diff --git a/src/shader_recompiler/ir/reinterpret.h b/src/shader_recompiler/ir/reinterpret.h
index 99819cbb9..2a18f394a 100644
--- a/src/shader_recompiler/ir/reinterpret.h
+++ b/src/shader_recompiler/ir/reinterpret.h
@@ -46,6 +46,10 @@ inline F32 ApplyReadNumberConversion(IREmitter& ir, const F32& value,
         const IR::F32 max = ir.Imm32(float(std::numeric_limits<u16>::max()));
         return ir.FPDiv(left, max);
     }
+    case AmdGpu::NumberConversion::Uint32ToUnorm: {
+        const auto float_val = ir.ConvertUToF(32, 32, ir.BitCast<U32>(value));
+        return ir.FPDiv(float_val, ir.Imm32(static_cast<float>(std::numeric_limits<u32>::max())));
+    }
     default:
         UNREACHABLE();
     }
@@ -92,6 +96,12 @@ inline F32 ApplyWriteNumberConversion(IREmitter& ir, const F32& value,
         const IR::U32 raw = ir.ConvertFToS(32, ir.FPDiv(left, ir.Imm32(2.f)));
         return ir.BitCast<F32>(raw);
     }
+    case AmdGpu::NumberConversion::Uint32ToUnorm: {
+        const auto clamped = ir.FPClamp(value, ir.Imm32(0.f), ir.Imm32(1.f));
+        const auto unnormalized =
+            ir.FPMul(clamped, ir.Imm32(static_cast<float>(std::numeric_limits<u32>::max())));
+        return ir.BitCast<F32>(U32{ir.ConvertFToU(32, unnormalized)});
+    }
     default:
         UNREACHABLE();
     }
diff --git a/src/video_core/amdgpu/types.h b/src/video_core/amdgpu/types.h
index ab0df689e..f7536f7e2 100644
--- a/src/video_core/amdgpu/types.h
+++ b/src/video_core/amdgpu/types.h
@@ -197,8 +197,9 @@ enum class NumberConversion : u32 {
     UintToUscaled = 1,
     SintToSscaled = 2,
     UnormToUbnorm = 3,
-    Sint8ToSnormNz = 5,
-    Sint16ToSnormNz = 6,
+    Sint8ToSnormNz = 4,
+    Sint16ToSnormNz = 5,
+    Uint32ToUnorm = 6,
 };
 
 struct CompMapping {
@@ -286,6 +287,17 @@ inline DataFormat RemapDataFormat(const DataFormat format) {
 
 inline NumberFormat RemapNumberFormat(const NumberFormat format, const DataFormat data_format) {
     switch (format) {
+    case NumberFormat::Unorm: {
+        switch (data_format) {
+        case DataFormat::Format32:
+        case DataFormat::Format32_32:
+        case DataFormat::Format32_32_32:
+        case DataFormat::Format32_32_32_32:
+            return NumberFormat::Uint;
+        default:
+            return format;
+        }
+    }
     case NumberFormat::Uscaled:
         return NumberFormat::Uint;
     case NumberFormat::Sscaled:
@@ -341,6 +353,17 @@ inline CompMapping RemapSwizzle(const DataFormat format, const CompMapping swizz
 
 inline NumberConversion MapNumberConversion(const NumberFormat num_fmt, const DataFormat data_fmt) {
     switch (num_fmt) {
+    case NumberFormat::Unorm: {
+        switch (data_fmt) {
+        case DataFormat::Format32:
+        case DataFormat::Format32_32:
+        case DataFormat::Format32_32_32:
+        case DataFormat::Format32_32_32_32:
+            return NumberConversion::Uint32ToUnorm;
+        default:
+            return NumberConversion::None;
+        }
+    }
     case NumberFormat::Uscaled:
         return NumberConversion::UintToUscaled;
     case NumberFormat::Sscaled: