Bandwidth limit of TOSLink?

I’ve been looking into building a music server for the DS, and there are TOSLink cards that appear to pass 192/24-bit audio, so should be able to do DSD64 also. Is that the limit for TOSLink, or will they someday pass DSD128 and DXD?

I’d like to know because it may prove far simpler/cheaper than I2S via HDMI for comparison with USB. (Pink Faun has a suitable I2S card but it is also limited to 192kHz PCM, and seems to require Windows.)

24/96 is the theoretical limit of TOSLink (it was designed for 48k) Still there are a lot of transceivers out there that, with good cables, can reliably pass 176.4k or even 192k. None (or almost none) will pass 352.8k (DXD and/or double rate DSD via DoP.)

Ah, that’s interesting. The lack of ground issues and the fact that you say it’s “just a TOSLink receiver module and a wire to the FPGA” in the DS makes it appealing. Has anyone compared TOSLink to I2S in the DS using the same source?

By “almost none”, do you mean that you know of a transceiver brand/type that is capable of passing 352.8k? I did a quick search and it seems that optical transceivers used in digital audio are not particularly fast. Fibre network PCIe cards seem to be able to pass 1000Mbps or more. Uh-oh, might be heading down the rabbit hole here …

That’s why I said “None (or almost none)”. I didn’t want to make a categorical statement, but I don’t know of any and I don’t know why someone would build one - it’s better bandwidth wouldn’t help with anything that’s already out there so you might as well use some other standardized interface.

A lot of the experiments I’ve heard of people doing with TOSLink against other inputs are flawed in that they leave the other input connected while they listen to TOSLink. The beauty of optical interfaces is galvanic isolation: keeping things electrically connected while listening to optical connections throws the baby out with the bath water.

The experiments I’ve done between I2S, AES/EBU, S/PDIF and TOSLink from the PWT to the DS show that all of the connections sound essentially identical if you don’t disconnect all wires except the one you are listening to. I.e. leaving everything connected means that you are masking effects of differing shielding, grounding, etc. and since the DS is relatively input jitter insensitive the jitter differences between the various inputs doesn’t matter much.

I’m certainly happy with TOSLink when I don’t need to play anything greater than 176.4k. Unfortunately I broke the TOSLink output from my PWT :slight_smile:

ted, the dsd specs state 24/96 as maximum for toslink input. true? seems to be …i have a new xmos u8 converter box that gives me silence at 24/192.

Yes, TOSLink was originally designed for 48k, but it does 96k reliably and 176.4k at times and 192k somewhat less often. It depends on the particular transmitter, receiver and cable. You may find that a better cable will work, I’d try an inexpensive glass cable instead of an expensive plastic one first. I don’t know what hardware you are using but you might also be bitten by DRM, each DVD, DVD-A or Blu-Ray disc can control whether it’s output will allow any sample rate over 48 (or 96k).

If you have problems finding a toslink cable that will pass the max reliably, go to Sys. Concepts Inc.
They sell a 1300 strand glass toslink at a very reasonable price that has always performed flawlessly for me.

1 Like

Ted, My bad …i just pushed in the TOSLINK cable further into the DS …a proper click. Now I can do full 24/192 no problem. And yes, i have pulled all other connections …so my DS is fully isolated from RF noise (apart from what creeps in thru the AC Mains). The sound is spectacular and, frankly, I am not wanting for USB with all its attendant RF noise problems. Wonderful, thanks. For those interested in the USB-to-SPDIF i used, it’s here. All of $68 shipped from China to my house.

Purchased the same toslink converter. Got it today. Best $50 upgrade ever. Beats curious usb with and without regen. Less grain more texture - more dimensional etc. PS should recommend that the DS be used with Toslink. Would think it would beat I2S also. It’s not a subtle improvement… Thanks for the great tip.

Thanks for some great foundation information on TOSLINK (in this and many other threads). Sorry to revive this old one, but I was wondering the very statement you made here for a while; “I depends on transmitter, receiver, and cable…”.
So is it safe/accurate to generalize and say that not all TOSLINK set ups are any more equal than cheap RCA on cheap components with cheap cables, verses expensive components with higher end RCA ports, and higher technology quality cables?

