I am trying to set up a simple scene (one spherical node and the default camera) in a square SceneView. Currently I set up the scene as below:
let scene = SCNScene()
let planet = SCNSphere(radius: 1.0)
let planetNode = SCNNode(geometry: planet)
scene.rootNode.addChildNode(planetNode)
To certain views, I also rotate the node as such:
let rotationNode = SCNNode()
rotationNode.addChildNode(planetNode)
scene.rootNode.addChildNode(rotationNode)
rotationNode.rotation = (SCNVector4: SCNVector4(x: 0, y: 0, z: 1, w: some_amount_of_radians))
What I noticed, however, is the objects that get rotated are smaller than the ones that don't get rotated. I am not really sure what the ratio is, but it seems to be dependent on how much rotation is added, to a point.
In the below screenshot, Earth is rotated 45 degrees, and the other 2 are not rotated. If I rotated it 90 degrees instead, there is no difference, which leads me to believe there is a square bounding box around the sphere and the default camera is forcing its point of view to contain this box.
I have also tried to change the euler angles, position, and scale of the rotated nodes to compensate, but no transormations I apply seem to have any effect. Any pointers for solving this camera issue would be perfect.
Related
I am using two virtual joysticks to move my camera around the scene. The left stick controls the position and the right one controls the rotation.
When using the right stick, the camera rotates, but it seems that the camera rotates around the center point of the model.
This is my code:
fileprivate func rotateCamera(_ x: Float, _ y: Float)
{
if let cameraNode = self.cameraNode
{
let moveX = x / 50.0
let rotated = SCNMatrix4Rotate(cameraNode.transform, moveX, 0, 1, 0)
cameraNode.transform = rotated
}
}
I have also tried this code:
fileprivate func rotateCamera(_ x: Float, _ y: Float)
{
if let cameraNode = self.cameraNode
{
let moveX = x / 50.0
cameraNode.rotate(by: SCNQuaternion(moveX, 0, 1, 0), aroundTarget: cameraNode.transform)
}
}
But the camera just jumps around. What is my error here?
There are many ways to handle rotation, some are very suitable for giving headaches to the coder.
It sounds like the model is at 0,0,0, meaning it’s in the center of the world, and the camera is tranformed to a certain location. In the first example using matrices, you basically rotate that transformation. So you transform first, then rotate, which yes will cause it to rotate around the origin (0,0,0).
What you should do instead, to rotate the camera in local space, is rotate the camera first in local space and then translate it to its position in world space.
Translation x rotation matrix results in rotation in world space
Rotation x translation matrix results in rotation in local space
So a solution is to remove the translation from the camera first (moving it back to 0,0,0), then apply the rotation matrix, and then reapply the translation. This comes down to the same result as starting with an identity matrix. For example:
let rotated = SCNMatrix4Rotate(SCNMatrixIdentity, moveX, 0, 1, 0)
cameraNode.transform = SCNMatrix4Multiply(rotated, cameraNode.transform)
I'm trying to make an ARKit application where an SCNNode, in this case a box, is placed in front of the camera, facing the camera. As the user moves the camera around, objects are placed when a certain distance has been moved. This would leave you with a series of nodes facing the camera in a line, equally spaced.
I have this working to a certain extent, but my problem is with the rotation. I'm currently taking all axes of rotation, so as the user re-orients their phone, the rotation of the node matches. I want to restrict this to just the rotation around the y-axis. The ideal outcome is a domino-trail like look, with all the objects having the same x and z rotations, but potentially different y rotations.
I hope I've explained this clearly enough!
Here's the code I'm currently using:
func createNode(fromCameraTransform cameraTransform: matrix_float4x4) -> SCNNode {
let geometry = SCNBox(width: 0.02, height: 0.04, length: 0.01, chamferRadius: 0)
let physicsBody = SCNPhysicsBody(type: .dynamic, shape: SCNPhysicsShape(geometry: geometry))
physicsBody.mass = 1000
let node = SCNNode(geometry: geometry)
node.physicsBody = physicsBody
var translationMatrix = matrix_identity_float4x4
translationMatrix.columns.3.x = 0.05 // Moves the node down in world space
translationMatrix.columns.3.z = -0.1 // Moves the object away from the camera
node.simdTransform = simd_mul(cameraTransform, translationMatrix)
return node
}
I've tried different combinations of extracting values from the second column of the cameraTransform and setting them as eulerAngles, rotation and simdRotation, but to no avail.
I've also tried extracting values from the pointOfView of the current sceneView and assigning them to the same values as listed above, but again, no luck.
Any help would greatly appreciated!
I know a little bit about this, but am really just starting out with SceneKit and 3D transformations/matrices so be gentle with me!
I think I know what your trying to do, basically automatically drop each new domino so its evenly spaced following the camera.pointOfView trail.
You can update the new nodes euler angles y-axis to the same as the cameras pointofView eularAngles.y. So as you move the camera around the next node you are placing is always facing towards the camera (only rotating around the y-axis).
The renderer function updateAtTime below gets called everytime
the camera moves
func renderer(_ renderer: SCNSceneRenderer, updateAtTime time: TimeInterval) {
// You set the camera’s pointOfView’s eularAngle for y-xis to the node you are
about to place.
node.eulerAngles.y = (sceneView.pointOfView?.eulerAngles.y)!
I had this working in a playground so it does work.
Edit: This solution above angle was having a gimbal lock problem (as you went around in a circle in would reset its angle back to the axis.
