Code

Frustum-cull small draws #17808

Code

Merged

hrydgard merged 10 commits intomasterhrydgard/ppsspp:masterfrom frustum-cull-small-drawshrydgard/ppsspp:frustum-cull-small-drawsCopy head branch name to clipboardDec 9, 2023

+371-37

Commits

Commits on Dec 9, 2023

• Frustum-cull small draws

  Show description for 0905b6a
  hrydgard

  committed
  0905b6a

  Copy full SHA for 0905b6a

• Use a less accurate but faster frustum cull for the general draws.

  hrydgard

  committed
  89d8ef8

  Copy full SHA for 89d8ef8

• Fastcull: SSE/NEON-optimize 16-bit position conversion

  hrydgard

  committed
  dbf796b

  Copy full SHA for dbf796b

• World space planes

  hrydgard

  committed
  a043962

  Copy full SHA for a043962

• Flip the cull plane data around to avoid transforming each vertex multiple times.

  hrydgard

  committed
  62c936b

  Copy full SHA for 62c936b

• SSE-optimize the frustum culling

  hrydgard

  committed
  5b44e25

  Copy full SHA for 5b44e25

• NEON-optimize the culling

  hrydgard

  committed
  6a7ef83

  Copy full SHA for 6a7ef83

• NEON culling: Use mla operations to shave off some more cycles. ARM32 compat.

  hrydgard

  committed
  99548be

  Copy full SHA for 99548be

• NEON: vcvtq can scale directly, no need for a mul by const.

  hrydgard

  committed
  4e2a1bf

  Copy full SHA for 4e2a1bf

• Disable the new culling on RISC-V for now.

  hrydgard

  committed
  7e85d3d

  Copy full SHA for 7e85d3d