I have followed a tutorial on how to create interactive iOS control center animation using UIViewPropertyAnimator:
http://www.swiftkickmobile.com/building-better-app-animations-swift-uiviewpropertyanimator/
when swiping up or down the bottom menu, after releasing the finger, I want to add pan velocity to UIViewPropertyAnimator and continue the animation:
popupViewPanned(recognizer:) {
switch recognizer.state {
.
.
.
// after finger released
case .end:
// continue all animations using pan velocity with spring timing
let normalizedPanVelocity: // how to normalize pan velocity
runningAnimators.forEach { $0.continueAnimation(withTimingParameters: spring(for: velocity()), durationFactor: 0) }
}
}
func velocity() -> CGVector {
let pan = panRecognizer
let progress = runningAnimators[0].fractionComplete
let fraction = popupOffset*(1 - progress)
return CGVector(with: pan.velocity(in: view), fraction: fraction)
}
func spring(for velocity: CGVector = .zero) -> UITimingCurveProvider {
return UISpringTimingParameters(dampingRatio: 0.9, initialVelocity: velocity)//UISpringTimingParameters(mass: 2.5, stiffness: 80, damping: 25, initialVelocity: velocity)
}
the problem is when I quickly swipe up or down menu and release the finer, it seems animation hit the wall (slow quickly), then continue to rest
so how can I fix the issue?
I have tried the whole day but I couldn't fix it
The documentation for the UISpringTimingParameters says:
https://developer.apple.com/documentation/uikit/uispringtimingparameters/1649832-init
A vector with a magnitude of 1.0 corresponds to an initial velocity that would cover the total animation distance in one second. For example, if the total animation distance is 200 points and the view’s initial velocity is 100 points per second, specify a vector with a magnitude of 0.5.
Meaning that you have to normalize the velocity using the with of the view.
And looking at the official documentation for CGVector the initializer you are using is confusingly not documented.
https://developer.apple.com/documentation/coregraphics/cgvector
What I've ended up doing was, calculating the normalized vector by myself.
You would need to calculate the total points the view is moving from the start of the animation to the end and then using this distanceToMove to make a "unit vector" from it / normalizing it:
let distanceToMove = newY - oldY
let velocity = recognizer.velocity(in: view)
let relativeVelocityY = velocity.x / distanceToMove
let relativeVelocity = CGVector(dx: 0, dy: relativeVelocityY)
let timing = UISpringTimingParameters(dampingRatio: 0.9, initialVelocity: relativeVelocity)
Let me know if this worked for you.
Related
I'm having a hard time setting boundaries and positioning camera properly inside my view after panning. So here's my scenario.
I have a node that is bigger than the screen and I want to let user pan around to see the full map. My node is 1000 by 1400 when the view is 640 by 1136. Sprites inside the map node have the default anchor point.
Then I've added a camera to the map node and set it's position to (0.5, 0.5).
Now I'm wondering if I should be changing the position of the camera or the map node when the user pans the screen ? The first approach seems to be problematic, since I can't simply add translation to the camera position because position is defined as (0.5, 0.5) and translation values are way bigger than that. So I tried multiplying/dividing it by the screen size but that doesn't seem to work. Is the second approach better ?
var map = Map(size: CGSize(width: 1000, height: 1400))
override func didMove(to view: SKView) {
(...)
let pan = UIPanGestureRecognizer(target: self, action: #selector(panned(sender:)))
view.addGestureRecognizer(pan)
self.anchorPoint = CGPoint.zero
self.cam = SKCameraNode()
self.cam.name = "camera"
self.camera = cam
self.addChild(map)
self.map.addChild(self.cam!)
cam.position = CGPoint(x: 0.5, y: 0.5)
}
var previousTranslateX:CGFloat = 0.0
func panned (sender:UIPanGestureRecognizer) {
let currentTranslateX = sender.translation(in: view!).x
//calculate translation since last measurement
let translateX = currentTranslateX - previousTranslateX
let xMargin = (map.nodeSize.width - self.frame.width)/2
var newCamPosition = CGPoint(x: cam.position.x, y: cam.position.y)
let newPositionX = cam.position.x*self.frame.width + translateX
// since the camera x is 320, our limits are 140 and 460 ?
