Mfcttrf

I use this script to build buildings on my terrain, and it works with the collider somehow. But I don't understand what it is that the raycast does to find the right place to put my building.

Ray ray = Camera.main.ScreenPointToRay(Input.mousePosition);
RaycastHit hitInfo;
if (Physics.Raycast(ray, out hitInfo))

 farmhide.transform.position = hitInfo.point;

 farmhide.transform.rotation = Quaternion.FromToRotation(Vector3.up, hitInfo.normal);

What is it that this code is doing?

A raycast is like shining a laser pointer in your scene, to see what the laser hits.

It takes a starting point and a direction as input (together, they make a "ray").

The physics engine takes this point and sweeps it along that direction. As it goes, it searches through all the physics colliders in your scene to check whether the point hits any of them during its travel.

(This is a simplification - in reality the engine has lots of clever acceleration structures and math so it can do this efficiently without looking at every collider or checking every point along the line one by one)

If it finds a collider that's in the path of the travelling point, it returns "true," meaning "yes, I found a hit." If you've provided an out RaycastHit parameter, it will store information about the first hit - the closest object along the ray.

From that RaycastHit, you can learn a lot about the intersection:

• hitInfo.collider will give you the Collider component that the ray hit. You can use this to get the hit object's GameObject and check its tag with hitInfo.collider.gameObject.CompareTag("someTag"), for example.




• hitInfo.point will tell you the position in world space where the ray hit the surface of the collider.




• hitInfo.normal will tell you the direction that the surface of the collider is facing at the point you hit. Your script uses this to orient the buildings to point directly out from the ground, instead of diagonal to the surface.




• hitInfo.rigidbody will give you the physics Rigidbody that owns the collider you hit, if it was a dynamic or kinematic object.




• etc.

If you don't want to search every collider in your scene, one of the best ways to control the raycast is to use Physics Layers.

You can expose a public LayerMask raycastLayers; variable in your script, so you can choose in the Inspector which physics layers you're interested in.

Then you can modify your raycast like so:

bool hitSomething = Physics.Raycast(
 ray, // Point + direction to sweep
 out hitInfo, // Destination for hit information
 float.positiveInfinity, // Change this to limit the ray length
 raycastLayers, // Only hit colliders on these layers
 QueryTriggerInteraction.Ignore // Ignore trigger colliders
 );

if (hitSomething) {...

This is more efficient than checking for specific tags, because it lets you filter out false positives at the source.