Closed jorgerobles closed 7 years ago
Try limiting the number of workers to the number of CPU threads - 1. Ex: my desktop is a 4-core 8-thread machine, so it'd be 7 on it. There is no browser API that reports this so it'd have to be a setting.
Is a setting indeed :) I've done a naive approach using setinterval to percent dispatching. Seems good but maybe there's a better way.
El 21 may. 2017 20:58, "Todd Fleming" notifications@github.com escribió:
Try limiting the number of workers to the number of CPU threads - 1. Ex: my desktop is a 4-core 8-thread machine, so it'd be 7 on it. There is no browser API that reports this so it'd have to be a setting.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/LaserWeb/LaserWeb4/issues/318#issuecomment-302955642, or mute the thread https://github.com/notifications/unsubscribe-auth/ABoIYHQty51o9MlxoQvaFGKoLvJPSm-Lks5r8IlfgaJpZM4Nhrm8 .
I'm quite in shock. A ~44sec computing to merely 1s540
I'm having hard time with parallel webworkers concurrency and react response. The most patent case is that LW is unresponsive on Gcode generation due to parallel processing choke. Can anyone lend me a hand?