Commit graph

8 commits

Author SHA1 Message Date
Bill Currie
cd4791c5d3 [vulkan] Get window resize working again
The biggest change was splitting up the job resources into
per-render-pass resources, allowing individual render passes to
reallocate their resources without affecting any others. After that, it
was just getting translucency and capture working after a window resize.
2023-07-03 20:05:27 +09:00
Bill Currie
614ca744ab [vulkan] Support multi-layer OIT rendering
This fixes fisheye rendering. I'm not too happy with always allocating
the cube OIT heads buffer, but that's for another day.
2023-06-26 18:00:46 +09:00
Bill Currie
b9bb841744 [vulkan] Use the descriptor set manager
This cleans up a lot of ugly code that I always thought was rather dumb.
2023-06-24 21:42:46 +09:00
Bill Currie
5140346c22 [vulkan] Nuke the old render pass code from orbit
Gotta be sure :)

With the new system mostly up and running (just bsp rendering and
descriptor sets/layout handling to go, and they're independent of the
old render pass system), the old system can finally be cleared out.
2023-06-24 10:42:27 +09:00
Bill Currie
bba82d3da1 [vulkan] Move frames from vulkan_ctx to renderctx
Part of the command buffer fix, but also a step towards cleaning up
vulkan_ctx.
2023-06-22 16:47:15 +09:00
Bill Currie
17c3f12ba2 [vulkan] Add a cvar for oit fragment buffer size
My laptop can't have a buffer larger than 128MB, but 16M fragments wants
512MB. Also, someone running at 4k will probably want a larger buffer.
2022-12-02 13:34:33 +09:00
Bill Currie
966ef949c5 [vulkan] Move oit blending into compose subpass
I had debated putting the blending in the compose subpass or a separate
pass but went with the separate pass originally, but it turns out that
removing the separate pass gains 1-3% (5-15/545 fps in a timedemo of
demo1).
2022-12-01 23:03:55 +09:00
Bill Currie
00cade072c [vulkan] Implement order implement transparency
It's a bit flaky for particles, especially at higher frame rates, but
that's due to supporting only 64 overlapping pixels. A reasonable
solution is probably switching to a priority heap for the "sort" and
upping the limit.
2022-12-01 03:00:47 +09:00