242,275 Commits over 3,867 Days - 2.61cph!
Bugfix: Don't send newly spawned trees to player if they land in a batch yet to be sent to them
- Added a profiling scope so we can track if it's taking too much time(it's a TxP complexity algo, but P tends to 0 very quickly, so we should be able to afford this)
Tests: Tested on 4.5k procgen world and connected from a separate client and chopped a bunch of trees saw no duplicates. That said, it's very difficult to proc this (<1s window).
Fix formatting of node titles with type params
https://files.facepunch.com/ziks/2024-10-07/NTc7SLTOVR.png
Add built-in nodes for Scene.GetAllComponents<T>()
https://files.facepunch.com/ziks/2024-10-07/SUbPeMLFFP.png
Added descriptions to Components that didn't have them. Most are self-explanatory, but should help reduce some confusion. Resolves Facepunch/sbox-issues#6560
brick_debris_a + debris_a vmat meta
Replace usage of "devshots_screenshot" with "screenshot"
Add "killserver" to concommand blacklist
Correct UTF8<->UTF16 conversion for close caption function & hook
Try to resolve issues caused by sv_max_userinfo_rate for extreme cases
Make DisplayInfo.Format() easier to use
Split out InputSettings into Input.config - backwards compatible
merge from physics_horses
Fixed string comparisons in SpriteComponent. Resolves Facepunch/sbox-spritetools#25
WorldPosition > Transform.Position ect..
▉▄▅█▄▄▆▄ ▍▄▅ ▊▆▋▇ ▉▋▉▌▅▆▅▄▍▋ █▌ ▅▇▄▋█▋ ▄▋▇▅▅▇, ▊█ ▋▉▌▉▄▉▇▌▆ ▅▍▉▆ ▆█▄▅▊▉
Add experimental r_tiled_rendering_conservative_culling so people can play with it, uses near-z approach for tiled culling for even better perfomance, doesn't play well with translucents, will be deprecated when we have two sets of light lists
Put counter text changes on the correct branch
Clean: forgot to save an extra comment
Tests: none, trivial change
Update: replacing cell budget with a time-per-player budget
- Setting this budget to 10micros initially
This budget doesn't evaluate every cell, but rather groupings of cells. I'll re-evaluate this once I get to testing big procedural worlds.
Tests: Confirmed entire Craggy grid gets streamed to the player
Update: added a runtime switch to disable tree batch streaming
- Enabled by default
Tests: booted with it being both turned on and off
Merge: from main
Tests: none
Don't do OnDisabled(); OnEnabled() on Skybox2D dirty but clean up probe cleanly if needed
Leaderboard backup, run #
14477
Fixed a bunch of props in radtown that were either missing LOD components entirely or using LODGroups
S2P and HLOD generation
Fixed some placeholder stuff for gesture slots that are empty
Fixed radial menu defaulting to the text of the last opened radial menu if opening a new menu that has no enabled options (bug is extremely old, very obvious when switching between gesture wheels)
▋▆▄▊▅▋█ ▉▇▆▉█▇ ▄▆ █▆█ ▉▄█▍▌██▅▊ ▄▋▉▍▄█, ▊▆ ▌▇▆▅▆▉▅ ▅▇▋▊▅▅ ▉▉ ▊▅'▄ ▉▊▌ █▉ ▅▄▄▇▌▌▆█
- ▍▊▌█▄ ▍ ▇▆▉▇▋█▋▌▍▉█▅ ▍▌▍▍▍ ▇▇▋ ▅▅▆▅▆▇▄▆▊ █▇ ▅█▊▋▌ ▄▇▄▄▌▆ ▄▉▍▊▌█▉▄▊█▄▆
- ▌▆▄█▄█▇ █▅▉▆▇▋▇▄▅▍▊ ▇▇▍▌▄▊▆▊ █▊▋▊▋▌ ▄▅ ▋▉▋ ▉▆ ▉▊█▅▉ ▇▉▍▉ █▇▅▄▅▋▇▄ ▅▇▆█▄
- ▋▋▉▋▇▆▉▊ ██▇█▋ ▊█ ▋▉▇ ▊▋▌▆▇█▋
▆▍▇█▋ ▌▄▄ ▆▌▋ ▊▊▉▉▌▌▌█▊ ▇▊ ▊▄██▌██▋▉▋.▄▊▌▋▉▉▍▊▋▌▉▆▅▋▋▅█▇, █▊▇▇ █▉ ▇▆ ▄▆▄▉
Scaling should not be applied to NormalToLocal and NormalToWorld but happy to be proven wrong
▆▇▍▄▌▄▅ ▄▄▅▇▅▆█▅▉▍▊█▍▄▅▇▇▆▅▄▆▌▅▌▆▆ ▊▅ ▊▌▆▊▉▆▆█▄▅ ▄▅▄▆▅▆▋▌▍▄▄▄▉▋ ▊▄▍▅▌▆
