Move totaliterations calculation to CPU. This frees 4 VGPRs.

Now the code is back to the limit of 32 VGPRs without the need to directly operate on the state buffer.
This means that now it's bound by memory access.

1 files changed, 4 insertions, 1 deletions

diff --git a/BuddhaTest/src/BuddhaTest.cpp b/BuddhaTest/src/BuddhaTest.cpp
index 3b99169..0a99fee 100644
--- a/BuddhaTest/src/BuddhaTest.cpp
+++ b/BuddhaTest/src/BuddhaTest.cpp
@@ -148,6 +148,7 @@ int main(int argc, char * argv[])
 
     glUseProgram(ComputeShader);
     GLint orbitLengthUniformHandle = glGetUniformLocation(ComputeShader, "orbitLength");
+    GLint totalIterationsUniformHandle = glGetUniformLocation(ComputeShader, "totalIterations");
     GLint widthUniformComputeHandle = glGetUniformLocation(ComputeShader, "width");
     GLint heightUniformComputeHandle = glGetUniformLocation(ComputeShader, "height");
     GLint iterationsPerDispatchHandle = glGetUniformLocation(ComputeShader, "iterationsPerDispatch");
@@ -155,6 +156,9 @@ int main(int argc, char * argv[])
     glUniform1ui(widthUniformComputeHandle, settings.imageWidth);
     glUniform1ui(heightUniformComputeHandle, bufferHeight);
 
+    const uint32_t maxOrbitlength = std::max(std::max(settings.orbitLengthBlue,settings.orbitLengthGreen),settings.orbitLengthRed);
+    glUniform1ui(totalIterationsUniformHandle, maxOrbitlength);
+
     glUseProgram(VertexAndFragmentShaders);
     GLint widthUniformFragmentHandle = glGetUniformLocation(VertexAndFragmentShaders, "width");
     GLint heightUniformFragmentHandle = glGetUniformLocation(VertexAndFragmentShaders, "height");
@@ -169,7 +173,6 @@ int main(int argc, char * argv[])
 
     uint64_t totalIterationCount{0};
     uint64_t lastMessage{0};
-    const uint64_t maxOrbitlength = std::max(std::max(settings.orbitLengthBlue,settings.orbitLengthGreen),settings.orbitLengthRed);
 
     /* Loop until the user closes the window */
     while (!glfwWindowShouldClose(window))