Not at all! DSPi is a firmware for microcontrollers, while CamillaDSP is software that runs on various operating systems. They have very different working principles separate codebases.
I have been looking into this and I believe that 10-15 ms should be achievable, provided both sender and receiver are connected to wired ethernet.
Because wifi is a packeted format and isn’t designed real-time operation.
So a bit better than AES67 and Dante, and worse than AVB. The first two need some reasonably common QoS support in the router(s) while AVB needs hardware timestamping and guaranteed latency support which are somewhat rare.
Because WiFi doesn't have protocols to offer the sort of QoS or latency that wired connections can, particularly the guaranteed latency that AVB uses to keep its latency so low. So you get a trade-off between making the buffer small to keep latency down and making the buffer large to avoid dropouts. And it's a hard one to balance because the latencies will be different in different environments depending on the local radio conditions - a busy city centre office is very different to an isolated house. We can't even run PTP over WiFi because it can't cope with the range of latency variation, hence wireless speakers using proprietary sync mechanisms.
Yes, a dedicated network or crossover is definitely much more robust. This does defeat the purpose of it for many people who just want to make use of their existing networks, of course.
Similar to AES67 but without guaranteed latency and significantly slower than hardware Dante (<1ms). In practice, I would probably target around 20-30ms for stability.
That’s essentially what WiSA is. Based on WiFi hardware but a different protocol than your normal WiFi that achieves lower latency
[former network engineer here] Yes, but those are pro standards requiring pro kit and proper planning to work reliably in their intended setting of studios and venues. You'd have to be doing something very odd in a home setting to truly require separate physical networks. The data rate for audio at home isn't ever realistically going to be over 10mbps, so 1% of a 1gig connection, and latencies per frame are in the microsecond range. It's possible if you have a really crappy switch that it might drop a frame or two, but if you are using AES67 you're using decent network hardware, right?
Thank you for your kind follow-up!
I'm not asking for real time - that's why I said 200ms latency. 200ms is a massive buffer.
I assumed that was a typo for 20ms because you also said 'low latency' and 200ms is certainly not low for latency.
Thank you!
sorry, Its a little confusing because its stated the diagram is "low-pass filter for conversion to analog audio." sub out PDM
Thank you!
Yes. The above diagram is for all coaxial SPDIF outputs.
Please always add a series cap, 10uF or so, for DC protection!
