# DX11 Display Quad in top left corner.

This topic is 1660 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

## Recommended Posts

Hi there,
I am very new to rendering, and i have a problem displaying a quad on the top left corner of my screen.
I already have a small 3D engine, and i wanted to use this code http://www.rastertek.com/dx11tut22.html to display my quad.
The issue is that the quad shows, but way off the -x axis and up the y axis. So i actually have to move my camera down the -z axis and tilt it to the left to see the quad.

Camera position and aiming when starting :

After going way back and turning a on the left

I just want it to stick in front of my camera in the top left corner of my screen.
My guts tell me there is something wrong with my view matrix, or at least, it is not setup the way the rastertek tutorial expected it, but i can't find exactly what it is.

cbuffer PerFrameBuffer
{
float4x4 gWorldViewProj;
};

PixelInputType VS(VertexInputType input)
{
PixelInputType output;

// Change the position vector to be 4 units for proper matrix calculations.
input.position.w = 1.0f;

// Calculate the position of the vertex against the world, view, and projection matrices.
output.position = mul(float4(input.position.xyz, 1.0f), gWorldViewProj);

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

return output;
}


XMMATRIX worldViewProj = world * camera->View() * camera->Proj();
mTextureEffect->mfxWorldViewProj->SetMatrix(reinterpret_cast<float*>(&(worldViewProj)));


XMMATRIX Camera::View()const
{
}

void Camera::UpdateViewMatrix()
{

// Keep camera's axes orthogonal to each other and of unit length.
L = XMVector3Normalize(L);
U = XMVector3Normalize(XMVector3Cross(L, R));

// U, L already ortho-normal, so no need to normalize cross product.
R = XMVector3Cross(U, L);

// Fill in the view matrix entries.
float x = -XMVectorGetX(XMVector3Dot(P, R));
float y = -XMVectorGetX(XMVector3Dot(P, U));
float z = -XMVectorGetX(XMVector3Dot(P, L));

XMStoreFloat3(&mRight, R);
XMStoreFloat3(&mUp, U);
XMStoreFloat3(&mLook, L);

mView(0,0) = mRight.x;
mView(1,0) = mRight.y;
mView(2,0) = mRight.z;
mView(3,0) = x;

mView(0,1) = mUp.x;
mView(1,1) = mUp.y;
mView(2,1) = mUp.z;
mView(3,1) = y;

mView(0,2) = mLook.x;
mView(1,2) = mLook.y;
mView(2,2) = mLook.z;
mView(3,2) = z;

mView(0,3) = 0.0f;
mView(1,3) = 0.0f;
mView(2,3) = 0.0f;
mView(3,3) = 1.0f;
}


The shader output vertices with very big negative values, when my camera is almost centered at the world origin, so there is no way it can see those.
So i am a bit confused about how i am supposed to use the view transformation (if the problem indeed lies there) so i can always see my quad no matter where i look.
Any help or suggestion would be greatly appreciated :-)

Thanks !

##### Share on other sites

You don't really need view matrix to render 2D image on screen.

Try to take a look at this tutorial, which is specifically made to render 2D things on the screen, not 3D space.

It should explain how to position things on the screen, and you should be able to easily see where's your mistake.

Edited by Zaoshi Kaba

##### Share on other sites

You don't really need view matrix to render 2D image on screen.

Try to take a look at this tutorial, which is specifically made to render 2D things on the screen, not 3D space.

It should explain how to position things on the screen, and you should be able to easily see where's your mistake.

Thanks a lot for the reply !

I read that tutorial, and i have to admit i am still confused because i think this is extacly what i am doing.

But i kept diging into http://www.d3dcoder.net/d3d11.htm and i found a similar code in the shadowmap sample which works for me :

I just don't know why yet.

Contrary to rastertek code, the d3dcoder code creates vertices directly in NDC space

	// Load the vertex array with data.
vertices[0].position = XMFLOAT3(-1.0f, -1.0f, 0.0f);
vertices[0].texture = XMFLOAT2(0.0f, 0.0f);

vertices[1].position = XMFLOAT3(-1.0f, +1.0f, 0.0f);
vertices[1].texture = XMFLOAT2(1.0f, 1.0f);

vertices[2].position = XMFLOAT3(+1.0f, +1.0f, 0.0f);
vertices[2].texture = XMFLOAT2(0.0f, 1.0f);

vertices[3].position = XMFLOAT3(+1.0f, -1.0f, 0.0f);
vertices[3].texture = XMFLOAT2(0.0f, 0.0f);


and only sends this world transform matrix to the shader :

XMMATRIX world(
0.5f, 0.0f, 0.0f, 0.0f,
0.0f, 0.5f, 0.0f, 0.0f,
0.0f, 0.0f, 1.0f, 0.0f,
0.5f, -0.5f, 0.0f, 1.0f);


Nothing else.

