Direct3D 11 使用向量而不是数组

Question

So basically I am trying to learn Direct3D 11. I have got my program running and importing OBJ files with good success. But now I have decided to use vectors instead of arrays since I may want to change the models that I load, and would like it to automatically allocate the required memory. Anyway, it now only seems to display half of the model, I believe it is to do with the ByteWidth and/or the stride, but I have now tried so many different combinations and different setups that I feel I am going round in circles, and I am certain that it is within this block of code here:

ID3D11Buffer *pVertexBuffer;
D3D11_BUFFER_DESC bd;
bd.BindFlags = D3D11_BIND_VERTEX_BUFFER;   
bd.ByteWidth = vertices.size() * sizeof(Vertex);  
bd.CPUAccessFlags = 0u;
bd.MiscFlags = 0u;
bd.Usage = D3D11_USAGE_DEFAULT;                
bd.StructureByteStride = sizeof(Vertex);
D3D11_SUBRESOURCE_DATA sd = {};
sd.pSysMem = vertices.data();
dev->CreateBuffer(&bd, &sd, &pVertexBuffer);
const UINT stride = sizeof(Vertex);
const UINT offset = 0u;
devcon->IASetVertexBuffers(0u, 1u, &pVertexBuffer, &stride, &offset);

ID3D11Buffer *pIndexBuffer;
D3D11_BUFFER_DESC ibd;
ibd.BindFlags = D3D11_BIND_INDEX_BUFFER;   
ibd.ByteWidth = indices.size() * sizeof(Index);  
ibd.CPUAccessFlags = 0u;
ibd.MiscFlags = 0u;
ibd.Usage = D3D11_USAGE_DEFAULT;                
ibd.StructureByteStride = sizeof(Index);
D3D11_SUBRESOURCE_DATA isd = {};
isd.pSysMem = indices.data();
dev->CreateBuffer(&ibd, &isd, &pIndexBuffer);
devcon->IASetIndexBuffer(pIndexBuffer, DXGI_FORMAT_R16_UINT, 0u);

for reference, I have pastebinned the full code: here

Answer 1

Here's my guess.

devcon->IASetIndexBuffer(pIndexBuffer, DXGI_FORMAT_R16_UINT, 0u);

Right here you are telling directx that your indices are uint16_t however you are declaring your indices as 'unsigned int'.

This is bad.

unsigned int is typically 32 bits (4 bytes). And uint16 is 16 bits (2 bytes).

Therefor there is a mismatch.

struct Index{
    unsigned int a, b, c;
};

You shouldn't use unsigned int. You should use uint16_t.

If this isn't the problem then you should make this change anyway because you are just asking for trouble in my opinion.

When working with graphics APIs you be very careful with your chosen types to avoid these types of runtime issues.

Here is the link to the reference on fixed width types: https://en.cppreference.com/w/cpp/types/integer