if newPositionX > self.frame.width/2 - xMargin && newPositionX < self.frame.width - xMargin {
newCamPosition.x = newPositionX/self.frame.width
}
centerCameraOnPoint(point: newCamPosition)
//(re-)set previous measurement
if sender.state == .ended {
previousTranslateX = 0
} else {
previousTranslateX = currentTranslateX
}
}
func centerCameraOnPoint(point: CGPoint) {
if cam != nil {
cam.position = point
}
}
Your camera is actually at a pixel point 0.5 points to the right of the centre, and 0.5 points up from the centre. At (0, 0) your camera is dead centre of the screen.
I think the mistake you've made is a conceptual one, thinking that anchor point of the scene (0.5, 0.5) is the same as the centre coordinates of the scene.
If you're working in pixels, which it seems you are, then a camera position of (500, 700) will be at the top right of your map, ( -500, -700 ) will be at the bottom left.
This assumes you're using the midpoint anchor that comes default with the Xcode SpriteKit template.
Which means the answer to your question is: Literally move the camera as you please, around your map, since you'll now be confident in the knowledge it's pixel literal.
With one caveat...
a lot of games use constraints to stop the camera somewhat before it gets to the edge of a map so that the map isn't half off and half on the screen. In this way the map's edge is showing, but the furthest the camera travels is only enough to reveal that edge of the map. This becomes a constraints based effort when you have a player/character that can walk/move to the edge, but the camera doesn't go all the way out there.
Our scene only has about 20 nodes. The code below lets the user pan, conducting a hit test on each pan -- the goal is to highlight blocks as the user pans around the screen.
However, it is noticeably sluggish on a iPhone 5S. It's not deterministic but happens often enough to be irritating (every 5-10 pans).
We considered using hitTestWithSegment since you could tightly bound the range for testing but believe that should be slower because you must first compute the two points required for the function.
Moreover, the SCNHitTestClipToZRangeKey option for hitTest should provide a comparable performance boost by tightening the hit range without requiring the computation of two additional points.
Any suggestions for speeding up the performance of hitTest?
func sceneViewPannedOneFinger(sender: UIPanGestureRecognizer) {
// Get pan distance & convert to radians
let translation = sender.translationInView(sender.view!)
var xRadians = GLKMathDegreesToRadians(Float(translation.x))
var yRadians = GLKMathDegreesToRadians(Float(translation.y))
// Get x & y radians
xRadians = (xRadians / 4) + curXRadians
yRadians = (yRadians / 4) + curYRadians
// Limit yRadians to prevent rotating 360 degrees vertically
yRadians = max(Float(-M_PI_2), min(Float(M_PI_2), yRadians))
// Set rotation values to avoid Gimbal Lock
cameraNode.rotation = SCNVector4(x: 1, y: 0, z: 0, w: yRadians)
userNode.rotation = SCNVector4(x: 0, y: 1, z: 0, w: xRadians)
// Save value for next rotation
if sender.state == UIGestureRecognizerState.Ended {
curXRadians = xRadians
curYRadians = yRadians
}
// Set preview block
setPreviewBlock(sender)
}
private func setPreviewBlock(recognizer: UIGestureRecognizer) {
let point = recognizer.locationInView(sceneView)
let options = [SCNHitTestRootNodeKey: sceneView.scene!.rootNode, SCNHitTestClipToZRangeKey: 15, SCNHitTestSortResultsKey: true]
let hits = sceneView.hitTest(point, options: options)
print(hits.first?.worldCoordinates)
}
I'm trying to rotate an SCNBox I created using swipe gestures. For example, when I swipe right the box should rotate 90degs in the Y-axis and -90degs when I swipe left. To achieve this I have been using the node's SCNAction.rotateByX method to perform the rotation animation. Now the problem I'm having is when rotating along either the X-axis or Z-axis after a rotation in the Y-axis and vice-versa is that the positions of the axes change.
