dumblob

Any news on this? If that still won't make it under [1ms margin](https://www.youtube.com/watch?v=vOvQCPLkPt4 ), then let me suggest giving a try to [Blend2D](https://github.com/blend2d/blend2d/ ) as the fastest existing vector backend...

> Thats an interesting video. Currently 1ms is not really possible, as not even the polling rate of drawing tablets allows for that. Oh, I didn't know that. That's interesting....

>But I am interested, which issues did you encounter while using piet? It mostly boils down to the fact that Blend2D does not use any pipelines and instead tries to...

Mind there might be more than one state - so `load_state()` and `save_state()` would need to work with a list of states rather then with exactly 2 states.

@salifm that's usually handled by a system of so called *capabilities* which is a really cool and [simple concept](https://github.com/WebAssembly/WASI/issues/1 ) (and fits well the os-level bill).

Thanks for the suggestion @Rubo3 . What is it expected to do? On the Linux machine I currently sit in front of (Arch Linux with systemd), this is what happens...

Ok, `cpud -d -init` under root seems to fix the bind issue and I can connect with `cpu localhost` on the same machine. Thanks for the pointer! Would you know...

@dswd how about periodical "reconfiguration" based on randomization of the list along with some latency/troughput measure over the last few seconds/minutes? That way it wouldn't require use interaction/prioritization and should...

> > I would rather add a logic that makes the first address that works win the race. Would this work in your scenario? > > In this case, this...

@biolim the current mechanism doesn't compute any price - it's binary "either or". But I'd prefer if it measured & estimated latency and bandwidth and also acted upon after reaching...