it's 480 Mbps). Originally, Toslink (the first digital stereo PCM interface) had a bandwidth of 3.1 Mbps (now it can get up to 125 Mbps: good enough for stereo PCM and lossy surround sound). It would be one thing if this is a claim for lossless, high resolution multichannel connections....but for stereo, the numbers say that any current digital audio connection is more than enough for high quality audio. This also seems ridiculous to me as I edit 4K video files: items that are large and taxing with current bandwidth (where I have been using USB 3.1 or even using external SSD hard drives): current interfaces can make quite a difference in time of processing for things that are more intensive than playback of stereo. The highest bandwidth for the latest digital standards has not been needed for stereo for quite some time.

 

[gregorio]

[1] So per their description it's an in-house implementation tailored specifically for delivering PCM and provides electromagnetic isolation. The creators claim it's better than spdif.
[2] I guess my question really is, how can a different USB interface provide better quality sound?

1. It's a marketing claim. spdif can deliver audibly perfect audio but if you try to run the signal over too long a distance or have a DAC that is poorly designed, then there could be audible issues.

2. The answer is similar to point #1. Given a faulty or very poorly designed DAC, with a USB interface that does not isolate it's output from EM interference, a DAC that is not faulty and does isolate it's output could provide better quality sound. I presume this wonderful bit of marketing is in response to the criticism of some of their other products. Some Schiit models were found to be very deficient in this regard, allowing noise/em interference from the USB connection to leak into it's analogue output at levels much higher than other DACs, even much cheaper ones. However, although significantly worse than other DACs in this regard, we're still only talking about noise leakage down at around -80dB and therefore it would be inaudible, unless you raised the volume during quiet sections of music. I presume their new "in-house tailored implementation" brings it more in line with other far cheaper DACs!!

G

 

[malocadi]

1. It's a marketing claim. spdif can deliver audibly perfect audio but if you try to run the signal over too long a distance or have a DAC that is poorly designed, then there could be audible issues.

2. The answer is similar to point #1. Given a faulty or very poorly designed DAC, with a USB interface that does not isolate it's output from EM interference, a DAC that is not faulty and does isolate it's output could provide better quality sound. I presume this wonderful bit of marketing is in response to the criticism of some of their other products. Some Schiit models were found to be very deficient in this regard, allowing noise/em interference from the USB connection to leak into it's analogue output at levels much higher than other DACs, even much cheaper ones. However, although significantly worse than other DACs in this regard, we're still only talking about noise leakage down at around -80dB and therefore it would be inaudible, unless you raised the volume during quiet sections of music. I presume their new "in-house tailored implementation" brings it more in line with other far cheaper DACs!!

G

This is what I suspected too, that the problem would be in the actual electronics of the interface versus the interface itself. The bits are bits. A better ATM isn't going to make the same amount of money worth more.

 

[manueljenkin]

I have a Schiit Bifrost 2 on the way and I'm trying to read up on what makes its new Unison USB interface so popular. Apart from Schiit's own description on what it actually is I'm having a hard time seeing what makes people so excited for it.

So per their description it's an in-house implementation tailored specifically for delivering PCM and provides electromagnetic isolation. The creators claim it's better than spdif. I guess my question really is, how can a different USB interface provide better quality sound? In my mind it's an interface for delivering bits, 1s and 0s, which may or may not have error correction to a source that then translates it to sound. Regardless of how the interface is implemented the bits coming in should always be the same right? Especially given that it's the same interface used for delivering data in computers where a single bit of difference corrupts the whole file.

Try to read up a little bit more on mixed signal ic design, pcb design, ground plane design, optocouplers, galvanic isolation, i2s signal generation etc!. You would eventually get an answer. Or, just buy one and find it out for yourself.

 

[bigshot]

And do a controlled listening test to make sure it isn’t just a solution without any problem.

 

[dazzerfong]

Try to read up a little bit more on mixed signal ic design, pcb design, ground plane design, optocouplers, galvanic isolation, i2s signal generation etc!. You would eventually get an answer. Or, just buy one and find it out for yourself.

That implies that have an EMI immunity problem to begin with, which isn't a good look if your DAC costs $1000+. Or, worse, it's an emissions problem so you're disturbing other equipment.

Anyways, I reckon this is an attempt to decouple themselves from XMOS or C-Media and owning one more part of the supply chain.