I had been assuming that TOSLINK was sort of, ‘absolute’ I guess? Meaning I get the most bandwidth, the most data at the most-close-to-source as possible that way WITHOUT any sort of interpolation, modulation or ‘processing’ other than the actual modulation of Digi to light and back again. I was assuming this was again, a ‘near absolute’ function, with the only real variant being a cheap quality plastic cable, vs a higher quality plastic or glass cable.

Thanks for any thoughts!

TOSLink is every bit as reliable (e.g. no bit errors with a reasonable setup) as any other digital connection.
Remember that in audio, S/PDIF, AES3, TOSLink, USB, I2S there is no error correction: fundamentally you can either have perfect data, but not know when it might get there (e.g. how many times does it have to retry?) or you can have data get there in a known time, but perhaps with errors. Between computers, disks, networks, etc. the choice of perfect data (perhaps a little late) is made. In audio, the choice is on time data, perhaps with errors.
But there’s no reason except bandwidth that an digital audio connection can’t be essentially perfect: the DS (and various other DACs) provide a bit perfect test (a known pattern that exercises many bit pattern changes) which can help verify that the connection is good and nothing is mangling the data.
The original TOSLink design only supported lower sample rates, but many transmitters and receivers can handle higher rates accurately, with a good enough cable. The cheapo cables that come for free with many consumer video gear has too low of a bandwidth, but most slightly better cables can do 24/176.4 or 24/192. If in doubt, run the bit perfect test.

1 Like

So interesting! I can see the reasoning of sacrificing of data time for data accuracy between computer or audio data. I am wondering if I am experiencing the issue of cheapo cable and suffering bit rates currently in the audio from my TV (cable signal or Chromecast streaming) with using the TOSLINK out from the TV, vs the RCA from the actual cable box. Weirdly, the RCA audio out from the TV to my amp (Rega Mira) sounds WORLDS better than the TOSLINK out from the TV to a Rega DAC~r (playing a movie). As in, the TOSLINK actually is quieter, it’s not properly balanced left and right channel, no dynamics, etc. When switching the amp to the RCA in from the cable box, it sounds at least proper/“good”. I think I will search and figure out about the Bit Perfect Test you mentioned and perform that, and get a couple different better quality/brand TOSLINK to test out as well.

Thanks for your expertise and help!! Have a good week,
(PS: my father has one of the earliest PS Audio Pre Amps (first or second gen?) Do you all ever do upgrades/testing on them at all?)

No, the kinds of differences you are likely hearing are caused by different streams of data, not changes in timing or corruption by TOSLink.

You can’t expect that the bits from RCA on a cable box are any where near the same as those from the TV connected to that cable box via RCA or TOSLink.

It takes work to keep most TVs, etc. from changing the data: music channels on cable, satelite, etc. are usually compressed in a lossy manner. TVs often have dynamic compression modes (for nighttime viewing, etc.) TVs often are doing Dolby Digital decoding (sometimes even when the source isn’t DD encoded.
Video sources also often have processing for speaker placement, sub crossovers, etc. as well as things like stadium emulation, etc.



Oh I see what you are saying. Sort of to be expected I would think: a TV is NOT concerned about audio accuracy or maintaining signal quality and purity, instead opting/needing as much various basic functions as possible, which might include things like, ‘always doing DD encoding whether required or not’. I was improperly assuming maybe that TOSLINK from cable box to TV, then TOSLINK out of TV was somehow like a ‘bypass’ or ‘as close to pure unfettered data’ as possible. Sort of like a complex piping network that you can shunt from going thru one set of electronic functions and just be a ‘thru put’ sort of function. But I do see what our are explaining, and why it ISN’T that way (unless ya wanna pay $1800 for 40" “hifi” TV!)

I did find the article ‘how to’ in performing the bit perfect test, and I’ll give that a try today just to see what results, and check out a couple other cables, etc.

Have a great week Ted!