Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Idea for affinity #20

Open
gllxflr opened this issue May 6, 2024 · 4 comments
Open

Idea for affinity #20

gllxflr opened this issue May 6, 2024 · 4 comments

Comments

@gllxflr
Copy link

gllxflr commented May 6, 2024

A useful option would be: when a world is chosen, give that instance all cores. remove all affinity limits from it.

I run a Ryzen 5 3600 so don't have the cores for advanced affinity, but also have lower single-core performance. Adding this option would greatly improve my in-game experience while not sacrificing reset efficiency.

@tesselslate
Copy link
Owner

In theory, advanced affinity should work fine on a 6-core CPU like the 3600. I used to use it when I had a 3600, and plenty of people still do on Windows with Julti, which has a similar affinity system. However, I don't remember if the 3600 has 1 or 2 CCXs.

If you're using sequence affinity instead, I believe this behavior should happen when setting active_cpus to 12. Lowering background_cpus may help a bit as well. Regardless of what you do, it's not unusual to have some very brief lag when initially tabbing into an instance while other instances are generating worlds.

@gllxflr
Copy link
Author

gllxflr commented May 12, 2024

Thanks for your help, I'll try this next time I do runs
About the lag - it was below 60fps for about 30 seconds upon any dimension load and sometimes otherwise, despite no other instances being active. Was just 3600 single-core performance I assume.

@tesselslate
Copy link
Owner

Hm, that is unusually bad. Does it still happen when not using resetti?

@gllxflr
Copy link
Author

gllxflr commented May 12, 2024

It doesn't, but it also doesn't happen if affinity is disabled. The issue appears to be mostly that single-core performance is not good.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants