Blender Git Commit Log

Fix compile error on gcc after last commit

error: static assertion failed: "invalid limits"

Fix T51297: Use A Higher Hair Segment Limit

Normally, segments up to 50 can be quite enough for most cases.
However, when dealing with things like braids,
the current limit can sometimes be quite a pain.

Update draw manager cache when change any RNA property

Gawain: further cleanup to VertexBuffer_prime

Get buffer size once, use it to both allocate and track VRAM.

Gawain: cleanup VRAM usage metric

same functionality

Gawain: revert recent VRAM usage metric

Revert 7a18ee62eb4d6c6028d05f1da259fe8695f49a3f and 1ff97bbfff78a0c375fb5256a9d9d37cd3973bbe after discussing with @fclem.

VertexBuffer_size should always report the same buffer size, but without asking/calling OpenGL.

Cycles: Fix native only kernel since denoiser commit

Mesh Batch Cache: Use compressed format.

We don't compress UV by default because that would mean restricting them to [0, 1]
Leaving all other code for later settings switch.

Use better prng algo and fix randomness UI

Implement matcaps for hair

Gawain: Fix memory manager.

Eevee: fix some glsl for low quality lamps.

Mesh Batch Cache: get rid of the ORCO VBO data, and reconstruct it in shader.

With only one MADD instruction we recover the orco data and reduce both the storage and the fetching cost of an attrib layer.

Gawain: Add assert for maximum attribute name.

Gawain: Better metric for VRAM usage.

Mesh Batch Cache: Define Compressed format for shading data.

Deactivated by default.
All shading attribs can be packed to take less VRAM at the cost of precision (not noticable in this case).
UVs can be packed into I16 but that limits their positions into the [-1, +1] range.

This could be a setting option in the future.

Mesh Batch Cache: Optimization, make use of gawain attrib aliases.

Gawain: Add support for multiple attrib names.

This adds the possibility to give multiple name to the same attrib.
This allow multiple binding points depending on the shader.

Edit Mode Engine: Fix loose edges/verts not displaying.

Mesh Batch Cache: Split Edit Mesh mode Vertex Buffers for faster update.