▄▌ ▊▆ █▊▋ ▍▌▋▋ ▋▆ ▆▅▄▍ ▍ ▋▌▄▇ ▇▉▅▊▇█▄▉ ▇▅ ▉▋▄ ▅█▋▌██ ▋▄▆▆ ▆▌▆▉▉▄▍▇ ▄▆▋ ▉▅ ▅▌▇▇▍▇, █▊▆▉ ▅▄▊▆▄ ▇▇▆ ▇▄▆ ▌▋ ▄▌█▉, ▇▋ ▉▉▇▊ ▆▊ ▍▅▅▌▊▌▊ ▄▇ ▅▉ ▍▌▍ ▄▊▍▇█▍ ▆▄ ▆▉█ ▌▅▍ ▊▍ ▌▇▇ ▉▅▅ ▆█▊▆▊▌ █▍▆▉▋▇▊▍. ▊▍▆▇█▌▌▉▇ ▌▋▋▇ ▇▇▅▉▆▋ ▍▇█▊▅▇▇▊▇▇ ▍▄▋▆ ▋▇▌ ▅▄▅▇▇▋ ▅▌█ ▌█ ▅▊▄▄█▊▋
▉▅▇▄ ▍▄▆▉▆▇ ▌▄█▊▍█▄█▅█▉, ▉▇▌ ▋█▌█▋▆ ▌▉▄ █▅▅▉▅▄▄▄ ▋▉ ▋▄█ █▌▅▅▄▊ ▇▅▅▍▌█ ▇▉▅ ▋▋▍▉ ▆▍ ▍█▋▍▍ ▇▄ ▇▌▄▉▌▆▄ ▄▇▊ ▆▆▆▌▇▆▇▄▌▉ ▍▇▍▉ ▆▋███▇▅▅
▆▇▌▇ ▌▇▊▄█ ▅█▅▋▆▄ ▅▊▆█-▍▄▋▄ ▄▌▉▇▄▋▊ ▉▍█▄▍▍▄ ▍▇▄█▋▄ ▌▍▌▋▍▋▇▋▆▍▆
Fixed transformation order for PointToLocal and PointToWorld, add some tests to make sure they match ToLocal and ToWorld
Revert WaterSystem.cs to the version from main, should make merges easier. Looks like it was missing a bunch of stuff due to various botched merges over the years
Added demo.saveShotFolderAsAnim, saves all shots of the current demo as anim files
Fixed large stables vendor getting killed on server restart
Leaderboard backup, run #
14470
Test tl, tr, bl and br corners instead of updownleftright for cone frustum culling
Tiled Frustum Culling (#1669)
SkyBox2D Indirect Lighting ( Reflections/IBL ) can be toggled, hammer maps don't expect skyboxes to be used for lighting since it has vrad
Be more conservative by testing 4 points of cone base rather than a sphere shape
Tiled Frustums
Start moving tiled_light_culling to use frusta
Iterate frustum tiled culling
Near/far tiled frustum planes
Very simple ConeIntersect and AABBIntersect
Cubemap frustum cullinng
Show cubemap complexity on tiled debug vis as well
https://files.facepunch.com/sampavlovic/1b2911b1/HM0UBknyYv.png
Set up things to use worldSpaceNear to fix orthographic rendering with tiled rendering, use depth min/max for building tiled frustums
https://imgur.com/a/tybd5Dv
Fix ortho projection for CalculateScreenTiles
fix DepthMinMax on ortho projections, fixes tiled lights on ortho projections
https://files.facepunch.com/sampavlovic/1b3011b1/ZiyxdtwHc1.png
Delete tiled_frustums_cs, main shader does everything, we recalculate it every frame anyway since we also match the depth chain for the far plane, can store it later again from it if we need it for anything
Transform plane directly to local instead of using inverse matrix
Light.GetInvertedRadius() not used anywhere, let's make it Light.GetRadius()
Push radius changes to lightbinner.hlsl
Fix frustum calculation so it wouldnt mess up with low farz, fix FetchDepthMinMax
Using invprojrow3 to linearize depth feels a bit incorrect specially when moving far-z use correct algorithm, can be accelerated later
Depth Near-Far was inverted on tiled_culling, perfect for both near/far planes now
Allow using conservative near-z culling with D_CONSERVATIVE_CULLING and remove elegibility of D_DEPTH_PREPASS, we always have it
https://i.imgur.com/9UlrmTc.png
https://i.imgur.com/joEz72S.png
Allow using conservative near-z culling with D_CONSERVATIVE_CULLING and remove elegibility of D_DEPTH_PREPASS, we always have it
https://i.imgur.com/9UlrmTc.png
https://i.imgur.com/joEz72S.png
Using invprojrow3 to linearize depth feels a bit incorrect specially when moving far-z use correct algorithm, can be accelerated later
Depth Near-Far was inverted on tiled_culling, perfect for both near/far planes now
Fix frustum calculation so it wouldnt mess up with low farz, fix FetchDepthMinMax
Push radius changes to lightbinner.hlsl