So I found this approach using SCNBillboardConstraint works without experiencing the gimbal lock problem as you go around in circle.
let yFreeConstraint = SCNBillboardConstraint()
yFreeConstraint.freeAxes = .Y
node.constraints = [yFreeConstraint]
node.eulerAngles = node.presentation.eulerAngles
node.position = position
I have a SCNNode - sceneNode - which is a child of rootNode and contains all of my child nodes. Upon the user tapping a button, I want to rotate the scene around a certain point on the y-axis. For example, the camera's point of view is known, and I want to rotate everything by 90º around the camera. The camera is no longer at 0, 0, 0.
I've tried playing around with the SCNMatrix4MakeTranslation function, and then changing the y value on the euler angles, but I've not been able to get it to work expectedly.
Your question is rather ambiguous, if you were to rotate the entire scene 90degrees “around the camera” the scene would end up next to the camera and you wouldn’t see it.
To rotate a SCNNode around a point other than its own pivot, as the question in the title, create a translation matrix that moves it to that point, multiply with a rotation matrix to the SCNNode, and then multiply with the inverse of the translation matrix.
However, what you probably want to do instead is do basically the same process with the camera node instead of the sceneNode. That will make the camera move around the sceneNode, giving the appearance the scene is rotating in place.
For example:
//get the current camera's transform and store it in cammat
GLKMatrix4 cammat = SCNMatrix4ToGLKMatrix4(self.myView.pointOfView.transform);
//create a rotation matrix of 90 degrees over axis Y
GLKQuaternion quaty = GLKQuaternionMakeWithAngleAndAxis(M_PI/2, 0, 1, 0);
GLKMatrix4 rotMat = GLKMatrix4MakeWithQuaternion(quaty);
//set the camera transform to rotMat * cammat, which basically rotates the camera first, and then moves it back to the same offset it was.
self.myView.pointOfView.transform = SCNMatrix4FromGLKMatrix4(GLKMatrix4Multiply(rotMat, cammat));
I am very new to SceneKit and your help will be really appreciated!
I have a 200x200 sized SCNView in my UIView, which is at the centre of super view.
I want to put a SCNCylinder inside, such that the SCNCylinder covers full SCNView. I read that all these views of Scenekit are defined in meters, so how do I form a relationship between the dimensions of my screen and the
SCNCylinder.
I tried:
var coinNode = SCNNode()
let coinGeometry = SCNCylinder(radius: 100, height: 2)
coinNode = SCNNode(geometry: coinGeometry)
coinNode.position = SCNVector3Make(0, 0, 0)
coinScene.rootNode.addChildNode(coinNode)
let rotate90AboutZ = SCNAction.rotateByX(-CGFloat(M_PI_2), y: 0.0, z: CGFloat(M_PI_2), duration: 0.0)
coinNode.runAction(rotate90AboutZ)
ibOutletScene.scene = coinScene
But this leaves a margin between my coinScene and the ibOutletScene. How do I remove this space?
I also tried adding Camera:
let cameraNode = SCNNode()
cameraNode.camera = SCNCamera()
cameraNode.position = SCNVector3Make(0, 0, 100)
coinScene.rootNode.addChildNode(cameraNode)
But I see random behaviour with this and the coinNode gets hidden! How should I position my camera? Or is there any other way to remove extra space from my ibOutletScene?
Edit:
This is how it looks if I don't add camera. There is a margin between red scene and green coin. I tried multiple sizes for the coin, but I am unable to remove this margin unless I add a camera. But, If I add camera, I get another problem, mentioned below this screenshot.
If I don't add the camera, The rotation animation on coin works perfectly, but If I add camera,the rotation enlarges the and then becomes small again with the animation. How can I rotate it on its axis, without increasing the size?
I am using following code to rotate the coin:
The same code works fine without camera, but enlarges the coin after adding camera. Checkout the snapshot.
let rotate = SCNAction.rotateByX(0, y: -2 * CGFloat(M_PI_2), z: 0, duration: 2)
coinNode.runAction(rotate)
The random behavior might be caused by the last line in your first code snippet. You're starting an animation, and then adding the scene to the view.
Instead, build your scene, attach it to the view, and then start your animation. Setting a non-zero duration for the action will give you a more pleasing transition.
As for the extra space, it would help us understand if you post a screenshot. But you're going to have to do a bit of trigonometry.
It looks like you have a scene that you want to be blocked by a coin, that then rotates out of the way? Simulate that yourself with real objects. Put your eye down at the edge of your desk. Put a coin out a ways from your eye. How far does that coin have to be in order to block particular objects farther away on your desk?
In SceneKit, you can query the field of view of the SCNCamera. You know the size of your coin and the size of the view. Calculate the distance from the camera needed for the projected diameter of your coin to equal the width of your view. Put the coin there.
I have a SCNScene that contains a SCNSphere encapsulating the SCNCamera (camera is inside # center of the SCNSphere, displaying its inner wall on screen).
Is there a method to query or a way to infer / calc the VISIBLE area of this sphere? I dont want the area of the screen in points, or the area of the scene itself, but what the camera is able to render from this sphere in a given device's screen (ipad and iphone do show a different amount of a 360 frame)
Help appreciated.
CameraNode.positions just set on the front
cameraRollNode.position = SCNVector3(x: 0, y: 0, z: X)
Set sphere material:
sphere_.firstMaterial!.cullMode = SCNCullMode.Front;
PIC had two sphere.
HOLP it useful.