The BNC clock connectors on the input modules all output wordsync (either 44.1Khz or 48Khz). How they work is the DAC syncs to the incomming stream and determines if it is in the 44.1Khz family or in the 48Khz family then it outputs the corresponding wordsync. So your wordsync input needs to be able to handle changing wordsync and wordsync that is at the wrong rate. For example if you have a ProUSB input and a S/PDIF input and you are using the wordsync output on the S/PDIF input but you have the ProUSB currently selected then the wordsync output will follow the ProUSB not the S/PDIF input. When you switch to the S/PDIF input then the wordsync output will follow the S/PDIF input frequency.
All multiples of 44.1Khz (88.2Khz, 176.4Khz, etc… including DSD) will output a 44.1Khz wordsync.
All multiples of 48Khz (96Khz, 192Khz, etc…) will output a 48Khz wordsync.
If the source switches from the 44.1Khz to 48Khz family, the wordsync will take a portion of a second to follow that change, and will momentarily be at the wrong rate.
Whatever source is selected is the source that the wordsync will follow. If you always want the wordsync to follow one source, then that is the only source you can select.
Sources like MSB transports and the Aurenderer all detect whether the wordsync in is at the correct frequency before they use it internally, and they have a backup internal clock they use when worsync in is at the wrong rate or isn’t present.
@Dustin_Symanski This thread and your helpful comments kind of answer my question whether it would be necessary to get an expensive, optional Innuos PhoenixUSB reclocker board installed on my Innuos Stream 3 server / streamer that I have on order, to run with my Discrete DAC over ProUSB / ProISL. My understanding from a recent podcast on The HiFi Five with Daniel and Jonathan Gullman, is that such a reclocker would not be necessary with an MSB DAC, given the internal clock of the MSB DAC, and all USB noise would be eliminated by ProUSB / ProISL.
Your advice on this would be appreciated as I want to save money by not spending on an unnecessary, expensive, optional extra, whereas the Stream 3 already has a dedicated DAC USB output.
Forgive me if this is a simple question, but I’m not a technical person.
The Pro USB will already reclock the data, I would not recommended adding another reclocker. I doubt this would make an audible improvement, and would be money better saved for other updates
That is not correct, except if video mode is on. If video mode is NOT on then the incoming data will be stored into RAM immediately on reception and played back using the DACs internal clock. The two interfaces that will change the clock source With Video Mode Enabled are I2S and S/PDIF, other inputs ignore the video mode setting since they are all synchronous interfaces. Since both S/PDIF and I2S are usually asynchronous (except if Word-sync feedback is used, such as with an Aurenderer).
I think you meant to say all other inputs ignore video mode because they are asynchronous. Correct? Just out of curiosity, does video mode also mean no oversampling since the source is used for the clock?
All other inputs are synchronous, IE they are synchronized to the DACs clock. S/PDIF and I2S without a clock feedback to the source are asynchronous, IE they are not synchronized to the DACs clock. The DAC stores the data from those in local memory which is then replayed synchronously. Turning on Video Mode uses the recovered clock from the source instead of the DACs internal clock. This causes those inputs to become synchronous and the internal memory is not used, significantly reducing latency and performance simultaniously.
Thanks for the explanation. I now understand how you are using the terms. I usually see USB and Ethernet protocols like Roon RAAT described as asynchronous because their clocks are just used for data transfer. The usb standard everyone uses, UAC2 Asynchronous Mode, even has asynchronous in its name. Anyway, what I am really curious about are any other consequences from using video mode. Are you still able to oversample and use the same digital filter without having the data in memory? I am just curious. I don’t really have a need to know. Thanks.
Video mode is compromised to achieve low latency. Only S/PDIF and I2S actually enable video mode, even though the display will still show it for other inputs. All other inputs are already low latency because they are synchronous so video mode is meaningless for them.
Is the latency without video mode constant, or does it vary? My TV has a video/audio time shift feature to adjust lipsync. I am wondering if it is worth trying to use non video mode in the DAC and adjust lipsync in the TV, or if I am better of staying with video mode in the DAC and no adjustment in the TV.
I tried this, but could not compensate enough with the TV controls. Switching to video mode works great though. I suppose it would be worth going through the TV system again for fine tuning. Does anyone know a method to test the lip synch issue? Perhaps some sort of test video with clear signs on the screen to align a sound with?
Put the DAC on video mode, and then do the TV’s lip-sync. This is the way..
I have an AppleTV, that AirPlay’s to a WiiM that’s connected via SPDIF to my cascade and with Video Mode on and doing the audio sync with the AppleTV’s process using the iPhone’s microphone the sync is perfect.
Yes, it’s in the AppleTV audio settings. It then pairs to your iPhone and plays some sounds and then its perfect. Even better is Apple made an update allowing the AirPlay to an external source to be a persistent setting so the AirPlay WiiM device doesn’t revert back to the TV speakers every time it turns on. I don’t have a conduit behind my TV to run a cable so this is the perfect solution and it’s as good as hardwired.
Have you tried the Apple TV sync with the DAC not in video mode? I watch a lot of concert videos and music television shows like Austin City Limits. It would be nice to have top performance from the DAC for this.
I have not, I don’t watch that kind of content but I just tried it with my existing calibration and the sync is really close. If you resync the appletv it should sync to whatever the buffer size is.
The picture shows the item you want to select to do the sync.
Thanks.