What I have notice is that any rotation perform on either of the X,Y,Z axes changes the direction in which the other axes point.
Example: Default position
Then after a rotation in the Z-axis:
Of course this pose a problem because now when I swipe left or right I no longer get the desire effect because the X-axis and Y-axis have now swapped positions. What I would like to know is why does this happen? and is there anyway to perform the rotation animation without it affecting the other axes?
I apologize for my lack of understanding on this subject as this is my first go at 3d graphics.
Solution:
func swipeRight(recognizer: UITapGestureRecognizer) {
// rotation animation
let action = SCNAction.rotateByX(0, y: CGFloat(GLKMathDegreesToRadians(90)), z: 0, duration: 0.5)
boxNode.runAction(action)
//repositoning of the x,y,z axes after the rotation has been applied
let currentPivot = boxNode.pivot
let changePivot = SCNMatrix4Invert(boxNode.transform)
boxNode.pivot = SCNMatrix4Mult(changePivot, currentPivot)
boxNode.transform = SCNMatrix4Identity
}
I haven't ran into any problems yet but it may be safer to use a completion handler to ensure any changes to X,Y,Z axes are done before repositioning them.
I had the same issue, here's what I use to give the desired behavior:
func panGesture(sender: UIPanGestureRecognizer) {
let translation = sender.translationInView(sender.view!)
let pan_x = Float(translation.x)
let pan_y = Float(-translation.y)
let anglePan = sqrt(pow(pan_x,2)+pow(pan_y,2))*(Float)(M_PI)/180.0
var rotVector = SCNVector4()
rotVector.x = -pan_y
rotVector.y = pan_x
rotVector.z = 0
rotVector.w = anglePan
// apply to your model container node
boxNode.rotation = rotVector
if(sender.state == UIGestureRecognizerState.Ended) {
let currentPivot = boxNode.pivot
let changePivot = SCNMatrix4Invert(boxNode.transform)
boxNode.pivot = SCNMatrix4Mult(changePivot, currentPivot)
boxNode.transform = SCNMatrix4Identity
}
}
I want to create a particle system on iOS using sprite kit where I define the colour of each individual particle. As far as I can tell this isn't possible with the existing SKEmitterNode.
It seems that best I can do is specify general behaviour. Is there any way I can specify the starting colour and position of each particle?
This can give you a basic idea what I was meant in my comments. But keep in mind that it is untested and I am not sure how it will behave if frame rate drops occur.
This example creates 5 particles per second, add them sequentially (in counterclockwise direction) along the perimeter of a given circle. Each particle will have different predefined color. You can play with Settings struct properties to change the particle spawning speed or to increase or decrease number of particles to emit.
Pretty much everything is commented, so I guess you will be fine:
Swift 2
import SpriteKit
struct Settings {
static var numberOfParticles = 30
static var particleBirthRate:CGFloat = 5 //Means 5 particles per second, 0.2 means one particle in 5 seconds etc.
}
class GameScene: SKScene {
var positions = [CGPoint]()
var colors = [SKColor]()
var emitterNode:SKEmitterNode?
var currentPosition = 0
override func didMoveToView(view: SKView) {
backgroundColor = .blackColor()
emitterNode = SKEmitterNode(fileNamed: "rain.sks")
if let emitter = emitterNode {
emitter.position = CGPoint(x: CGRectGetMidX(frame), y: CGRectGetMidY(frame))
emitter.particleBirthRate = Settings.particleBirthRate
addChild(emitter)
let radius = 50.0
let center = CGPointZero
for var i = 0; i <= Settings.numberOfParticles; i++ {
//Randomize color
colors.append(SKColor(red: 0.78, green: CGFloat(i*8)/255.0, blue: 0.38, alpha: 1))
//Create some points on a perimeter of a given circle (radius = 40)
let angle = Double(i) * 2.0 * M_PI / Double(Settings.numberOfParticles)
let x = radius * cos(angle)
let y = radius * sin(angle)
let currentParticlePosition = CGPointMake(CGFloat(x) + center.x, CGFloat(y) + center.y)
positions.append(currentParticlePosition)
if i == 1 {
/*
Set start position for the first particle.
particlePosition is starting position for each particle in the emitter's coordinate space. Defaults to (0.0, 0,0).