No projection or view matrix sent at all. And no matter where i look, the quad remains visible.

Its like the quad is not part of the rest of the "world" and "bypassed" the whole 3d part.

How is this possible ?

##### Share on other sites

Its like the quad is not part of the rest of the "world" and "bypassed" the whole 3d part.

How is this possible ?

Its just how graphics work. Your GPU does not know nor care what a view, projection matrix is, or whats drawn 3d and what not. It just sees vertices, and performs transformations on those in the vertex shader. You could even display a quad from (-1, -1, 0) to (1, 1, 0) without any transformation applied. What that matrix you supplied here does is scale the quad to 50% of its size, and move it 0.5 to the right, and 0.5 down. You don't need anything else, try to replace that matrix with identity and see what happens.

##### Share on other sites

Every vertices in the frustrum ends up being projected on the screen, and my screen surface IS the NDC space, ranging from -1 to 1.

So the whole point of world * view * proj operation is to convert my vertex from its 3D position to a 2D position in the NDC space.
And nothing stops me from sending vertices with NDC coordinates directly to the GPU.
I feel very stupid for realizing this only now... The whole thing makes sense now !

Thank you very much :-)

Note: Please do not hesitate to correct me if i came to a wrong conclusion, i would much rather know it now ;-)

• 10
• 11
• 9
• 16
• 19
• ### Similar Content

• I wanted to see how others are currently handling descriptor heap updates and management.
I've read a few articles and there tends to be three major strategies :
1 ) You split up descriptor heaps per shader stage ( i.e one for vertex shader , pixel , hull, etc)
2) You have one descriptor heap for an entire pipeline
3) You split up descriptor heaps for update each update frequency (i.e EResourceSet_PerInstance , EResourceSet_PerPass , EResourceSet_PerMaterial, etc)
The benefits of the first two approaches is that it makes it easier to port current code, and descriptor / resource descriptor management and updating tends to be easier to manage, but it seems to be not as efficient.
The benefits of the third approach seems to be that it's the most efficient because you only manage and update objects when they change.

• hi,
until now i use typical vertexshader approach for skinning with a Constantbuffer containing the transform matrix for the bones and an the vertexbuffer containing bone index and bone weight.
Now i have implemented realtime environment  probe cubemaping so i have to render my scene from many point of views and the time for skinning takes too long because it is recalculated for every side of the cubemap.
For Info i am working on Win7 an therefore use one Shadermodel 5.0 not 5.x that have more options, or is there a way to use 5.x in Win 7
My Graphic Card is Directx 12 compatible NVidia GTX 960
the member turanszkij has posted a good for me understandable compute shader. ( for Info: in his engine he uses an optimized version of it )
Now my questions
is it possible to feed the compute shader with my orignial vertexbuffer or do i have to copy it in several ByteAdressBuffers as implemented in the following code ?
the same question is about the constant buffer of the matrixes
my more urgent question is how do i feed my normal pipeline with the result of the compute Shader which are 2 RWByteAddressBuffers that contain position an normal
for example i could use 2 vertexbuffer bindings
1 containing only the uv coordinates
2.containing position and normal
How do i copy from the RWByteAddressBuffers to the vertexbuffer ?

(Code from turanszkij )
Here is my shader implementation for skinning a mesh in a compute shader:

• Hi, can someone please explain why this is giving an assertion EyePosition!=0 exception?

It looks like DirectX doesnt want the 2nd parameter to be a zero vector in the assertion, but I passed in a zero vector with this exact same code in another program and it ran just fine. (Here is the version of the code that worked - note XMLoadFloat3(&m_lookAt) parameter value is (0,0,0) at runtime - I debugged it - but it throws no exceptions.
and here is the repo with the alternative version of the code that is working with a value of (0,0,0) for the second parameter.

• Hi, can somebody please tell me in clear simple steps how to debug and step through an hlsl shader file?
I already did Debug > Start Graphics Debugging > then captured some frames from Visual Studio and
double clicked on the frame to open it, but no idea where to go from there.

I've been searching for hours and there's no information on this, not even on the Microsoft Website!
They say "open the  Graphics Pixel History window" but there is no such window!
Then they say, in the "Pipeline Stages choose Start Debugging"  but the Start Debugging option is nowhere to be found in the whole interface.
Also, how do I even open the hlsl file that I want to set a break point in from inside the Graphics Debugger?

All I want to do is set a break point in a specific hlsl file, step thru it, and see the data, but this is so unbelievably complicated