wgpu: Some optimisations, some refactors, much less buffers #14919
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
pony-creator-v3-generalzoi.zip
|
|
@SN902 web or desktop? Do they look normal to you before this PR, no rendering artifacts or crashes? |
I tested your PR on the desktop build, these games works fine on the latest nightly 2024-01-24 |
|
Try now. I think I misunderstood the max_uniform_buffer_binding_size limit. |
Fixed, now works perfectly. |
torokati44
added
A-rendering
Aaron1011
approved these changes
Jan 27, 2024
…ms, no longer dynamically allocated
…nd it should be same perf now
…Builder when making meshes
Dinnerbone
force-pushed
the
staging_belt
branch
from
fda8cf2 to
efb6ef9
Compare
Lord-McSweeney
removed
the
waiting-on-review
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This reduces the amount of buffers needed to run Heartcore from about 8k to 3k peak, which should improve performance and reduce memory usage. We also don't need a second command buffer per draw anymore. Filters, especially, should be cheaper to do now.
It also simplifies the codebase a lot since we no longer have to juggle lifetimes with our old staging belt system, and we also don't specialcase push constants since the performance should be comparable now. This opens up future improvements, especially to BMD draws (and making BMD operations on gpu viable, to avoid readbacks)