feat(endfield): update for version 1.4.4

src/games/endfield/ao/gtao-spatial-avg_0xA683B186.cs_5_0.hlsl

@@ -0,0 +1,43 @@
+// ---- Created with 3Dmigoto v1.4.1 on Tue Feb 10 15:11:49 2026
+Texture2D<float4> t0 : register(t0);
+
+SamplerState s0_s : register(s0);
+
+cbuffer cb0 : register(b0)
+{
+  float4 cb0[1];
+}
+
+
+#include "../shared.h"
+
+#define cmp -
+
+RWTexture2D<unorm float4> u0 : register(u0);
+
+[numthreads(8, 8, 1)]
+void main(uint3 vThreadID : SV_DispatchThreadID)
+{
+  float4 r0,r1;
+  r0.xy = (uint2)vThreadID.xy;
+  r0.xy = float2(0.5,0.5) + r0.xy;
+  r0.zw = r0.xy * cb0[0].zw + -cb0[0].zw;
+  r0.z = t0.SampleLevel(s0_s, r0.zw, 0).x;
+  r1.xy = cb0[0].zw * r0.xy;
+  r0.xy = r0.xy * cb0[0].zw + cb0[0].zw;
+  r0.x = t0.SampleLevel(s0_s, r0.xy, 0).x;
+  r0.y = t0.SampleLevel(s0_s, r1.xy, 0).x;
+  r1.xyzw = cb0[0].zwzw * float4(1,-1,-1,1) + r1.xyxy;
+  r0.y = r0.y + r0.z;
+  r0.z = t0.SampleLevel(s0_s, r1.xy, 0).x;
+  r0.w = t0.SampleLevel(s0_s, r1.zw, 0).x;
+  r0.y = r0.y + r0.z;
+  r0.y = r0.y + r0.w;
+  r0.x = r0.y + r0.x;
+  r0.x = 0.200000003 * r0.x;
+  if (shader_injection.disable_game_ao >= 0.5f) {
+    r0.x = 1.0f;
+  }
+  u0[vThreadID.xy] = r0.xxxx;
+  return;
+}

