It’s doubtful the DAC will support anything higher than 96kHz on optical. The connectors just won’t support anything higher. Not sure what makes you think the Mac will send it. Even the maker, Toshiba, will not spec anything higher than 96kHz reliably.
Hi Paul. I’m telling you, I have been listening to a bunch of 192kHz content on a DS DAC via optical input, both from the Squeezebox and from the MacBook Pro. I also played some DSD64 using DoP from Audirvana on the Mac.
The DAC’s screen confirms PCM 192 or DSD64 accordingly. And it sounds wonderful.
You’re lucky. Not all DS will do that. Luck of the draw.
Well this is the store’s demo machine. When my new DS arrives in a few days I’ll test it and let you know whether my luck holds.
I’ve tried the SBT to the optical input of the DSD. I can play 96KHz but I don’t get a lock on 192KHz.
Re SQ - its good considering the cost of the SBT but not in the same class as even an optimized mac mini running A+ into the USB port of the DSD.
My luck has held again: just returned the shop demo unit in exchange for my very own shiny new (black) DS… and it too syncs immediately with 192kHz SPDIF over optical from the SqueezeBox Touch. 
That is good, can you play the bit perfect test file? http://www.psaudio.com/wp-content/uploads/2014/07/BitPerfectTest-24-192.flac_.zip
Yep. I get the little check mark and the words “Bit Perfect” on my DS screen, along with “Optical Input” and “192k 24”.
I’m confused about this. My understanding from this thread is that 24-192 was a limitation on the toslink cable itself, not the DS. Why then is Dvorak able play 24-192 files? It shouldn’t matter what DS he is using, right?
TOSLink was originally designed for cheap connections at 44.1/48k. In a sense the fact that it runs reliably at 96 was luck. With standard connectors and cables you can’t be guaranteed that 192k will work - in fact it probably doesn’t. But with, say, good quality cables, a good transmitter and a good receiver you may get lucky and be able to run at 176.4 or 192k. The DS doesn’t have any artificial limits in that the output of the TOSLink receiver runs directly into the FPGA, but that receiver is only speced to work up to 96k as are most of the TOSLink transmitters out there. Using a longer cable (say 3 meters vs. 1 meter) will cause the upper sample rate limit to be lower - tho once again you might get lucky with the right set of components.
dvorak said Yep. I get the little check mark and the words "Bit Perfect" on my DS screen, along with "Optical Input" and "192k 24".
Hi dvorak, Ted and Paul
dvorak, how long is your toslink cable?
Ted or Paul, with a toslink connection there is other electronics in circuit, enabling electrical conversion TX to light and light back to RX electrical.
Interested to understand please advise
Thanks
Johno
The cable I’ve been using is just under 5’ / 1.5m long. The jacket is printed with: “cinema IQ Polyfiber TOSlink Optical Digital Audio Interconnect”
It’s just something I picked up for around twenty bucks a few years ago from a regular store, not a specialist hi-fi place.
If I try using two cables (I do have something even cheaper than the one above) with a passive optical switchbox in between them, the DAC struggles to lock on even at 88.2kHz. I’ve ordered a $30 active optical input selector (3 in, 1 out) which claims to support 192kHz and even has remote control, so I’m hoping I can have my SBT for high res and DSD64, plus the TV and a spare input for something like an AirPort Express all coming into the DS via optical.
EDIT: Actually, even better… I did the 192kHz bit-perfect test not from the SBT but from my MacBook Pro (Retina, Late 2013, which absolutely does support 192kHz from the built-in optical output). The cable I used in that case was this awesome $10 one:
Johno said Ted or Paul, with a toslink connection there is other electronics in circuit, enabling electrical conversion TX to light and light back to RX electrical.TOSLink was designed to be cheap - there's very little to it.
A TOSLink transmitter can be as simple as a LED and resistor. A receiver is just about as simple except it needs some sort of an amplifier, but they are generally integrated with the receiver module so you have a receiver module on your board with a single wire running to the decoder, in the case of the DS, the FPGA.
Ted and dvorak Thanks very much for the replies - appreciated, be interesting to see weather a longer optical cable can be used and still achieve the bit perfect test.
I’m thinking a cable length of 10 meters
I might have to try this myself I have 1.5 meter length Ill try that first.
Cheers
John
John, bit-perfect testing is mostly about making sure no additional computation is being performed on your audio data (such as level adjustments, EQ, sample rate conversion). When it comes to digital cables, anything which fails bit-perfect transmission at a given rate is broken and won’t play any audio reliably at that rate. For example, with my passive switcher in the middle of two cables I could play 44.1 or 48 audio (bit-perfect) but I could not play 88.2 or higher - at all.
For 10m lengths, and especially at rates higher than 44.1/48, you might be needing a glass cable rather than the common plastic types. For example, http://www.lifatec.com/toslink2.html
mine does not work higher than 96 kz but I have a sound a little better than my Very basic printer USB cable. I also try the coaxial link (very good Lavardin cable) and the sound is as good as the Toslink but pass the 192/32bit without any problem.
I have now an Auralic Aries that is more efficient than the bridge. (minimserver on a NAS 1513+) through a regular RJ45 cable.
I have now to test better USB cable