fix(deferred): llf visualization debug mode

[alandtse]

Summary by CodeRabbit

• Bug Fixes
  • Corrected albedo visualization output in lighting and debug visualization modes across rendering pipelines.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

📝 Walkthrough

Walkthrough

The changes introduce local variables (outputAlbedo and albedo) that capture diffuse color values within debug visualization paths (light limit fix and lights visualization branches) across three shader files, then propagate these values to the albedo output in conditional branches.

Changes

Cohort / File(s)

Summary

Debug Albedo Visualization package/Shaders/Effect.hlsl, package/Shaders/Lighting.hlsl, package/Shaders/RunGrass.hlsl

Modified shader files to assign diffuse color to albedo variables within debug visualization conditional blocks (light limit fix and lights visualization). Effect.hlsl introduces outputAlbedo variable and propagates it to psout.Albedo in all shader paths; Lighting.hlsl and RunGrass.hlsl assign diffuse/color values to outputAlbedo/albedo within LIGHT_LIMIT_FIX and LightLimitFix branches respectively.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Possibly related PRs

• fix: fix effect blending #1490 — Modifies shader Albedo output declaration and assignment in Effect.hlsl using similar outputAlbedo variable pattern
• fix(pbr): fix inconsistencies #1526 — Updates albedo propagation across Lighting.hlsl and RunGrass.hlsl with comparable debug visualization changes
• build(hlsl): fix warnings #1370 — Modifies pixel shader output handling in Effect.hlsl for psout initialization and albedo assignment

Suggested reviewers

• jiayev
• doodlum

Poem

🐰 Through shader paths so bright and clear,
Albedo whispers, "I am here!"
From diffuse light to debug's glow,
The colors dance and brightly show,
Three files blend in visualization's cheer! ✨

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and specifically identifies the fix as addressing a light limit fix (llf) visualization debug mode, which aligns with the changes made across three shader files to correct albedo output in debug visualization paths.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[github-actions]

Using provided base ref: 44eb788
Using base ref: 44eb788
Base commit date: 2026-01-19T23:55:15Z (Monday, January 19, 2026 11:55 PM)
No actionable suggestions for changed features.

[Copilot]

Pull request overview

This PR fixes the Light Limit Fix (LLF) debug visualization mode to correctly display in the deferred rendering albedo buffer. When LLF debug visualization is enabled, the debug colors (showing light counts via color mapping) should appear in both the diffuse output and the deferred albedo buffer.

Changes:

• Modified three shader files to ensure debug visualization colors are written to the albedo buffer in deferred rendering mode
• RunGrass.hlsl: Sets albedo variable to debug visualization color
• Lighting.hlsl: Sets outputAlbedo to debug visualization color
• Effect.hlsl: Introduces outputAlbedo variable and uses it for deferred albedo output

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
package/Shaders/RunGrass.hlsl	Sets albedo to debug visualization color when LLF debug mode is active
package/Shaders/Lighting.hlsl	Sets outputAlbedo to visualization color when debug mode is active, used in deferred albedo buffer
package/Shaders/Effect.hlsl	Introduces outputAlbedo variable, sets it for debug visualization, and uses it in deferred albedo output

[Copilot]

When debug visualization is enabled, psout.Diffuse is never assigned. The else branch (line 829) assigns psout.Diffuse = float4(diffuseColor, 1), but when the visualization is enabled, only albedo is set. This leaves psout.Diffuse uninitialized, which will cause undefined behavior. Add psout.Diffuse = float4(diffuseColor, 1); before this line to match the pattern used in Effect.hlsl (line 854 sets psout.Diffuse, then line 865 sets outputAlbedo) and Lighting.hlsl (where psout.Diffuse.xyz is always set before line 3149).

Suggested change

	}
	}
	psout.Diffuse = float4(diffuseColor, 1);

[coderabbitai]

Actionable comments posted: 0

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

package/Shaders/Effect.hlsl (1)

855-889: Restore non-debug albedo behavior for MULTBLEND.

outputAlbedo is only set inside the LLF debug path, so deferred albedo becomes zero in normal rendering, including MULTBLEND/MULTBLEND_DECAL paths that previously carried psout.Diffuse. Initialize outputAlbedo to the old behavior and keep the debug override.

✅ Proposed fix

-	float3 outputAlbedo = 0;
+	float3 outputAlbedo = 0;
+#if defined(MULTBLEND) || defined(MULTBLEND_DECAL)
+	outputAlbedo = psout.Diffuse.xyz;
+#endif
 #	if defined(LIGHTING) && defined(LIGHT_LIMIT_FIX) && defined(LLFDEBUG)
 	if (SharedData::lightLimitFixSettings.EnableLightsVisualisation) {
 		if (SharedData::lightLimitFixSettings.LightsVisualisationMode == 0) {
 			psout.Diffuse.xyz = LightLimitFix::TurboColormap(0.0);
 		} else if (SharedData::lightLimitFixSettings.LightsVisualisationMode == 1) {
 			psout.Diffuse.xyz = LightLimitFix::TurboColormap(0.0);
 		} else {
 			psout.Diffuse.xyz = LightLimitFix::TurboColormap((float)lightCount / MAX_CLUSTER_LIGHTS);
 		}
 		outputAlbedo = psout.Diffuse.xyz;
 	}
 #	endif

[alandtse]

Taking to draft as looking at another PR LLF may be working?

[github-actions]

✅ A pre-release build is available for this PR:
Download

[alandtse]

Must be crazy. Closing as dev apparently works.