5b5dc00c52
2D: Remove redundant assignment
2024-09-17 06:32:49 +10:00
8c6210a3eb
Merge pull request #92290 from clayjohn/LOD-fixes
...
Use distance to AABB surface to calculate Mesh LOD instead of using supports
2024-09-16 13:34:04 +02:00
c113e5b143
2D: Fix use-after-free in batch rendering
...
Closes #96960
Fixes regression of #95574 using fix from #95666
2024-09-14 07:50:39 +10:00
7ff106f828
Use distance to AABB surface to calculate Mesh LOD instead of using supports
2024-09-11 14:17:09 -07:00
a657ea42f1
2D: Add batching to RendererCanvasRenderRD
2024-09-12 05:26:06 +10:00
bc4c60c933
Merge pull request #94893 from rune-scape/no-const-cast-mesh-storage
...
Avoid `const_cast` in `mesh_storage.h`
2024-09-10 09:51:22 +02:00
af92fdb0ac
Add basic support to evaluate operator value in shader language
2024-09-09 15:12:11 +03:00
aa07333abe
Merge pull request #96114 from BlueCube3310/sh-lightmap-packing
...
LightmapGI: Pack L1 SH coefficients for directional lightmaps
2024-09-06 22:38:29 +02:00
a89f4fa5a9
LightmapGI: Pack L1 SH coefficients for directional lightmaps
2024-09-05 22:46:58 +02:00
82d7531a2e
Merge pull request #95990 from Breush/76166-sky-auto-radiance
...
Sky: Adapt radiance size if `AUTOMATIC_MODE` resolves to `REALTIME`
2024-09-04 17:12:02 +02:00
44e526d3d5
Sky: No more auto-selecting REALTIME mode if radiance is not 256
2024-09-04 10:54:08 +02:00
49ed6c558c
Merge pull request #86809 from clayjohn/TAA-disocclusion
...
Tune TAA disocclusion scale to avoid rejecting all samples during motion.
2024-09-03 16:13:26 +02:00
d1c27f239d
Merge pull request #96486 from ComycSans/black_sky
...
Fix `SkyRD::Sky::free()` freeing shared material
2024-09-03 11:44:05 +02:00
63a8548693
Merge pull request #96426 from clayjohn/RD-reflection-probe-roughness
...
Use non-linear mapping for ReflectionProbe mip levels to match sky
2024-09-03 11:43:52 +02:00
a2b64ed544
Merge pull request #95961 from RadiantUwU/fix-mesh-recursion
...
Fix shadow mesh recursion.
2024-09-03 11:43:16 +02:00
667778cf4d
Merge pull request #93448 from clayjohn/transmittance-fixes
...
Various fixes for transmittance effect
2024-09-03 11:42:57 +02:00
09e59fbec5
Fix SkyRD::Sky::free() freeing material
2024-09-02 18:59:31 +02:00
f546bf6aea
Merge pull request #96431 from viksl/fix-alpha-reflections-cull-order
...
Fix alpha material's reverse cull order.
2024-09-02 12:13:57 +02:00
f50ead48b0
Merge pull request #96128 from BlueCube3310/vram-profiler-texture-mem
...
Fix incorrect parameters for layered textures in VRAM texture memory profiler
2024-09-02 12:13:11 +02:00
929c69bad8
Fixes alpha materials being rendered with a reverse cull which is visible in reflections from reflection probes.
2024-09-01 12:10:04 +02:00
e74bc3079a
Fix incorrect parameters for layered textures in Video RAM texture memory profiler
2024-09-01 11:49:30 +02:00
64f5443b14
Use non-linear mapping for ReflectionProbe mip levels to match sky
2024-09-01 01:40:42 -07:00
70860aafd8
Fix shadow mesh recursion.
2024-08-31 22:31:43 +03:00
f04a9bb630
Avoid const_cast in mesh_storage.h
2024-08-28 14:46:03 -07:00
346cbc7f1f
Add support for compiling with VS clang-cl toolset
2024-08-28 13:30:44 -05:00
f538376c3b
Add `CLIP_SPACE_FAR` built-in to spatial shader
2024-08-27 08:48:46 +03:00
68d188d521
Merge pull request #95888 from clayjohn/Lightmap-SH-coefficients
...
Use correct lightmap coefficients to ensure that the directional lightmap mode looks correct
2024-08-25 20:18:18 +02:00
568589c9d8
Merge pull request #90993 from darksylinc/matias-TheForge
...
Add debug utilities for Vulkan
2024-08-22 00:38:22 +02:00
0badddaccb
Merge pull request #95917 from BlueCube3310/lightmap-align-fix
...
LightmapGI: Fix shader data alignment after #89919
2024-08-22 00:10:53 +02:00
39b77ea04e
Merge pull request #95790 from aaronfranke/rect-aabb-support
...
Simplify Rect2/AABB `get_support` function
2024-08-22 00:10:39 +02:00
364f916f3f
Add debug utilities for Vulkan
...
Features:
- Debug-only tracking of objects by type. See
get_driver_allocs_by_object_type et al.
- Debug-only Breadcrumb info for debugging GPU crashes and device lost
- Performance report per frame from get_perf_report
- Some VMA calls had to be modified in order to insert the necessary
memory callbacks
Functionality marked as "debug-only" is only available in debug or dev
builds.
