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

Severe performance regression due to soft particles in 1.5.4 #604

Open
user21944 opened this issue Aug 8, 2024 · 3 comments
Open

Severe performance regression due to soft particles in 1.5.4 #604

user21944 opened this issue Aug 8, 2024 · 3 comments

Comments

@user21944
Copy link

1.5.4 has significantly higher GPU usage than 1.5.3 whenever particles are on screen, going from ~35% to over 80% on my system (Ryzen 5 5600G, Vega 8 iGPU on Windows). This is caused by soft particles, which are now enabled by default.

I'm not sure if the poor performance on my system is a symptom of my hardware or a bug/optimization issue.
Regardless, soft particles should probably be disabled by default so that unaware users aren't faced with significant performance degradation after updating (especially on Intel iGPUs where performance is already limited).

No soft particles, 36% usage:
image
No soft particles, capture depth buffer to texture, 40% usage:
image
Soft particles, 80% usage:
image
Performance is the same in both windowed and fullscreen, windowed was used here to take a screenshot

@xmansch
Copy link

xmansch commented Aug 12, 2024

,

1.5.4 has significantly higher GPU usage than 1.5.3 whenever particles are on screen, going from ~35% to over 80% on my system (Ryzen 5 5600G, Vega 8 iGPU on Windows). This is caused by soft particles, which are now enabled by default.

I'm not sure if the poor performance on my system is a symptom of my hardware or a bug/optimization issue. Regardless, soft particles should probably be disabled by default so that unaware users aren't faced with significant performance degradation after updating (especially on Intel iGPUs where performance is already limited).

No soft particles, 36% usage: image No soft particles, capture depth buffer to texture, 40% usage: image Soft particles, 80% usage: image Performance is the same in both windowed and fullscreen, windowed was used here to take a screenshot

what about perfomance, when you disable soft particales ?

@DanielGibson
Copy link
Member

DanielGibson commented Aug 15, 2024

I'm not sure I want to disable this by default - do you think it would be sufficient if it were disabled when setting a Performance Preset below Ultra?
That would mean that it's still enabled by default, but if someone sets the performance preset to "high" or "medium" or "low" (because performance is bad on their system), soft particles get disabled.
Because of how those performance presets work, having it already set to high won't disable it, though (that would be bad, because you want to allow people to select a preset and then tweak a few settings afterwards)

@user21944
Copy link
Author

I'm not sure I want to disable this by default - do you think it would be sufficient if it were disabled when setting a Performance Preset below Ultra? That would mean that it's still enabled by default, but if someone sets the performance preset to "high" or "medium" or "low" (because performance is bad on their system), soft particles get disabled. Because of how those performance presets work, having it already set to high won't disable it, though (that would be bad, because you want to allow people to select a preset and then tweak a few settings afterwards)

I guess that would be fine.
It just seems like this feature is very expensive for what it does

DanielGibson added a commit that referenced this issue Aug 15, 2024
When applying the currently configured graphics quality preset
(it's set in the `com_machineSpec` CVar and applied with the
`execMachineSpec` command, or by using the menu), now soft particles
are disabled for all quality presets except for ultra, because this
feature has a noticeable impact on performance with some (slower) GPUs.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Development

No branches or pull requests

3 participants