Thaddeus Crews
3c1e479290
Merge pull request #110077 from clayjohn/mobile-glow
...
Overhaul and optimize Glow in the mobile renderer
2025-10-31 09:23:32 -05:00
clayjohn
2e59cb41f4
Optimize glow and tonemap gather step in the mobile renderer
...
Mobile devices are typically bandwidth bound which means we need to do as few texture samples as possible.
They typically use TBDR GPUs which means that all rendering takes place on special optimized tiles. As a side effect, reading back memory from tile to VRAM is really slow, especially on Mali devices.
This commit uses a technique where you do a small blur while downsampling, and then another small blur while upsampling to get really high quality glow. While this doesn't reduce the renderpass count very much, it does reduce the texture read bandwidth by almost 10 times. Overall glow was more texture-read bound than memory write, bound, so this was a huge win.
A side effect of this new technique is that we can gather the glow as we upsample instead of gathering the glow in the final tonemap pass. Doing so allows us to significantly reduce the cost of the tonemap pass as well.
2025-10-30 21:56:26 -07:00
David Snopek
ca0eb5da24
OpenXR: Fix resizing viewports used by OpenXRCompositionLayer
2025-10-30 17:40:52 -05:00
Stuart Carnie
230adb7511
Add Persistent Buffers
...
This work is a heavily refactored and rewritten from TheForge's initial
code.
TheForge's original code had too many race conditions and was
fundamentally flawed as it was too easy to incur into those data races
by accident.
However they identified the proper places that needed changes, and the
idea was sound. I used their work as a blueprint to design this work.
This PR implements:
- Introduction of UMA buffers used by a few buffers
(most notably the ones filled by _fill_instance_data).
Ironically this change seems to positively affect PC more than it does
on Mobile.
Updates D3D12 Memory Allocator to get GPU_UPLOAD heap support.
Metal implementation by Stuart Carnie.
Co-authored-by: Stuart Carnie <stuart.carnie@gmail.com >
Co-authored-by: TheForge team
2025-10-24 08:16:19 +11:00
clayjohn
f61ee7bdf6
Use half float precision buffer for 3D when HDR2D is enabled
...
This is necessary for Environment effects like Glow to work correctly.
2025-10-21 13:44:46 -07:00
Dario
ba268416d5
Push pipeline compilation of various effects to the worker thread pool.
2025-10-13 12:00:23 -03:00
Thaddeus Crews
8150cb9200
Merge pull request #111403 from AeioMuch/fix_double_precision_wrong_indexes
...
Fix wrong indices used for transform & UBO matrix for double precision build
2025-10-10 10:25:59 -05:00
AeioMuch
dae2122388
Fix wrong indexes for double precision
2025-10-09 22:01:41 +02:00
Kaleb Reid
f2d0ea6d40
Ensure reflection atlas is valid before rendering
2025-10-07 22:05:22 -07:00
Thaddeus Crews
fdc235f496
Merge pull request #110884 from Kaleb-Reid/fix-spotlight-aabb
...
Use correct AABB for SpotLight3Ds when `spot_angle > 90`
2025-10-06 14:34:05 -05:00
Kaleb Reid
91167c3c23
Use correct AABB for SpotLight3Ds when spot_angle > 90
2025-10-04 02:12:51 -07:00
Thaddeus Crews
2b7f39ea28
Merge pull request #110505 from Calinou/tweak-draw-command-labels
...
Tweak draw command label names for consistency
2025-09-30 20:10:37 -05:00
Aaron Franke
9fbf5808a0
Rename server "free" functions to "free_rid" to match exposed API
2025-09-30 16:52:25 -07:00
Thaddeus Crews
f9694033a5
Merge pull request #107923 from clayjohn/RD-mat3x4
...
Use Mat3x4 for model and view transforms to save bandwidth and ALUs
2025-09-30 18:34:59 -05:00
clayjohn
aa8bc0b56d
Prompt editor restart when reflection probe size is updated
...
Also formally deprecate the RS function for updating an individual probes size. The functionality was removed in 4.0, but the function itself was mistakenly left exposed.
2025-09-29 18:35:34 -07:00
clayjohn
14b60f2264
Optimize vertex shader using mat3x4 to reduce bandwidth, load/store operations and ALUs
2025-09-26 23:20:08 -07:00
Thaddeus Crews
7b8073d122
Merge pull request #107763 from Ivorforce/missing-vpv
...
Add missing `vpv.push_back(pv);` in `render_target_get_sdf_texture`.
2025-09-24 09:59:11 -05:00
Thaddeus Crews
cd3a6c88fd
Merge pull request #106200 from BlueCube3310/image-16-u16
...
Image: Implement 16-bit unorm and uint formats
2025-09-23 12:08:46 -05:00
Thaddeus Crews
e721af5775
Merge pull request #110330 from DarioSamo/reflection-probe-size-leak
...
Fix reflection probes not recreating downsampled textures when mode changes.
2025-09-17 11:34:12 -05:00
Thaddeus Crews
853884ba05
Merge pull request #109143 from sievaxx/more_sampler_default_texs
...
Add and enable default textures for other samplers
2025-09-16 11:44:51 -05:00
Hugo Locurcio
204e310914
Tweak draw command label names for consistency
...
