• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

SMSL SU-2: a dumb, wasted opportunity

No volume control but that's not so unusual, more DACs lack volume control than lack USB.
No volume control is why I would consider this DAC in the first place! This is not a "lack" for my use case, but rather, a sought-after feature.

It is now hard to find DACs that do not include volume control knob on the front panel. From a human factors (and aesthetics) perspective, since my RME ADI-2 Pro is the device I use for control, I do not want any other device in my audio rack with volume control knob sticking out of the front panel.
 
please stop me from making 95$ useless purchase just because i always wanted to try a balanced DAC, though i think the SU2 would look good on top of the 3e Audio A5..

1769849905786.png
 
No digital volume means no way to fix inter-sample clipping once the signal has hit the DAC. If you're buying, make sure your source can digitally lower the volume before sending audio to the converter.
 
No digital volume means no way to fix inter-sample clipping once the signal has hit the DAC. If you're buying, make sure your source can digitally lower the volume before sending audio to the converter.
it's going to replace (haven't decided yet on the purchase) SU1 actually that doesn't have digital volume as well
 
Why would you want two DACs in the chain?
Because I have multiple consumer-audio digital sources (laptops, WiiM, SqeezeBox) that route into an analog stereo mixer.

Why an analog mixer? Because NOBODY makes a device that can mix digital audio from multiple consumer-audio devices. I even have an RME Digiface USB that is gathering dust in box because I didn't realize it could NOT accept digital inputs whose clocks are not synchronized.

@MC_RME : an updated version of the Digiface USB that can support a simple re-clocking buffer (at a minimum) or even sample rate correction (ideally). There are a lot of digital devices that cannot accept an external clock source and thus cannot be used as an input source. And a lot of audio use cases where precision clocking across-devices is not required.
 
Last edited:
@MC_RME : an updated version of the Digiface USB that can support a simple re-clocking buffer (at a minimum) or even sample rate correction (ideally). There are a lot of digital devices that cannot accept an external clock source and thus cannot be used as an input source. And a lot of audio use cases where precision clocking across-devices is not required.
Do you realize that what you wish is one of the most expensive features, as you'd need multiple ASRCs with decent quality, not your CS8422 shiit? Even in DSPs, that stuff gets expensive when realized in decent quality.
The Digiface USB already is the cheapest "All-in-one" stuff one can buy from RME (thats why I have one as well).
Go out, find a device that supports your wish, below $1000 (and above) you'll not find anything.
 
Do you realize that what you wish is one of the most expensive features, as you'd need multiple ASRCs ...
As noted in my comment, I only need multiple inputs with a simple buffer and re-clocker. Added functionality of sample rate conversion (SRC and/or ASRC) would be even nicer; but I can manage all inputs to provide a common sample rate.

... with decent quality, not your CS8422 shit?
Why wouldn't a CS8422 or similar chip work?

Everyone on ASR says that DACs have been a solved problem for decades. And all that is needed here is re-clocking (at a minimum) or potentially even sample rate conversion if full SRC/ASRC is being provided. Not PEQ, room correction, time / phase alignment, etc.

Besides, since I don't have access to anything else, I'll take a CS8422 crappy chip if that allows me to eliminate the need to convert everything into analog just to route it through an analog stereo mixer (since I can't route those devices into a Digiface USB for mixing).

Even in DSPs, that stuff gets expensive when realized in decent quality.
How much more is the unit cost and added engineering complexity to use a better performing chip than the CS8422 (or similar)?

And even relatively low-cost devices (e.g., Beni_Skate's Automatic SPDIF (Optical/RCA) Audio Switch) are regarded as providing exceptionally high quality re-clocking. And that is NOT an expensive device.

Please help me understand why re-clocking at an already known sample rate is so expensive? Compute power has drastically changed since the Digiface USB was originally released. And I believe a lot of chips can now support re-clocking to synchronize external digital input against an internal clock (but perhaps that is just my ignorance).

The Digiface USB already is the cheapest "All-in-one" stuff one can buy from RME (thats why I have one as well).
That was why I bought it. Fantastic device, with exceptional quality, and TotalMix included.

What is not to love, beside that fact that every single dollar I spent on it was 100% wasted because it does not function for a home office environment. Of course, that is my fault since I did not yet understand the issue of optical clocking and why consumer devices are incompatible with it.

Go out, find a device that supports your wish, below $1000 (and above) you'll not find anything.
I know, I've tried. The closest device out there was a MOTU 8D. But, that only had coax outputs, which are incompatible with the RME optical inputs. And no longer on the market.

As more and more people begin to use RME devices for home offices, podcastings, etc., there are now use cases for accepting input from external consumer-grade input devices such as laptops, WiiMs, CDs, DVDs. That is, use cases where exact time alignment (between the input sources) is NOT required. Contrasted against use cases such as audio engineering where precise time alignment is critical, such as recording audio from vocals, guitars, keyboards, drums, etc.

Which is why I make the plea to RME (@MC_RME) to add this capability to a future release of the Digiface USB — at least with a simple buffer & re-clock for those of us who want to mix multiple digital inputs into the same audio stream; but who do not require precise timing alignment across the inputs.

Edit: alternatively, please also release versions of the ADI-2 Pro and UCX / UFX that can perform re-clocking on all of the digital input channels. The fundamental challenge is that consumer-grade devices do not accept external master-clock for synchronized digital output. But is it that much of a challenge for modern RME converters and interfaces to re-clock internally.
 
Last edited:
Just received the SMSL-SU-2 today. So far, I'm loving it. Excellent neutrality and that slight warmth in the low midrange of the SU-1 has been replaced with a tighter and cleaner sound. A very open and clear sounding DAC even though the SU-1 is already quite neutral.
 
Just received the SMSL-SU-2 today. So far, I'm loving it. Excellent neutrality and that slight warmth in the low midrange of the SU-1 has been replaced with a tighter and cleaner sound. A very open and clear sounding DAC even though the SU-1 is already quite neutral.

They should not differ sonically from the SU-1 unless one of them you have is broken
 
Just received the SMSL-SU-2 today. So far, I'm loving it. Excellent neutrality and that slight warmth in the low midrange of the SU-1 has been replaced with a tighter and cleaner sound. A very open and clear sounding DAC even though the SU-1 is already quite neutral.

SU-1 and SU-2 will sound the same unless they're broken. :)

SU-1 does have an ever-so-slight high-frequency roll-off, but it's likely inaudible. SU-2 should perform better on paper.
 
I have 2 SU-1s and they sound identical. So yes, maybe my SU-2 is indeed broken and sounds better (at least for my taste).

it just a placebo (imagination). You should do a poper blind test

I am sure the difference will disappear once you cannot see what device you are using, as is always the case in such situations.
 
Back
Top Bottom