Hey.

I have been following this tutorial , but while experimenting I reached a dead end. The addition I was trying to implement is a dynamic updating rotation per instance.

So first, i define


// TriangleTest.h
struct InstanceType
{
D3DXVECTOR3 position;
D3DXVECTOR3 rotation;
};

for my Instance buffer Type, which leads me to define


// vertexshader.vs
struct VertexInputType
{
float3 position : POSITION;
float4 color : COLOR;
float3 instancePosition : POSITION1;
float3 instanceRotation : POSITION2;
};

in my vertex Shader file. Note that I have changed the usage of semantic type TEXCOORD from the tutorial, as I feel more comfortable having positions use POSITION, as well as giving the position a float3 type.

I set up my InputElementDescription array like this


//shaderclass.cpp
polygonLayout[0].SemanticName = "POSITION";
polygonLayout[0].SemanticIndex = 0;
polygonLayout[0].Format = DXGI_FORMAT_R32G32B32_FLOAT;
polygonLayout[0].InputSlot = 0;
polygonLayout[0].AlignedByteOffset = 0;
polygonLayout[0].InputSlotClass = D3D11_INPUT_PER_VERTEX_DATA;
polygonLayout[0].InstanceDataStepRate = 0;
polygonLayout[1].SemanticName = "COLOR";
polygonLayout[1].SemanticIndex = 0;
polygonLayout[1].Format = DXGI_FORMAT_R32G32B32A32_FLOAT;
polygonLayout[1].InputSlot = 0;
polygonLayout[1].AlignedByteOffset = D3D11_APPEND_ALIGNED_ELEMENT;
polygonLayout[1].InputSlotClass = D3D11_INPUT_PER_VERTEX_DATA;
polygonLayout[1].InstanceDataStepRate = 0;
polygonLayout[2].SemanticName = "POSITION";
polygonLayout[2].SemanticIndex = 1;
polygonLayout[2].Format = DXGI_FORMAT_R32G32B32_FLOAT;
polygonLayout[2].InputSlot = 1;
polygonLayout[2].AlignedByteOffset = 0;
polygonLayout[2].InputSlotClass = D3D11_INPUT_PER_INSTANCE_DATA;
polygonLayout[2].InstanceDataStepRate = 1;
polygonLayout[3].SemanticName = "POSITION";
polygonLayout[3].SemanticIndex = 2;
polygonLayout[3].Format = DXGI_FORMAT_R32G32B32_FLOAT;
polygonLayout[3].InputSlot = 1;
polygonLayout[3].AlignedByteOffset = D3D11_APPEND_ALIGNED_ELEMENT;
polygonLayout[3].InputSlotClass = D3D11_INPUT_PER_INSTANCE_DATA;
polygonLayout[3].InstanceDataStepRate = 1;


and create the input layouts.
In my model class (TriangeTest), i use this to update the dynamic instance buffer ( completely created as D3D11_USAGE_DYNAMIC, with D3D11_CP_ACCESS_WRITE)


void TriangleTest::Frame(ID3D11DeviceContext* deviceContext) {
for (int i = 0; i < this->i_instanceCount; i++) {
D3DXVec3Add(&(m_instanceList.rotation), &(m_instanceList.rotation), &D3DXVECTOR3(1,0,0.0f));
}
UpdateBuffers(deviceContext);
}
void TriangleTest::UpdateBuffers(ID3D11DeviceContext* deviceContext) {
HRESULT result;
D3D11_MAPPED_SUBRESOURCE mappedResource;
InstanceType* instancePtr;

result =deviceContext->Map(m_instanceBuffer, 0, D3D11_MAP_WRITE_DISCARD, 0, &mappedResource);
if(FAILED(result)) throw CException("Could not remap instance buffer");
instancePtr = (InstanceType*)mappedResource.pData;
memcpy(instancePtr, (void*)m_instanceList, (sizeof(InstanceType) * i_instanceCount));
deviceContext->Unmap(m_instanceBuffer, 0);
}

, where m_instanceList is an array of InstanceType holding all information on the position and rotation per triangle. Of course, theFrame(...) function gets called once per frame.

Then I use the shader as provided and replace instancePosition with instanceRotation,which should make my triangles slide out to the right.


PixelInputType TextureVertexShader(VertexInputType input)
{
PixelInputType output;

// Change the position vector to be 4 units for proper matrix calculations.
input.position.w = 1.0f;
Here is where we use the instanced position information to modify the position of each triangle we are drawing.
// Update the position of the vertices based on the data for this particular instance.
input.position.x += input.instanceRotation.x;
input.position.y += input.instanceRotation.y;
input.position.z += input.instanceRotation.z;
// Calculate the position of the vertex against the world, view, and projection matrices.
output.position = mul(input.position, worldMatrix);
output.position = mul(output.position, viewMatrix);
output.position = mul(output.position, projectionMatrix);

// Store the texture coordinates for the pixel shader.
output.tex = input.tex;

return output;
}


But when I run it (complete without errors), the triangles stay still on the screen. Even when using instancePosition again. They move however, if I change the semantic type of inputPosition to TEXCOORD and update the position in the Frame procedure, but even then I get no motion from the instanceRotation.

Which leads me to think, that my InputElementLayoutDescription is wrong. But how? I have tried many a combination already, but don't understand.

Can anyone enlighten me with their ideas?

I'm taking a bit of a guess here but it might be because your first semantic usage of POSITION is not POSITION0 (zero) which is causing the binding to fail for some reason; try giving it the semantic POSITION0 and see if that helps.

That actually did the trick.
Thank you so very much, i thought I had tried this before, but my mind was so messed up with this, I screwed it completely and tested it with some update code disabled. I was already going to tell you how it doesn't work, but kept trying once more to consider your answer, and Voila.

It actually works.