spring-multitheaded.exe is only ~1% slower and by default runs in single threaded mode with no known compatibility issues. It is not worth it to have two executables imo. Speak up now or forever hold your peace :)
Edit: To clarify, the current spring-multithreaded.exe would be renamed to spring.exe
anything against removal of spring.exe?
Re: anything against removal of spring.exe?
k I am working on my benchmark system again, but all GML stuff is enabled by default yet!? Please bring it in a testable/request state first ...
Re: anything against removal of spring.exe?
No, it is only enabled in BA by default, because of modinfo.
Edit: And if you want to run BA with 1 thread use
MultiThreadCount = 1
Don't forget the SetCoreAffinity if you do any kind of benchmarking.
Edit: And if you want to run BA with 1 thread use
MultiThreadCount = 1
Don't forget the SetCoreAffinity if you do any kind of benchmarking.
Re: anything against removal of spring.exe?
odds are spring-mt & evens are default
Wished the error would be less :/
Wished the error would be less :/
- Attachments
-
- benchmark_total2.png
- (34.57 KiB) Downloaded 4 times
Re: anything against removal of spring.exe?
Yeah, hard to make any conclusion from that, except that the two versions are within +/-5% of each other. But nice anyway.
Re: anything against removal of spring.exe?
@jk:
can you upload the plain data so other things can be calculdated? (like average sim time in total or so...)
can you upload the plain data so other things can be calculdated? (like average sim time in total or so...)
Re: anything against removal of spring.exe?
IMO it should be total sim frame time and total number of frames rendered, like in my benchmark:
http://pastebin.com/9HkhftQd
http://pastebin.com/9HkhftQd
Re: anything against removal of spring.exe?
hmm, thats my "plan":
-fix:
1. http://springrts.com/mantis/view.php?id=3340 (assertion failed)
2. http://springrts.com/mantis/view.php?id=3339 (desync)
3. merge https://github.com/abma/spring/commit/d ... 21fbfd9298
- test-release
with 1. i need help... :-/ is the assertion wrong, or not?
i think i can do 2 (at least find the cause) & 3 is a signle command :)
-fix:
1. http://springrts.com/mantis/view.php?id=3340 (assertion failed)
2. http://springrts.com/mantis/view.php?id=3339 (desync)
3. merge https://github.com/abma/spring/commit/d ... 21fbfd9298
- test-release
with 1. i need help... :-/ is the assertion wrong, or not?
i think i can do 2 (at least find the cause) & 3 is a signle command :)