Schiit Modius Balanced DAC Review

[dinglehoser]

The reason they say "assembled" is because of California's extremely strict labeling laws. It's basically impossible to source components (resistors, caps, silicon) from US manufacturers because it just isn't made here any more for the most part.

Yes. Adding to the complication, the silicon ecosystem is heavily fragmented, and it's nearly impossible for a end-product designer to specify a supply chain route with any level of granularity. Most silicon vendors are not vertically integrated; for a given bare IC, the design team, the foundry, and the packaging/assembly/test house might all be in different places ... and in many cases, are different companies altogether. Any of those sub-teams might also multisource themselves - e.g., the foundry might use a first party fab in the US and also second source via a third party subcontractor in, say, Taiwan.

And that's just for the chip itself. Then you start thinking about module-level implementation, the passives that go with it, etc. ...

tl;dr It's impossible for an electronic end-product to be "Made in ___" if you interpret that phrase strictly. "Assembled in ___" ... where you do the final assembly of the parts in ___, is practically the best you can do. Which is what Schiit appears to be doing.

 

[Rene]

I emailed Schiit about whether or not the XLR outs are differential or impedance balanced, and I'm happy to report that they are true differential outputs!

Do we care? It's the input circuits that do the trick.

 

[gvl]

@amirm, any chance you can test for incoming jitter immunity over SPDIF?

 

[PeteL]

Yes. Adding to the complication, the silicon ecosystem is heavily fragmented, and it's nearly impossible for a end-product designer to specify a supply chain route with any level of granularity. Most silicon vendors are not vertically integrated; for a given bare IC, the design team, the foundry, and the packaging/assembly/test house might all be in different places ... and in many cases, are different companies altogether. Any of those sub-teams might also multisource themselves - e.g., the foundry might use a first party fab in the US and also second source via a third party subcontractor in, say, Taiwan.

And that's just for the chip itself. Then you start thinking about module-level implementation, the passives that go with it, etc. ...

tl;dr It's impossible for an electronic end-product to be "Made in ___" if you interpret that phrase strictly. "Assembled in ___" ... where you do the final assembly of the parts in ___, is practically the best you can do. Which is what Schiit appears to be doing.

Yes, for the silicon, but in my view it's more than "Assembled in USA" When the metal shop is in Cali, the PCB maker is in Cali, the Pick and place shop is in Cali and the mechanical assembly is at Schiit Quarters. Here in Canada, it would warrant the label "made in Canada", I'm not sure about the USA regulations, but yes, if this not qualify, then nothing in electronics can qualify

 

[Theriverlethe]

It's a drawback nonetheless. I like DSD. And some people like to use a media player which upsamples PCM.

DSD can easily be resampled to PCM, which you have to do anyway if you want any kind of DSP. There is absolutely no sane reason to use a sample rate above 192kHz.

 

[mrdrewk]

I was poking around to see if this thing was a Roon endpoint, but I realized it's a DAC, not a network streamer. So, this could be paired with an Allo
USB Bridge? $240 costs more than this DAC. Is that a wise way to split up $440? Or would that get something more turnkey, possibly better?

 

[theoblue80]

Not having MQA is a drawback... Not having a 1000 other features is a drawback. The reality is that most music is PCM and going past 16/44.1 is quickly hitting diminishing returns.

It's $199 for a fully balanced DAC.

The thing is that we compare the performance and features of similarly-priced DACs, not the features of this DAC against an ideal one.
And IMHO supporting only up to 24bit/192kHz in 2020 is a significant drawback in the competition of the value for money DACs.

 

[Theriverlethe]

The thing is that we compare the performance and features of similarly-priced DACs, not the features of this DAC against an ideal one.
And IMHO supporting only up to 24bit/192kHz in 2020 is a significant drawback in the competition of the value for money DACs.

The problem with that value assessment is that MQA, DSD and sample rates above 192kHz have no intrinsic value whatsoever. They’re pure snake oil.

 

[PeteL]

It will be the same or worse than SPDIF because AES is decoded through the same (AKM?) device.
Advice: Balanced for analog; USB or coax for digital.

I agree that it's likely the same jitter performance, but why do youadvise against going balanced AES/EBU for digital exactly? What's wrong with AES? that's not my experience, not that I would hear any difference, but AES/EBU is a solid, mature and trusted interface for years, why do you advise against?

 

[solderdude]

AES/EBU is just a balanced version of SPDIF. It is likely to be less problematic, certainly with longer cables.
Jitter will be the same as RCA SPDIF, at longer distances AES may be somewhat better when it comes to cable induced jitter because of the balanced character in the cable.
In the end it's the jitter reduction circuits in the DAC that create the result.
It is very clear that USB is better as the USB clock is not used for the audio path. Clock will be sync'd with SPDIF (in all its forms)
Now that Schiit is finally able to make a good USB implementation it is highly recommended to use it and not the technically inferior SPDIF connections.

Use this DAC with USB, you should only use the other inputs when USB is not available (say a CD transport or other devices without USB connection)

 

[PeteL]

AES/EBU is just a balanced version of SPDIF. It is likely to be less problematic, certainly with longer cables.
Jitter will be the same as RCA SPDIF, at longer distances AES may be somewhat better when it comes to cable induced jitter because of the balanced character in the cable.
In the end it's the jitter reduction circuits in the DAC that create the result.
It is very clear that USB is better as the USB clock is not used for the audio path. Clock will be sync'd with SPDIF (in all its forms)
Now that Schiit is finally able to make a good USB implementation it is highly recommended to use it and not the technically inferior SPDIF connections.

Sure, if you have a USB source. I don't in the use case I have in mind for the modius. I know AES is a balanced version of SPDIF, my question was why we should prefer Coax for digital, which is what Rene said.

 

[T.M.Noble]

If they add a volume control and remote for $50 more there will be no more reason to buy Topping DACs.

Baby steps.

 

[solderdude]

When you have AES that is preferred over RCA. Assuming the cable used and termination is correct in both cases.
Use optical when you only have TOSLINK or when there are nasty groundloop issues. TOSLINK may or may not work well with 24/192.

 

[Matias]

@T.M.Noble hi what is the impedance for balanced and unbalanced outputs? Thanks.

 

[T.M.Noble]

They implemented their own USB interface rather than buying an off-the-shelf USB interface. Likely that saved them a few important dollars.

To be honestly it took 2 years of R&D to get our own USB interface. It certainly hasn't saved us money. I attribute it to Mr. Moffat's, lets be kind, eccentricity.

 

[T.M.Noble]

I wonder what the Ethernet-like coil is on the left side of the PCB

Firmware.

 

[T.M.Noble]

They should start offering a discount to ASR members.

We put out a fantastic DAC at a great price made in the US with quality components and you want it for cheaper? There truly is no pleasing some people.

 

[T.M.Noble]

@T.M.Noble hi what is the impedance for balanced and unbalanced outputs? Thanks.

75 ohms.

 

[T.M.Noble]

Schiit is improving because they're adjusting to conform to the poor performance exposure they've gotten from ASR. This kind of adjustment is a huge win for consumers. An excellent example of exactly how a free market is suppose to work. They bought an APx555 for obvious reasons, display their improved measurements on product pages and then send new products to Amir to confirm the measurements. Schiit deserves to be congratulated and appreciation should reflect in sales. We should be looking forward to ASR's influence on other brands and pay attention to those who just make excuses for poor measurements.

We realized we were not serving a meaningful portion of our industry. We are now doing our best to address those customers. In the end, everyone benefits.

 

[T.M.Noble]

Not having MQA is a drawback... Not having a 1000 other features is a drawback. The reality is that most music is PCM and going past 16/44.1 is quickly hitting diminishing returns.

It's $199 for a fully balanced DAC.

Honestly, we thought the little blue light totally threw off the aesthetics.