Entity Culling Fabric/Forge

Entity Culling Fabric/Forge

80M Downloads

Incompability with DynamicFPS?

andreasdc opened this issue ยท 12 comments

commented

Incompability with DynamicFPS? After some time my cpu is at 20% not 4% when minimalized. juliand665/Dynamic-FPS#59
Lagged:
image
Normal:
image

commented

Funky! If you can, maybe try without Dynamic FPS and see if it still happens?

How can it happen without Dynamic FPS?

commented

Funky! If you can, maybe try without Dynamic FPS and see if it still happens?

commented

I need the full thing. 36% on hasNext sounds like something going wrong with mc?

commented

Well I can tell you that the cullthread won't care about being in the background/DynamicFPS doing anything (it's a separate thread just in a while loop with a sleep). Can you send the lagged spark timings?

commented

Well I can tell you that the cullthread won't care about being in the background/DynamicFPS doing anything (it's a separate thread just in a while loop with a sleep). Can you send the lagged spark timings?

In the lagged profile there is no sleeping if i'm not wrong.

commented

I can't tell you that, since spark can't tell you cpu usage based on threads. (and the fact that the profiling like I said is broken)

commented

I'm pretty sure that spark screwed up the profiling there(it's missing a ton of info and the info that is there is wrong), the two mods work perfectly well together, and whatever causes the CPU usage to not lower correctly is not related to EntityCulling.

What is causing it then? You have 2 profilers.

commented

I'm pretty sure that spark screwed up the profiling there(it's missing a ton of info and the info that is there is wrong), the two mods work perfectly well together, and whatever causes the CPU usage to not lower correctly is not related to EntityCulling.

commented

Hm, is it possible it's Spark itself that's getting messed up by Dynamic FPS's backgrounding stuff? Though generally it should only affect the rendering thread, parking it for a short time so as to avoid a busy loop.

commented

Hm, is it possible it's Spark itself that's getting messed up by Dynamic FPS's backgrounding stuff? Though generally it should only affect the rendering thread, parking it for a short time so as to avoid a busy loop.

99% chance that it's not spark's fault, only thing that is different from the normal profile is CullThread's behaviour, probably stuck at the loop and not going to sleep.

commented

Also my guess is that something bad happens near that line and it gets stuck on the loop.
image

commented

Also my guess is that something bad happens near that line and it gets stuck on the loop.
image

That's block entities, they don't even use the Google iterator. Like I said, the timings are garbo.