-
-
Notifications
You must be signed in to change notification settings - Fork 2.4k
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
Fps drops in beatmaps with heavy storyboards #3756
Comments
Heavy storyboards will result in some frame drops. They can perhaps be optimised a little bit more but they’re already pretty well off and pretty much on-par with osu!stable. Marking as low priority. |
I'd like to disagree with that statement. Heavy storyboards drain fps much more in lazer than they do in stable. For example in this map, I can get ~400 fps in stable, yet only ~45 in lazer, making it already lower than the monitor refresh rate, which is always bad. So I'd argue either storyboards are way off from being optimised or something else is causing fps to drop that low. |
I see, indeed the game work well with the other maps. |
There is similar issue on ginkiha - EOS by alacat. The fps becomes lower and lower during play (minimum is 40 fps for me). Here is a video: |
Similar issue with this Taiko map, with storyboard enabled |
Lazer lags pretty hard on my pc when playing Trail of Dust - TRakker, and that does not happen on stable at all... |
This is a case where the transforms for many sprites starts at 0 with a duration of 0, and we can probably optimise this by adjusting such sprites to take their lifetime according to their alpha. I imagined this as a post-processing step only during gameplay or similar and should not be baked into the transforms, as this wouldn't be desired in the editor. |
Whatever solution is implemented should scale across to the editor. |
Currently stoyboard (script and player) could be optimzed by these method:
osu!stable hasn't impl anything for optimzing osb script and other yet , so there are some storyboard run bad in old PC : |
Recently I came across this map, which should be added to the list of "fps-dropping storyboards". Another one I came across has some very slight but still noticeable fps drops. |
The same issue happened in this map Sendan Life (416153) which is a heavy storyboard
As you can see,
EnvironmentVersion:
Specs:
LogsAfter viewing the logs, it doesn't have much special, so I'll replace with video. (All links are Youtube) |
Yes, it will happen for all storyboards. |
Would it theoretically be possible to "render" the storyboard beforehand, and when the storyboard is played, display this "rendered" version instead of making the storyboard "from scratch" every time? |
No further comments are required here for now. We'll get to fixing it. |
Another map: https://osu.ppy.sh/beatmapsets/139525#osu/348549 |
Maybe related: #31678 |
It looks like #31506 enabled storyboards even when 100% dim, so the issue became more apparent for me, and maybe others. I like watching storyboard and artwork in breaks, so disabling it decreases the game quality for me. And I suspect I'm not alone in this. Found two other songs that are affected by this issue: Can this issue get retriaged? |
I'm not sure what you mean by "retriaged", but it's still on our radar. |
I was playing the the map "Sukisuki Zecchoushou - Kano" in the osuLazer that have a heavy storyboard.
So in certain part of the game-play the fps have a huge drop, like 40fps and 20fps.
I testing with vsyn mode and unlimited fps mode and the same thing happens.
I test with other maps too, that have lighter storyboard but nothing happens, so could be something like a functionality that only this map have.
So I the problem could be in the storyboard because if you deactivate it, the game will work normally in 300fps or more with no problem.
My PC:
OS: Linux Manjaro KDE 64bits
Intel Pentium G4560
Nvidia GTX 1060 3gb
Ram 8 gb DDR4
HD 1 Tera
Video that shows the issue: Link of the Video
Osu!lazer version: osu-lazer-git 2018.1120.0-1 get for AUR
Logs:
database.log
debug.log
performance.log
runtime.log
The text was updated successfully, but these errors were encountered: