NullReferenceException - Camera Selection

[justinwasilenko]

Hi there.

Was hoping to give this a try for a project using Unity 6, and especially VR as the default URP Lens Flare system is not rendering in VR. So I am looking for a more performant Lens Flare system that I could use. I stumbled on your project, but checking our your example scene I get this error when selecting the camera with the MFLensFlareSimple component.

[Exception] NullReferenceException: Object reference not set to an instance of an object
MFLensflareEditor.OnInspectorGUI() at /MoonFlowLensFlare/MFLensFlare.cs:326
324: public override void OnInspectorGUI()
325: {
--> 326: EditorGUILayout.LabelField("Light Count: ", _target.FlareDict.Count.ToString());
327: base.OnInspectorGUI();
328: }

InspectorElement+<>c__DisplayClass79_0.b__0() at :0

GUIUtility.ProcessEvent()

Separate issue is that it only renders in one eye in VR.

[Reguluz]

I'm so sorry that I can't support VR render pipeline, because I don't have any VR device for debugging.
The solution only aimed to equip during the time that Unity didn't update their old lens flare system to URP yet.
but the error console maybe can be fixed.

[Reguluz]

I've checked the project. MFLensFlareSimple is deprecated long time before. Sorry that I forgot to update the release pack. Please download the 3.1 version pack or download the source code directly.
And again, for I do not have any VR device till now, I cannot debug this function on VR at the moment. So I can't promise when to solve the problem on VR.

[justinwasilenko]

Thank you for the response. I started working on it because I don't think Unitys Lens Flare system as a post processing effect will work performance wise on the Meta Quest.

Here is an updated MFLensflare.shader that will render in both eyes in VR. The one thing that it still needs is to be able to control the depth of the effects to give it that full 3d look. I'm thinking in the FlareAsset where we are storing all the values, add a depth variable for each part of the effect, on how far it renders away from the source.

Shader "Moonflow/Lensflare"
{
    Properties
    {
        _MainTex ("Texture", 2D) = "white" {}
    }
    SubShader
    {
        Tags
        {
            "RenderPipeline"="UniversalPipeline"
            "RenderType"="Transparent"
            "Queue"="Transparent"
            "IgnoreProjector"="True"
        }

        Pass
        {
            Blend One One
            ZWrite Off
            ZTest Always
            Cull Off

            HLSLPROGRAM
            #pragma vertex vert
            #pragma fragment frag
            #pragma multi_compile_instancing
            #pragma multi_compile _ _STEREO_MULTIVIEW_ON _STEREO_INSTANCING_ON
            #pragma multi_compile_fragment _ _DEBUG_DISPLAY
            #include "Packages/com.unity.render-pipelines.universal/ShaderLibrary/Core.hlsl"

            struct appdata
            {
                float4 vertex : POSITION;
                float2 uv     : TEXCOORD0;
                half4  color  : COLOR;
                UNITY_VERTEX_INPUT_INSTANCE_ID
            };

            struct v2f
            {
                float4 vertex : SV_POSITION;
                float2 uv     : TEXCOORD0;
                half4  color  : COLOR;
                UNITY_VERTEX_OUTPUT_STEREO
            };

            TEXTURE2D(_MainTex);
            SAMPLER(sampler_MainTex);

            v2f vert(appdata v)
            {
                v2f o;
                UNITY_SETUP_INSTANCE_ID(v);
                UNITY_INITIALIZE_VERTEX_OUTPUT_STEREO(o);

                float3 worldPos = TransformObjectToWorld(v.vertex.xyz);