src/games/endfield/ao/gtao-temporal_0xDD16F0F8.cs_5_0.hlsl

@@ -0,0 +1,122 @@
+// ---- Created with 3Dmigoto v1.4.1 on Tue Feb 10 15:11:49 2026
+Texture2D<float4> t3 : register(t3);
+
+Texture2D<float4> t2 : register(t2);
+
+Texture2D<float4> t1 : register(t1);
+
+Texture2D<float4> t0 : register(t0);
+
+SamplerState s0_s : register(s0);
+
+cbuffer cb0 : register(b0)
+{
+  float4 cb0[5];
+}
+
+
+#include "../shared.h"
+
+#define cmp -
+
+RWTexture2D<float4> u0 : register(u0);
+
+[numthreads(8, 8, 1)]
+void main(uint3 vThreadID : SV_DispatchThreadID)
+{
+  float4 r0,r1,r2,r3,r4;
+  r0.xy = (uint2)vThreadID.xy;
+  r0.xy = float2(0.5,0.5) + r0.xy;
+  r0.zw = cb0[4].zw * r0.xy;
+  r1.xy = t2.SampleLevel(s0_s, r0.zw, 0).xy;
+  r1.zw = abs(r1.xy) * float2(2,2) + float2(-1,-1);
+  r1.zw = r1.zw * r1.zw;
+  r1.zw = r1.zw * r1.zw;
+  r1.xy = float2(-0.5,-0.5) + r1.xy;
+  r2.xy = cmp(float2(0,0) < r1.xy);
+  r1.xy = cmp(r1.xy < float2(0,0));
+  r1.xy = (int2)-r2.xy + (int2)r1.xy;
+  r1.xy = (int2)r1.xy;
+  r1.xy = r1.zw * r1.xy;
+  r1.zw = r0.xy * cb0[4].zw + -r1.xy;
+  if (IMPROVED_GTAO >= 0.5) {
+    // Manual bilinear interpolation for previous AO (reduces wobble/ghosting)
+    // Only interpolate AO (.x), keep depth (.y) point-sampled
+    float2 texelPos = r1.zw * cb0[4].xy - 0.5;
+    float2 f = frac(texelPos);
+    float2 base = (floor(texelPos) + 0.5) * cb0[4].zw;
+    float ao00 = t3.SampleLevel(s0_s, base, 0).x;
+    float ao10 = t3.SampleLevel(s0_s, base + float2(cb0[4].z, 0), 0).x;
+    float ao01 = t3.SampleLevel(s0_s, base + float2(0, cb0[4].w), 0).x;
+    float ao11 = t3.SampleLevel(s0_s, base + cb0[4].zw, 0).x;
+    r2.x = lerp(lerp(ao00, ao10, f.x), lerp(ao01, ao11, f.x), f.y);
+    r2.y = t3.SampleLevel(s0_s, r1.zw, 0).y; // depth: point-sampled
+  } else {
+    r2.xy = t3.SampleLevel(s0_s, r1.zw, 0).xy;
+  }
+  r2.z = t0.SampleLevel(s0_s, r0.zw, 0).x;
+  r2.z = 0.00999999978 * r2.z;
+  r3.y = min(1, r2.z);
+  r3.x = t1.SampleLevel(s0_s, r0.zw, 0).x;
+  r2.zw = cmp(r1.zw < float2(0,0));
+  r2.z = (int)r2.w | (int)r2.z;
+  r1.zw = cmp(float2(1,1) < r1.zw);
+  r1.z = (int)r1.w | (int)r1.z;
+  r1.z = (int)r1.z | (int)r2.z;
+  if (r1.z != 0) {
+    r3.z = 0;
+  } else {
+    r1.xy = cb0[4].xy * r1.xy;
+    r1.x = dot(r1.xy, r1.xy);
+    r1.x = sqrt(r1.x);
+    r1.x = cb0[2].z * -r1.x;
+    r1.x = 1.44269502 * r1.x;
+    r1.x = exp2(r1.x);
+    r1.x = 0.970000029 * r1.x;
+    r1.y = r3.y + -r2.y;
+    r1.y = -1442.69507 * abs(r1.y);
+    r1.y = exp2(r1.y);
+    r3.z = r1.x * r1.y;
+  }
+  r1.x = cmp(0.000000 == cb0[2].y);
+  if (r1.x == 0) {
+    r1.xy = r0.xy * cb0[4].zw + -cb0[4].zw;
+    r1.x = t1.SampleLevel(s0_s, r1.xy, 0).x;
+    r4.xyzw = cb0[4].zwzw * float4(0,-1,1,-1) + r0.zwzw;
+    r1.y = t1.SampleLevel(s0_s, r4.xy, 0).x;
+    r1.z = t1.SampleLevel(s0_s, r4.zw, 0).x;
+    r4.xyzw = cb0[4].zwzw * float4(-1,0,1,0) + r0.zwzw;
+    r1.w = t1.SampleLevel(s0_s, r4.xy, 0).x;
+    r2.y = t1.SampleLevel(s0_s, r4.zw, 0).x;
+    r4.xyzw = cb0[4].zwzw * float4(-1,1,0,1) + r0.zwzw;
+    r0.z = t1.SampleLevel(s0_s, r4.xy, 0).x;
+    r0.w = t1.SampleLevel(s0_s, r4.zw, 0).x;
+    r0.xy = r0.xy * cb0[4].zw + cb0[4].zw;
+    r0.x = t1.SampleLevel(s0_s, r0.xy, 0).x;
+    r0.y = min(r1.w, r3.x);
+    r2.z = min(r2.y, r0.w);
+    r0.y = min(r2.z, r0.y);
+    r0.y = min(r1.y, r0.y);
+    r2.z = min(r1.x, r1.z);
+    r2.w = min(r0.z, r0.x);
+    r2.z = min(r2.z, r2.w);
+    r0.y = min(r2.z, r0.y);
+    r1.w = max(r1.w, r3.x);
+    r0.w = max(r2.y, r0.w);
+    r0.w = max(r1.w, r0.w);
+    r1.x = max(r1.x, r1.z);
+    r0.x = max(r0.z, r0.x);
+    r0.xw = max(r1.xy, r0.xw);
+    r0.x = max(r0.w, r0.x);
+    r0.y = max(r2.x, r0.y);
+    r0.x = min(r0.y, r0.x);
+    r0.x = r0.x + -r3.x;
+    r3.x = r3.z * r0.x + r3.x;
+  }
+  if (shader_injection.disable_game_ao >= 0.5f) {
+    r3.x = 1.0f;
+  }
+  r3.w = 0;
+  u0[vThreadID.xy] = r3.xyzw;
+  return;
+}

src/games/endfield/bloom/bloom-prefilter_0x2B198371.cs_5_0.hlsl → src/games/endfield/bloom/bloom-prefilter_0xB3082097.cs_5_0.hlsl

@@ -22,7 +22,7 @@ cbuffer cb1 : register(b1)
 
 cbuffer cb0 : register(b0)
 {
-  float4 cb0[110];
+  float4 cb0[28];
   // cb0[82].zw = sample offset scale
   // cb0[109].x = bloom intensity multiplier
 }
@@ -61,7 +61,7 @@ void main(uint3 vThreadID : SV_DispatchThreadID)
   while (true) {
     r1.w = cmp((int)r0.w >= 13);
     if (r1.w != 0) break;
-    r2.xy = icb[r0.w+0].xy * cb0[82].zw;
+    r2.xy = icb[r0.w+0].xy * cb0[0].zw;
     r2.xy = r0.xy * cb1[0].zw + r2.xy;
     r2.xyz = t0.SampleLevel(s0_s, r2.xy, 0).xyz;
     r1.w = max(r2.x, r2.y);
@@ -76,7 +76,7 @@ void main(uint3 vThreadID : SV_DispatchThreadID)
     r1.w = r2.w / r1.w;               
     r2.xyz = r2.xyz * r1.www;
 
-    r2.xyz = cb0[109].xxx * r2.xyz * (BLOOM_STRENGTH / 50.0f);
+    r2.xyz = cb0[27].xxx * r2.xyz * (BLOOM_STRENGTH / 50.0f);
 
     r1.w = dot(r2.xyz, float3(0.212672904,0.715152204,0.0721750036)); 
     r1.w = 1 + r1.w;
@@ -99,4 +99,4 @@ void main(uint3 vThreadID : SV_DispatchThreadID)
   r0.xyzw = r1.xyzw * r0.xxxx;           
   u0[vThreadID.xy] = r0.xyzw;
   return;
-}
+}