 

[manueljenkin]

That implies that have an EMI immunity problem to begin with, which isn't a good look if your DAC costs $1000+. Or, worse, it's an emissions problem so you're disturbing other equipment.

Anyways, I reckon this is an attempt to decouple themselves from XMOS or C-Media and owning one more part of the supply chain.

Not EMI, but ground plane noise. But I understand your sentiments. Dunno about xmos and c-media how good/bad they are (my xmos based chepo DACs respond to cable changes, so I assume it's not perfect). At 2400$ they should have come up with a good DDC to begin with instead of asking you to "upgrade" the interfaces.

He asked me what's the difference, not what's the audible difference. I just told him what the design changes try to control. Whether it's audible or not depends on your ears, mood, the component in place.

 

[dazzerfong]

Not EMI, but ground plane noise. But I understand your sentiments. Dunno about xmos and c-media how good/bad they are (my xmos based chepo DACs respond to cable changes, so I assume it's not perfect). At 2400$ they should have come up with a good DDC to begin with instead of asking you to "upgrade" the interfaces.

Which falls directly under EMI. Also, DDC is a completely different protocol (hint: the first D is 'display').

What's funny is that their old USB interface had this to say:

Forget decrapifiers, regenerators, isolators, and all the USB dongles and boxes you’ve been told you need for USB sound. Our all-new Gen 5 USB input doesn’t require any of that stuff. It’s this simple: Gen 5 is USB, solved.

The new Gen 5 board is an upgrade for any Bifrost, Gungnir, or Yggdrasil with the Gen 1, 2, or 3 USB inputs, and now ships standard with every new Bifrost, Gungnir, or Yggdrasil.

 

[malocadi]

Not EMI, but ground plane noise. But I understand your sentiments. Dunno about xmos and c-media how good/bad they are (my xmos based chepo DACs respond to cable changes, so I assume it's not perfect). At 2400$ they should have come up with a good DDC to begin with instead of asking you to "upgrade" the interfaces.

He asked me what's the difference, not what's the audible difference. I just told him what the design changes try to control. Whether it's audible or not depends on your ears, mood, the component in place.

I get that there's a difference in technology behind the interface but I was trying to see if there's any validity to the claim that a better USB interface actually provides better sound quality. I don't believe it because the UBS layer should be completely isolated from the interpolation or D->A portion of the DAC. That is of course unless their previous USB interfaces were producing interference that was adversely affecting the output.

I can't find the post now but I remember a specific quote that suggested that the new Unison USB interface produced much better soundstage, which in my mind sounds irrelevant for something that's simply receiving 1s and 0s from a source.

edit: as an aside I'm aware of some of the possible side effects of replaying digital, such as clock syncing to avoid jitter. If the improved interface somehow handles this better or eliminates it altogether then I could see how it might improve the sound quality.

 

[bigshot]

Improved soundstage is almost always a word to describe expectation bias. It disappears in controlled listening tests. I don't see any reason why a properly functioning USB port would sound any different than any other. For something to sound different, something has to be wrong somewhere. You can't sound better than transparent.

 

[Davesrose]

Given the bandwidth of USB 2, I'm not sure how jitter or interference comes into play. A digital cable is not as sensitive to EMI and given that the bandwidth for 2 channel audio is relatively low for data transfer standards, you've got to be talking about either really poorly designed interface/cable or insanely long usb cabling with extenders. I personally feel some devices may sound different due to factors, but have never experienced any change in audio difference with different digital cables or device's digital interface (only if there is a problem with cabling, it manifests as obvious video/audio drop outs). If the only claim that the Unison USB interface produces "much better sound stage", well that seems like subjective validation.

 

[bigshot]

Jitter, in the amounts it occurs in even the cheapest digital home audio component, is well below the threshold of audibility, usually by an order of magnitude.

 

[dazzerfong]

Remember, a DAC has an analog section. That's what it's doing after all: it doesn't stay digital forever. That analog stage is prone to EMI, especially if you have a poorly-grounded or poorly-isolated design.

That being said, you're not exactly on the money with USB and EMI. USB is prone to EMI: that's why there's shielding and TPs all over the place. However, these manifest as errors, not less soundstage.

Would be nice to have someone from Schiit explain their line of reasoning, but I don't think this forum section is quite ready for that.