- Use Title Case for all labels, and add hyphens where relevant.
- Mention Roughness in SSR Filter's label name, as it's only enabled
when the SSR roughness quality is not set to Disabled.
2025-09-14 01:23:47 +02:00
Dario
7b923609c2
Fix reflection probes not recreating downsampled textures when mode changes.
2025-09-08 15:54:37 -03:00
clayjohn
b0d7a241b2
Handle the case where VRS is a two byte per pixel format when creating default VRS texture
2025-08-26 12:22:23 -07:00
Pedro J. Estébanez
919006eb35
Fix material removal clearing all instances of shared texture arrays
2025-08-18 12:21:15 +02:00
Stuart Carnie
c8b5982d0f
Metal: Ensure correct output texture format selection
2025-08-08 12:50:16 +10:00
Riley Willows
fbe2710843
Add and enable default textures for other samplers
2025-07-31 00:08:43 +01:00
Thaddeus Crews
6bc7fcb7af
Merge pull request #108567 from beicause/multimesh-null-checks
...
Add some multimesh null checks to avoid crash
2025-07-14 10:30:35 -05:00
Thaddeus Crews
35a2d7843e
Merge pull request #104302 from brennennen/reflection_probe_count_error_check
...
Add error check for reflection probe invalid atlas index.
2025-07-14 10:30:31 -05:00
LuoZhihao
5352638f95
Add some multimesh null checks to avoid crash
2025-07-13 10:09:36 +08:00
brennen
2583aa4a68
Add error check for reflection probe invalid atlas index.
2025-07-10 19:03:21 -05:00
Allen Pestaluky
a1591512f8
Always perform color correction and debanding on nonlinear sRGB values.
...
Fixes #107730
Co-authored-by: LuoZhihao <luo_zhihao@outlook.com >
2025-07-08 13:44:53 -04:00
Thaddeus Crews
e750950dc6
Merge pull request #107568 from Rudolph-B/Issue-92708
...
Fix floating point precision errors when setting particle trail length
2025-07-01 17:57:58 -05:00
BlueCube3310
16b9ee6f50
Image: Implement 16-bit unorm and uint formats
2025-06-28 00:30:52 +02:00
Mingxi Zhang
de2aa1bc4d
Fix: Ensure sky orientation is set when reflection uses sky
2025-06-22 23:50:39 +08:00
Lukas Tenbrink
28fd557c58
Add missing vpv.push_back(pv); in render_target_get_sdf_texture.
2025-06-20 14:46:03 +02:00
Thaddeus Crews
c9511255de
Merge pull request #107684 from dsnopek/revert-shadow-normal-bias-relative
...
Revert "Use lower shadow normal bias for distant directional shadow splits"
2025-06-19 09:53:54 -05:00
David Snopek
4cc1d0ba67
Revert "Use lower shadow normal bias for distant directional shadow splits"
...
This reverts commit dfc05caf06
2025-06-18 18:14:22 -05:00
Lukas Tenbrink
2b36c79f7b
Use Span<uint8_t> in RenderingDevice allocation APIs to avoid intermediary arrays on calls.
2025-06-18 12:31:48 +02:00
Rudolph Bester
3c5d4a2410
Fix floating point precision errors when setting particle trail length
2025-06-15 20:38:50 +02:00
LuoZhihao
7419e4429d
Fix global shader texture uniform
2025-06-13 13:24:38 +08:00
Thaddeus Crews
fc523ec5f6
Merge pull request #107230 from dsnopek/rd-depth-image-swizzle
...
RendererRD: Fix swizzle on depth formats
2025-06-09 17:08:49 -05:00
David Snopek
316b9aa6d2
RendererRD: Fix swizzle on depth formats
2025-06-06 16:30:17 -05:00
David Snopek
fa30719814
Fix wrong default texture for global uniforms of type sampler2DArray
2025-06-05 11:12:03 -05:00
Rémi Verschelde
37982d4a01
Merge pull request #107116 from beicause/expose-mesh-surface-update-index-buffer
...
Expose `RS.mesh_surface_update_index_region`
2025-06-05 17:24:54 +02:00
LuoZhihao
63c124fa45
Expose RS.mesh_surface_update_index_region
...
And `mesh_surface_get_format_index_stride`
2025-06-05 19:35:59 +08:00
Rémi Verschelde
088ef7b6f6
Merge pull request #100283 from devloglogan/vk-motion-vectors
...
Implement motion vectors in mobile renderer
2025-06-05 13:10:20 +02:00
Thaddeus Crews
7a3d2f6779
Merge pull request #106606 from aaronfranke/pow2_64
...
Add 64-bit versions of core power of 2 functions
2025-06-02 18:51:56 -05:00
Thaddeus Crews
167a9e33d3
Merge pull request #106880 from dsnopek/meta-depth-api-nonhacks
...
Minor rendering and XR changes to allow Meta enivornment depth API to work entirely from GDExtension
2025-06-02 18:51:19 -05:00
Aaron Franke
f6f1df7d73
Add 64-bit versions of core power of 2 functions
2025-06-01 23:11:12 -07:00
devloglogan
3b70fbdc3c
Implement motion vectors in mobile renderer
2025-05-30 13:09:54 -05:00
