Switch to batched CPU side primitive processing #16

Open
opened 2026-04-24 01:14:09 +00:00 by zack · 1 comment
Owner

Right now the basic draw procedures immediately process the CPU side operations and transform the data to what will be uploaded to GPU. We can probably batch these operations on the CPU more efficiently by caching the CPU side primitives and operating on like kind primitives together (likely with significant SIMD).

Just need to make sure we do this in a way that maintains proper ordering.

Right now the basic draw procedures immediately process the CPU side operations and transform the data to what will be uploaded to GPU. We can probably batch these operations on the CPU more efficiently by caching the CPU side primitives and operating on like kind primitives together (likely with significant SIMD). Just need to make sure we do this in a way that maintains proper ordering.
zack added the
enhancement
proposed
package
draw
labels 2026-04-24 01:14:09 +00:00
Author
Owner

This will probably be the same change in which we add multi-window support and move away from storing all draw caching globally.

This will probably be the same change in which we add multi-window support and move away from storing all draw caching globally.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: BFPOWER/levlib#16