                // ✅ Use per-eye view-projection matrix (Core.hlsl defines this)
                float4x4 stereoVP = UNITY_MATRIX_VP;
                #if defined(USING_STEREO_MATRICES)
                    stereoVP = unity_StereoMatrixVP[unity_StereoEyeIndex];
                #endif

                o.vertex = mul(stereoVP, float4(worldPos, 1.0));
                o.uv     = v.uv;
                o.color  = v.color;
                return o;
            }

            half4 frag(v2f i) : SV_Target
            {
                UNITY_SETUP_STEREO_EYE_INDEX_POST_VERTEX(i);
                return SAMPLE_TEXTURE2D(_MainTex, sampler_MainTex, i.uv) * i.color;
            }
            ENDHLSL
        }
    }
}

[Reguluz]

Thank you for the response. I started working on it because I don't think Unitys Lens Flare system as a post processing effect will work performance wise on the Meta Quest.感谢您的回复。我开始研究它，因为我认为 Unity 的镜头光晕系统作为后期处理效果不会在 Meta Quest 上发挥性能作用。

Here is an updated MFLensflare.shader that will render in both eyes in VR. The one thing that it still needs is to be able to control the depth of the effects to give it that full 3d look. I'm thinking in the FlareAsset where we are storing all the values, add a depth variable for each part of the effect, on how far it renders away from the source.下面是一个更新的 MFLensflare.shader，它将在 VR 中双眼渲染。它仍然需要的一件事是能够控制效果的深度，使其具有完整的 3D 外观。我正在考虑在存储所有值的 FlareAsset 中，为效果的每个部分添加一个深度变量，以确定它渲染与源的距离。

Shader "Moonflow/Lensflare"
{
    Properties
    {
        _MainTex ("Texture", 2D) = "white" {}
    }
    SubShader
    {
        Tags
        {
            "RenderPipeline"="UniversalPipeline"
            "RenderType"="Transparent"
            "Queue"="Transparent"
            "IgnoreProjector"="True"
        }

        Pass
        {
            Blend One One
            ZWrite Off
            ZTest Always
            Cull Off

            HLSLPROGRAM
            #pragma vertex vert
            #pragma fragment frag
            #pragma multi_compile_instancing
            #pragma multi_compile _ _STEREO_MULTIVIEW_ON _STEREO_INSTANCING_ON
            #pragma multi_compile_fragment _ _DEBUG_DISPLAY
            #include "Packages/com.unity.render-pipelines.universal/ShaderLibrary/Core.hlsl"

            struct appdata
            {
                float4 vertex : POSITION;
                float2 uv     : TEXCOORD0;
                half4  color  : COLOR;
                UNITY_VERTEX_INPUT_INSTANCE_ID
            };

            struct v2f
            {
                float4 vertex : SV_POSITION;
                float2 uv     : TEXCOORD0;
                half4  color  : COLOR;
                UNITY_VERTEX_OUTPUT_STEREO
            };

            TEXTURE2D(_MainTex);
            SAMPLER(sampler_MainTex);

            v2f vert(appdata v)
            {
                v2f o;
                UNITY_SETUP_INSTANCE_ID(v);
                UNITY_INITIALIZE_VERTEX_OUTPUT_STEREO(o);

                float3 worldPos = TransformObjectToWorld(v.vertex.xyz);

                // ✅ Use per-eye view-projection matrix (Core.hlsl defines this)
                float4x4 stereoVP = UNITY_MATRIX_VP;
                #if defined(USING_STEREO_MATRICES)
                    stereoVP = unity_StereoMatrixVP[unity_StereoEyeIndex];
                #endif

                o.vertex = mul(stereoVP, float4(worldPos, 1.0));
                o.uv     = v.uv;
                o.color  = v.color;
                return o;
            }

            half4 frag(v2f i) : SV_Target
            {
                UNITY_SETUP_STEREO_EYE_INDEX_POST_VERTEX(i);
                return SAMPLE_TEXTURE2D(_MainTex, sampler_MainTex, i.uv) * i.color;
            }
            ENDHLSL
        }
    }
}


Thanks for your solution, I'll update code later on.
Turn to the question of depth, do you mean that every flare mesh need their independent Z depth? I'm not sure if I understand your requirement correctly