Joachim Breitner

(I don't use Wayland yet, and have no idea how arbtt will behave there)

Ah, that might be an explanation. `arbtt-capture` is a bit naive in it’s interval calculation, and will always sleep the configured 3s between taking captures; not taking the time it...

My theory is that even with 60s interval, this will even out (you’ll catch them less often), but yes, depending on your needs and use cases this might be too...

#145 might improve the situation. Do you want to try it out? Can you build from a branch or need help with that?

This might be an old file, look at the time-stamp. Modern `cabal` puts its files in `dist-newstyle`.

Hmm, maybe my changes on the branch are just broken. :-/. Will have to careful look at them again.

Did you run multiple arbtt-capture instances at the same time maybe?

Actually, the mailing list still works (I think) so you can also ask there https://lists.nomeata.de/mailman/listinfo/ Is it none of the tools at https://arbtt.nomeata.de/doc/users_guide/contributed.html

Thanks for the report. `arbtt` gets the “last activity” information from the X server using the [`XScreenSaverQueryInfo`](https://linux.die.net/man/3/xscreensaverqueryinfo) function call. I would assume that a screensaver would also not get activated...

Sorry for the late reply, I was traveling offline. Can `arbtt-recover` fix the issue?