JF
Thanks for the clarification @mark9064. As I said, I try to limit the use of the NRF SDK as much as possible because I had some very difficult time debugging...
Closing this PR since the original issue was fixed by #1869
Hi @liamcharger ! If it' s not too much to ask, could you provide a before/after picture (from a camera or using [InfiniSim](https://github.com/InfiniTimeOrg/InfiniSim)) of your changes?
I've never been satisfied with the synchronization mechanism in the ST7789/SPI driver but I haven't tried to find a better design/implementation since I wrote it a while ago, so I'm...
> No numbers unfortunately - just based off how it seems. I suspect as going through the RTOS adds some waiting due to task switching etc. But equally this could...
> It looks like that if the capture list is small enough, the captured variables fit within the std::function instance and so the heap is not used Yes, that's also...
@mark9064 btw, I can apply all those small changes if you wish :)
> As we don't need to store the lambda anymore, we can just pass in a reference. Yes, you're right. Passing the reference works fine and does the job, but...
I'm a little worried of this new mutex. I applied the original design on purpose : SystemTask is the only class with "write accesss" to the clock. `DateTimeController` is `const`...
Thank you very much for this detailed explanation. Fast, accurate and precise time is indeed a core functionality of a watch so this PR makes sense. However, if you don't...