Misc fixes:
- Early break optimization in RenderingDevice::uniform_set_create
============================
The work was performed by collaboration of TheForge and Google. I am
merely splitting it up into smaller PRs and cleaning it up.
2024-08-21 23:48:08 +02:00
4aa145624e
LightmapGI: Fix shader data alignment after #89919
2024-08-21 21:37:13 +02:00
f4ccba7508
Use correct lightmap coefficients to ensure that the directional lightmap mode looks correct
...
Also remove the metallic option from directional lightmap as it is guaranteed to return negative numbers in many cases
2024-08-21 10:24:32 -07:00
2d0165574d
Add Metal support for macOS (arm64) and iOS
2024-08-20 12:11:06 +02:00
6a9ecdcf0c
Merge pull request #91818 from rburing/fti_multimesh
...
Physics interpolation: `MultiMesh`
2024-08-20 10:01:46 +02:00
8acd82f70e
Merge pull request #89919 from BlueCube3310/bicubic-lightmap
...
Implement bicubic sampling for lightmaps
2024-08-20 10:01:41 +02:00
333f0f910f
Merge pull request #86000 from jsjtxietian/add-missing-model_normal_matrix
...
Add `model_normal_matrix` for fragment shader
2024-08-20 10:01:36 +02:00
7db24a9ad5
Simplify and fix Rect2/AABB get_support function
2024-08-19 23:55:31 -07:00
e698351db2
Add model_normal_matrix for fragment shader
2024-08-20 12:39:29 +08:00
6bf64027b7
Merge pull request #94785 from Chaosus/shader_fix_samplers_order
...
Fix texture samplers to not being last in the property list
2024-08-19 16:05:21 +02:00
8b39d7f326
Merge pull request #95666 from kleonc/parallax2d_repeat_offsets_relative_to_source
...
Fix `Parallax2D` repeats being not relative to its transform
2024-08-19 14:34:21 +02:00
10b91ee950
Merge pull request #95705 from zeux/fsr-skin-relax
...
Relax motion vector updates to allow skipped frames for skeletons
2024-08-19 12:09:34 +02:00
9cc53a134e
Merge pull request #95682 from zeux/fsr2-fix-free
...
Fix double free in FSR2 destructor
2024-08-19 12:09:20 +02:00
c6400a8fe4
Merge pull request #95662 from clayjohn/GLES3-sky-fog
...
Add fixed fog to the sky in the Compatibility renderer
2024-08-19 12:09:10 +02:00
824a97120e
Merge pull request #92213 from clayjohn/ambient-disabled
...
Disable all sources of ambient light when `ambient_light_disabled` render mode is used
2024-08-19 12:08:31 +02:00
1bd8372813
Fix Parallax2D repeats being not relative to its transform
2024-08-19 11:43:16 +02:00
ef9bb1a207
Implement support for bicubic lightmap filtering
...
Co-authored-by: Calinou <hugo.locurcio@hugo.pro>
2024-08-19 09:52:09 +02:00
1728f80e7c
Fixed Timestep Interpolation: MultiMesh
...
Adds fixed timestep interpolation to multimeshes.
Co-authored-by: lawnjelly <lawnjelly@gmail.com>
2024-08-18 17:14:59 +02:00
578049b7b9
Add fixed fog to the sky in the Compatibility renderer
...
And apply luminance multiplier after fog in RD renderer
2024-08-17 23:33:26 -07:00
92f2bc70dd
Relax motion vector updates to allow skipped frames for skeletons
...
Before this change, a skeleton that was not updated every frame would
result in a difference of 2+ between last_change and frame index every
frame, which would disable the buffer rotation and set motion vectors to
zero. This results in significant visual artifacts for FSR2 that are
especially prominent on the characters that move together with the view
such as the main character in third person mode.
This is a significant problem for high refresh rate displays: at 120 Hz,
we are effectively guaranteed to skip skeleton updates every other frame
with skeleton update happening during physics processing, and the lack
of physics interpolation for skeletons. This happens by default in TPS
demo when FSR2 is enabled.
In other places where motion vectors are disabled, such as multi-mesh
and mesh rendering (where previous transform is updated), the logic
effectively allows for a single-frame gap in updates, because it
compares the frame where the update happened (which is the current frame
if updates are consistent) with the current frame, so the latency of 0
means "update just happened", but both multi-mesh and mesh transform
updates permit a latency of 1 as well.
Here, however, last_change is updated *after* the frame processing has
concluded, so a zero-latency update has a distance of 1. Allowing a
distance of 2 (latency 1) reduces the severity of the problem and aligns
the logic with transform updates.
Note that the problem will still happen when refresh rate is noticeably
higher than physics rate times 2. For example, it still happens at 240
Hz. However, a longer latency allowance is inconsistent with other
transforms and could lead to issues, so ideally long term physical
interpolation of skeleton transforms would completely solve this.
2024-08-17 11:10:41 -07:00
Stuart Carnie
Rémi Verschelde
clayjohn
Yuri Rubinsky
BlueCube3310
Alexis Breust
ComycSans
viksl
Radiant
rune-scape
Alula
Rémi Verschelde
Matias N. Goldberg
Aaron Franke
jsjtxietian
kleonc
Ricardo Buring
Arseny Kapoulkine