feat(render): relative-to-eye for mesh, image, face3d and hatch pipelines

Switch the remaining world-space shaders (mesh, image, face3d, hatch,
hatch_batched) to the same relative-to-eye transform the wire pipeline
uses: subtract the eye (high+low f32) per vertex and multiply by the
rotation-only view-projection. Each shader's Uniforms struct is widened
to the full shared layout so the appended RTE fields land at the correct
offsets. Math is unchanged (proj·R·(pos−eye)); large coordinates no
longer jitter on camera movement.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Hakan SevenandClaude Opus 4.8 committed 2026-06-23 18:28:52 +03:00
1 parent 2a9416464e
commit a0b4968ad2
5 files changed
+67 -27

No files matched your search

+7 -1
View File
@@ -12,6 +12,12 @@ struct Uniforms {
flat_shade: f32,
transparency_enable: f32,
_pad: vec2<f32>,
// Relative-to-eye (double-single): see wire.wgsl.
view_rot: mat4x4<f32>,
eye_high: vec3<f32>,
_pad_eh: f32,
eye_low: vec3<f32>,
_pad_el: f32,
};
@group(0) @binding(0) var<uniform> u: Uniforms;
@@ -46,7 +52,7 @@ struct VertOut {
@vertex
fn vs_main(in: VertIn) -> VertOut {
var out: VertOut;
out.clip_pos = u.view_proj * vec4<f32>(in.pos, 1.0);
out.clip_pos = u.view_rot * vec4<f32>((in.pos - u.eye_high) - u.eye_low, 1.0);
out.clip_pos.z = out.clip_pos.z - img_params.draw_depth * DRAW_ORDER_BIAS * out.clip_pos.w;
out.uv = in.uv;
return out;