Hi!

I cant seem to figure out how to structure my Vertex and Index buffers in my game engine.

Should i make one method for Rendering any Vertex Array?
or should i make a class for each "Model" in my game that creates own buffers?

Should i use lots of buffers / models or just one that i refill during the game?
If this is the case, how can i reset the size of my vertex at runtime?

Vertex and Index buffers are handles to resources that youve have sent or intend to send to the GPU. The VB/IB structures are not expensive resources in themselves but if neccessary you can create and dispose them as you need. Generally you have one or many VB/IBs per model, rather than attempting to reuse them.

In that sense they are similar to programming in IO streams (in fact if they were called VertexStream and IndexStream you might get a better analogy - although its not perfect). Like IO stream programming, once youve committed data into a stream you rarely modify the contents, as you have explicitly "shared" the contents between other holders of the stream pointer - in this case the GPU. You can modified the contents of the VB/IB but its only done in specific circumstances for specific effects.

If you attempt to channel all your data per frame through one VB/IB you are preventing the GPU from seeing the "whole picture" and force it to process your draw call serially rather than use the parallelism that they specialise in.

Thank you very much. My approach will be as you say and suggest. I think it will be easier code to manage with each VB/IB / model.
I have found this example and this is exactly how i will structure my Models.
The Model itself will have their own VB/IB and fill them up and then on Render, you set this VB/IB to active before you render.

So it would be something like this is pseudo code.


Model triangle
Model qube
Model Level

triangle.init();
qube.init();
Level.init();


render()
{
SetActiveBuffer->triangle
DrawPrimitives(triangle->count);
SetActiveBuffer->cube
DrawPrimitives(cube->count);
SetActiveBuffer->level
DrawPrimitives(leve->count);

}

[source lang="cpp"][/source]

My code, was in pseudo and really bad
Here is how i mean for me render code (still in pseudo but you get the point)
Dont you think this is a good approach?

I was planning to user DrawIndexed for my whole GameLeve (about 1000 triangles)
is this not a good idea?

[source lang="cpp"] // select which primtive type we are using
devcon->IASetPrimitiveTopology(D3D11_PRIMITIVE_TOPOLOGY_TRIANGLELIST);

// Model1

devcon->IASetVertexBuffers(0, 1, &Model1->pVBuffer, &stride, &offset);
devcon->IASetIndexBuffer(Model1->pIBuffer, DXGI_FORMAT_R32_UINT, 0);


devcon->UpdateSubresource(Model1->pCBuffer, 0, 0, &matFinal, 0, 0);
devcon->DrawIndexed(Model1->count, 0, 0);

// Model2

devcon->IASetVertexBuffers(0, 1, &Model2->pVBuffer, &stride, &offset);
devcon->IASetIndexBuffer(Model2->pIBuffer, DXGI_FORMAT_R32_UINT, 0);


devcon->UpdateSubresource(Model2->pCBuffer, 0, 0, &matFinal, 0, 0);
devcon->DrawIndexed(Model2->count, 0, 0);

// Model3

devcon->IASetVertexBuffers(0, 1, &Model3->pVBuffer, &stride, &offset);
devcon->IASetIndexBuffer(Model3->pIBuffer, DXGI_FORMAT_R32_UINT, 0);


devcon->UpdateSubresource(Model3->pCBuffer, 0, 0, &matFinal, 0, 0);
devcon->DrawIndexed(Model3->count, 0, 0);

[/source]

My pseudo code isnt finished.
I will of course hook up a Rendered / each Shader that manages the geometry.
The end code will be like this: in pseudo also =)

ColorShader->Render->SomeModel

My question was only about using ID3D11Buffer in every model for reading the correct Vertex Buffer.

I didnt know if you only could have ONE vertexBuffer and ONE indexBuffer for the program or if you could have this in your model class.


the tutorials at DirectX 11 Tutorials have the same approach i am looking for and that code is slick, so i will try to implement it in the same way.

Sorry about blurry sample-code i think my code only made stuff more confusing... =)