Commit Graph
7 Commits
Author SHA1 Message Date
Ampandheaust 4784ad7e87 Render continuously and optimize the glTF viewer
Remove vsync and dirty-flag frame gating, optimize PBR sampling and lighting, add a direct-to-canvas loadout, and preserve camera state on viewer resize.

Add regression coverage and update rendering documentation.

Amp-Thread-ID: https://ampcode.com/threads/T-01a0b382-1e79-712b-abd1-d29201c9bb5e
Co-authored-by: Heaust Azure <heaust.azure@gmail.com>
2026-09-18 08:58:27 +00:00
Ampandheaust bc957cf40d handles: indirect draw buckets for opaque and mask meshes
Opaque/mask draws sharing a pipeline and bind set become rows of one
persistent `draws.<id>` indirect array (32-byte indexed commands,
geometric capacity, inactive tail zeroed) drawn by a single graph pass,
so the graph holds one pass per bucket instead of one per mesh.
Transmission and blend draws stay direct and ordered after the buckets.
Scene requires the indirect-first-instance feature. Compiles re-assert
the bundle block when they start, since an earlier switch in the queue
clears the one set at enqueue time.

Amp-Thread-ID: https://ampcode.com/threads/T-01a0b2dd-db32-712d-99ce-12ce4a86d979
Co-authored-by: Heaust Azure <heaust.azure@gmail.com>
2026-09-18 07:14:37 +00:00
Ampandheaust f324ff6bb2 handles: one interleaved vertices/indices arena
Geometry becomes a span in a single 80-byte interleaved `vertices` array
plus a span in a single `indices` array, allocated through the core's
span allocator instead of one arena array per mesh attribute. Draws are
always indexed with baseVertex; face-material overrides emit one draw per
run of same-material faces rather than one per face. Vertex data is
validated as a whole before any span is touched, same-geometry mutations
are serialised, and render signal 4 is held across a span change until
the graph reflecting it is published. Sibling arrays grow to the primary
array's capacity in one batch; reserve() takes vertex/index totals.

Amp-Thread-ID: https://ampcode.com/threads/T-01a0b2dd-db32-712d-99ce-12ce4a86d979
Co-authored-by: Heaust Azure <heaust.azure@gmail.com>
2026-09-18 07:00:36 +00:00
Ampandheaust 31d3ae21ff core: manual buffer sync with sync-rows, reuse unchanged buffers and pipelines across activations
Also lets the screenshot suite point at another reference directory and settle longer.

Amp-Thread-ID: https://ampcode.com/threads/T-01a0b2dd-db32-712d-99ce-12ce4a86d979
Co-authored-by: Heaust Azure <heaust.azure@gmail.com>
2026-09-18 05:53:44 +00:00
Ampandheaust 49c80a209e core: indirect draw passes, device feature list, drop handles-specific bundle heuristics
Amp-Thread-ID: https://ampcode.com/threads/T-01a0b2dd-db32-712d-99ce-12ce4a86d979
Co-authored-by: Heaust Azure <heaust.azure@gmail.com>
2026-09-18 05:31:11 +00:00
Ampandheaust 6153223fd5 Add FxNode render graph playground
Add persistent render graph loadouts, grouped node authoring, code-appending apply behavior, explicit playground lighting, and resilient worker startup.

Amp-Thread-ID: https://ampcode.com/threads/T-01a033d4-0e44-708e-af0e-6166283416ff
Co-authored-by: Heaust Azure <heaust.azure@gmail.com>
2026-08-24 17:45:48 +00:00
Ampandheaust 951449db0b Add local glTF viewer and visual coverage
Amp-Thread-ID: https://ampcode.com/threads/T-01a02d1c-27fc-7761-9f51-ed97b2dd4e3e
Co-authored-by: Heaust Azure <heaust.azure@gmail.com>
2026-08-24 10:51:26 +00:00