S/PDIF Toslink or Coax? Computer => DAC

[Lupin]

I know, I know, the question has been asked a thousand times before but couldn't find an definitive answer for my situation.

I have a DAC (SMSL M200) of which the USB input is busted so I only have the optical or coax input available.
What is the best way to go?

1. Toslink out from my motherboard (Gigabyte X399 Aursus Gaming 7) to Toslink in on the SMSL M200

2. USB from computer to Singxer F-1 Audio Bridge and from there Coax to SMSL M200

I've tried to search a bit here and on DuckDuck but couldn't find much that relates to my situation as described above.

My common sense tells me that Toslink and Coax are both using the same S/PDIF protocol so there shouldn't be any difference there. But then there is the part of implementation..
Is the Singxer F-1 with it's own clock (audibly) better then whatever comes out of the Toslink connection straight from the motherboard?
Have the feeling that the Toslink out on the motherboard is more a "convenience" option rather something they spend time and money on to get it the best it can be.

Any recommendations?

 

[Thalis]

I would just use the Toslink cable.

 

[3125b]

Is the Singxer F-1 with it's own clock (audibly) better then whatever comes out of the Toslink connection straight from the motherboard?

Normally I would say clearly no, digital is digital, however I just recently checked the TOSLINK output from my mainboard (Asus Crosshair VI Hero) with some DACs - turns out there is a major problem with it (might be different with yours).

Running the Topping E30 from the TOSLINK output of my RME ADI-2 Pro FS (connected via USB to the PC) I get the same >110dB SINAD as with USB to the DAC, however running it from the TOSLINK output of my mainboard I get 74dB with both a high noise floor and some clipping-like artifacts. Regradless of Windows audio setting (16 or 24bit, 44.1 or 1192kHz, doesn't matter) and largely Windows volume level (at -10dB the clipping.like artifacts fade, however the SNR gets even worse with comparatively higher noise floor).
I'm not sure what that is or why, still looking into it, and haven't yet done proper listening comparisons, but you might want to be aware of that.

A TOSLINK cable costs 5$, just try that first and see how it sounds. If it's fine, then there is no need for anything else. If it's audibly worse than say your phone (just about all smartphones are audibly transparent at full volume) as the DAC, you might want to find another solution.

 

[Vincent Kars]

Toslink is limited to 24 / 96 kHz, coax in general does 24 / 192.
In practice Toslink can often do 176 or even 192 depending on the hardware used.
Toslink is optical hence a perfect galvanic isolation.
In general Toslink generates more input jitter than coax.
Modern DAC’s are very good in the rejection of input jitter.
You probably will have a hard time to hear the differences between coax and toslink or even async USB because of this.

https://www.thewelltemperedcomputer.com/Intro/SQ/USB_SPDIF.htm

 

[mansr]

In general Toslink generates more input jitter than coax.

Why do people keep saying that?

 

[Lupin]

Seems like I forgot to mention that I use the DAC 99% for music listing, that means 16/44.1 FLAC.
So I'm not worried about the potential extra bandwidth of 24/96 vs. 24/192 for Coax. From what I understand games are all 16/48 so even for the little time that I game (only stereo, no surround sound or anything) the bandwidth of optical is more then enough.

Reading the replies it seems that optical should be just fine and that the price of the Singxer F-1 is not worth it (in this case).

 

[Lupin]

Why do people keep saying that?

That what I read everywhere too..
So what is the real deal then??

 

[Bullwinkle J Moose]

I use both

Optical goes to Bluetooth Transmitter
Coax goes to DAC

Both supply a signal simultaneously so either is usable instantly

 

[Thalis]

I have a Pi3B+ with a SPDIF HAT to Topping E30.... Toslink and Coax connected at the same time... AB'd and no difference. Stuck with Toslink.

 

[simbloke]

Why do people keep saying that?

It seems to have been written all over the place for years.

 

[mansr]

It seems to have been written all over the place for years.

That doesn't make it true.

 

[abdo123]

Why do people keep saying that?

i think it has to do with the fact most people use short cables, optical will obliterate coax in any distances above 1 meter, but below that coax slowly takes advantage.

either way the galvanic isolation of optical makes it superior SINAD wise.

 

[mansr]

i think it has to do with the fact most people use short cables, optical will obliterate coax in any distances above 1 meter, but below that coax slowly takes advantage.

Got any evidence for this?

 

[abdo123]

Got any evidence for this?

that jitter increases with cable length? it's common knowledge.

 

[mansr]

that jitter increases with cable length? it's common knowledge.

It's a common misconception. That doesn't make it true.

 

[UCrazyKid]

The Singxer and your DAC use the same XMOS chip and reclock, so the Singxer is not going to gain you anything over just using your DAC, either USB or optical in.

 

[Vincent Kars]

Why do people keep saying that?

Because they do.

https://www.thewelltemperedcomputer.com/Intro/SQ/Toslink_Coax.htm

 

[mansr]

Because they do.

https://www.thewelltemperedcomputer.com/Intro/SQ/Toslink_Coax.htm

I see a quote of Charley Hansen making that same claim with no evidence. He also insisted that myrtle wood cable lifters were beneficial, as well as other outlandish things. Then there are a couple of graphs without a clear description of what is being measured (the link to the source is dead). At most, they show some kind of difference for one particular product.

 

[Vincent Kars]

Probably this one
https://hometheaterhifi.com/reviews...3se-special-edition-universal-blu-ray-player/

 

[somebodyelse]

either way the galvanic isolation of optical makes it superior SINAD wise.

Coax often has galvanic isolation too - look for a pulse transformer close to the connector, usually at the sending end.