# DX11 Picking triangle

This topic is 2139 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

## Recommended Posts

Based on this tutorial I have try to make it work with triangles with no success.

this tutorial is demonstrating how to how to do picking on the a sphere, so in theory if i take the rayOrigin and rayDirection can work with a  triangle.(without the MatrixTranslation for the center of the sphere), but don't because something needs to be changed regarding the rayOrigin / rayDirection.

all i want is to get the intersection point on the triangle.

struct triangle {
D3DXVECTOR3 V0;
D3DXVECTOR3 V1;
D3DXVECTOR3 V2;
};
struct Ray {
D3DXVECTOR3 P0;
D3DXVECTOR3 P1;
};
...
intersect3D_RayTriangle(Ray, triangle, out intersection point);//this method works fine, if you don't put the ray data that comes from this tutorial

Edited by xlope01

##### Share on other sites

http://www.realtimerendering.com/intersections.html has multiple methods for ray/trianlge intersection (and is pretty much the to-go-to page for any intersection related questions)

##### Share on other sites

Maybe i didn't make my self clear I already have a ray/trianlge intersection method that is working, the problem is the ray that is created by mouse clicking is some how wrong.

TestIntersection(int visibleMouseX, int visibleMouseY)
{
float pointX, pointY;
D3DXMATRIX projectionMatrix, viewMatrix, inverseViewMatrix, worldMatrix, translateMatrix, inverseWorldMatrix;
D3DXVECTOR3 direction, origin, rayOrigin, rayDirection;

// Move the mouse cursor coordinates into the -1 to +1 range.
pointX = ((2.0f * (float)visibleMouseX) / (float)m_ResolutionWidth) - 1.0f;
pointY = (((2.0f * (float)visibleMouseY) / (float)m_ResolutionHight) - 1.0f) * -1.0f;

// Adjust the points using the projection matrix to account for the aspect ratio of the viewport.
m_D3D->GetProjectionMatrix(projectionMatrix);
pointX = pointX / projectionMatrix._11;
pointY = pointY / projectionMatrix._22;

// Get the inverse of the view matrix.
m_Camera->GetViewMatrix(viewMatrix);
D3DXMatrixInverse(&inverseViewMatrix, NULL, &viewMatrix);

// Calculate the direction of the picking ray in view space.
direction.x = (pointX * inverseViewMatrix._11) + (pointY * inverseViewMatrix._21) + inverseViewMatrix._31;
direction.y = (pointX * inverseViewMatrix._12) + (pointY * inverseViewMatrix._22) + inverseViewMatrix._32;
direction.z = (pointX * inverseViewMatrix._13) + (pointY * inverseViewMatrix._23) + inverseViewMatrix._33;

// Get the origin of the picking ray which is the position of the camera.
origin = m_Camera->GetPosition();

// Get the world matrix and translate to the location of the sphere.
m_D3D->GetWorldMatrix(worldMatrix);
//D3DXMatrixTranslation(&translateMatrix, 1800.0f, 20.0f, 1900.0f);
D3DXMatrixMultiply(&worldMatrix, &worldMatrix, &translateMatrix);

// Now get the inverse of the translated world matrix.
D3DXMatrixInverse(&inverseWorldMatrix, NULL, &worldMatrix);

// Now transform the ray origin and the ray direction from view space to world space.
D3DXVec3TransformCoord(&rayOrigin, &origin, &inverseWorldMatrix);
D3DXVec3TransformNormal(&rayDirection, &direction, &inverseWorldMatrix);

// Normalize the ray direction.
D3DXVec3Normalize(&rayDirection, &rayDirection);

Ray r;
r.P0 = rayOrigin;
r.P1 = rayDirection;
triangle tri;
tri.V0 = D3DXVECTOR3(1.0f, 0.0f, 1.0f);
tri.V1 = D3DXVECTOR3(100.0f, 0.0f, 100.0f);
tri.V2 = D3DXVECTOR3(0.0f, 0.0f, 100.0f);
D3DXVECTOR3 intersect;
intersect3D_RayTriangle(r, tri, &intersect);

return intersect;
}

##### Share on other sites

"r.P1" sounds like a point to me, and yet you're passing in a unit vector instead.

I don't know you could expect any useful advice without telling anyone what's wrong.

##### Share on other sites

"r.P1" sounds like a point to me, and yet you're passing in a unit vector instead.

I don't know you could expect any useful advice without telling anyone what's wrong.

I take rayOrigin as point 1 and rayDirection as point 2 crating a line, and i believe that is not the correct way to do it ?

##### Share on other sites
Nicely spotted, Pink Horror.

Edit: If your triangle intersection function defines a ray as two points, then you need to give it two points, not a point and a direction (vector).

Picking (or any collision/intersection test for that matter) need to happen in the same space. I think rastertek - although the code presumably works for the sphere - is making a bit of a fuss with the spaces. E.g. using the inverse world transform goes from world to local (model) space, not from view to world. Check if you got that right. Edited by unbird

##### Share on other sites

for ray generation, i used the code from the raypick example in directx. works great. hooked it up to a ray-plane test for the plane y=0 to do skycam picks onto a world map for RTS type camera views.

i've noted that rastertek, while good, sometimes does weird stuff - like the code at the start of their "calculating frustum cull planes" examples.  always compare their stuff to at least one other source if unsure.

##### Share on other sites

ok i get it I will use the vector "rayDirection", however all the methods( triangle intersection ) I find take as input a ray that require point 1 and point 2.

##### Share on other sites

r.P0 = rayOrigin;
r.P1 = rayOrigin + rayDirection;

• 18
• 29
• 11
• 21
• 16