I am changing equipment and have an issue. The new USB interfaces I like (Komplete Audio 6/Roland UR-55/Scarlet 6i6) don`t have an optical output. My Parasound DAC1000 has a single optical input and a single coax which I use for the Parasound transport so optical is the only one left. I do not want to swap cables every time so thought of two options:
1. Buy an SPDIF to optical converter. Question is, is it going to degrade sound in any way and any idea of a good one?
2. Rework the DAC1000 by removing the optical input and replacing it with another SPDIF input. Question is, has anyone done it?
Thanks,
Mario
1. Buy an SPDIF to optical converter. Question is, is it going to degrade sound in any way and any idea of a good one?
2. Rework the DAC1000 by removing the optical input and replacing it with another SPDIF input. Question is, has anyone done it?
Thanks,
Mario
I did this many years ago to add an optical output to a Micromega:
SPDIF with Optical DIY transmitter and super clean drive.
SPDIF with Optical DIY transmitter and super clean drive.
Optical is also called TOSLINK. It is more jittery than SPDIF and is not reliable above 96kHz sample rate, whereas SPDIF can work well up to 192kHz. If you don't care about some hi-res audio signals, a SPDIF to TOSLINK adapter should probably work adequately.
Thanks for inputs. I use quad shielded digital cables of very good quality (made by Ixos) so probably best option would be ot modify the dac so it accepts another SPDIF instead of the TOSLINK.
I’d like to see some support for this claim, since ADAT Lightpipe carries eight multiplexed 48kHz channels (for an equivalent wordclock rate of 384kHz) on the same hardware with little enough jitter to have been used for commercial audio mastering for a decade, and the only difference between Toslink and coaxial is the transmission medium (they’re both S/PDIF).
When I played with optical S/PDIF the jitter was exacerbated by the unequal rise and fall times of the photoreceiver. Also known as 'pulse width distortion'.
If you go to the DS for TORX147 Toslink receiver as an example, the max pulse width distortion quoted (for both TX and RX) is +/-15nS, they don't give a typical.
If you go to the DS for TORX147 Toslink receiver as an example, the max pulse width distortion quoted (for both TX and RX) is +/-15nS, they don't give a typical.
@okto,
TOSLINK can be de-jittered. ASRC can attenuate the jitter. FIFO can eliminate it at the cost of a time delay (and or possible FIFO buffer underrun/overrun). Nothing new there.
SPDIF and AES have the same issue as TOSLINK, but tend to work better at higher data rates than the typical optical gear used for TOSLINK. Therefore its pretty standard now to use ASRC as needed in professional applications, as opposed to a SPDIF/TOSLINK/AES derived PLL master clock.
TOSLINK can be de-jittered. ASRC can attenuate the jitter. FIFO can eliminate it at the cost of a time delay (and or possible FIFO buffer underrun/overrun). Nothing new there.
SPDIF and AES have the same issue as TOSLINK, but tend to work better at higher data rates than the typical optical gear used for TOSLINK. Therefore its pretty standard now to use ASRC as needed in professional applications, as opposed to a SPDIF/TOSLINK/AES derived PLL master clock.
Last edited:
But an ASRC will rebuild and interpolate the actual data to accomodate difference between the in and out clocks. They do what they are suppsoed to do.... but is the net effect better sound than the jiiter? I would think not but would like to see it tested.
There are some pretty good ASRC chips (e.g. SRC4392). If the surrounding circuitry it well implemented and if there is a high quality crystal reference clock for the ASRC PPLL (polyphase locked loop) then usually ASRC a pretty good solution. Better than SPDIF derived PLL master clock. FIFO is probably best if the time delay is tolerable, but delay would often be a problem for professional use.