*/
emitter.particlePosition = positions[0]
emitter.particleColor = colors[0]
self.currentPosition++
}
}
// Added just for debugging purposes to show positions for every particle.
for particlePosition in positions {
let sprite = SKSpriteNode(color: SKColor.orangeColor(), size: CGSize(width: 1, height: 1))
sprite.position = convertPoint(particlePosition, fromNode:emitter)
sprite.zPosition = 2
addChild(sprite)
}
let block = SKAction.runBlock({
// Prevent strong reference cycles.
[unowned self] in
if self.currentPosition < self.positions.count {
// Set color for the next particle
emitter.particleColor = self.colors[self.currentPosition]
// Set position for the next particle. Keep in mind that particlePosition is a point in the emitter's coordinate space.
emitter.particlePosition = self.positions[self.currentPosition++]
}else {
//Stop the action
self.removeActionForKey("emitting")
emitter.particleBirthRate = 0
}
})
// particleBirthRate is a rate at which new particles are generated, in particles per second. Defaults to 0.0.
let rate = NSTimeInterval(CGFloat(1.0) / Settings.particleBirthRate)
let sequence = SKAction.sequence([SKAction.waitForDuration(rate), block])
let repeatAction = SKAction.repeatActionForever(sequence)
runAction(repeatAction, withKey: "emitting")
}
}
}
Swift 3.1
import SpriteKit
struct Settings {
static var numberOfParticles = 30
static var particleBirthRate:CGFloat = 5 //Means 5 particles per second, 0.2 means one particle in 5 seconds etc.
}
class GameScene: SKScene {
var positions = [CGPoint]()
var colors = [SKColor]()
var emitterNode: SKEmitterNode?
var currentPosition = 0
override func didMove(to view: SKView) {
backgroundColor = SKColor.black
emitterNode = SKEmitterNode(fileNamed: "rain.sks")
if let emitter = emitterNode {
emitter.position = CGPoint(x: frame.midX, y: frame.midY)
emitter.particleBirthRate = Settings.particleBirthRate
addChild(emitter)
let radius = 50.0
let center = CGPoint.zero
for var i in 0...Settings.numberOfParticles {
//Randomize color
colors.append(SKColor(red: 0.78, green: CGFloat(i * 8) / 255.0, blue: 0.38, alpha: 1))
//Create some points on a perimeter of a given circle (radius = 40)
let angle = Double(i) * 2.0 * Double.pi / Double(Settings.numberOfParticles)
let x = radius * cos(angle)
let y = radius * sin(angle)
let currentParticlePosition = CGPoint.init(x: CGFloat(x) + center.x, y: CGFloat(y) + center.y)
positions.append(currentParticlePosition)
if i == 1 {
/*
Set start position for the first particle.
particlePosition is starting position for each particle in the emitter's coordinate space. Defaults to (0.0, 0,0).
*/
emitter.particlePosition = positions[0]
emitter.particleColor = colors[0]
self.currentPosition += 1
}
}
// Added just for debugging purposes to show positions for every particle.
for particlePosition in positions {
let sprite = SKSpriteNode(color: SKColor.orange, size: CGSize(width: 1, height: 1))
sprite.position = convert(particlePosition, from: emitter)
sprite.zPosition = 2
addChild(sprite)
}
let block = SKAction.run({
// Prevent strong reference cycles.
[unowned self] in
if self.currentPosition < self.positions.count {
// Set color for the next particle
emitter.particleColor = self.colors[self.currentPosition]
// Set position for the next particle. Keep in mind that particlePosition is a point in the emitter's coordinate space.
emitter.particlePosition = self.positions[self.currentPosition]
self.currentPosition += 1
} else {
//Stop the action
self.removeAction(forKey: "emitting")
emitter.particleBirthRate = 0
}
})
// particleBirthRate is a rate at which new particles are generated, in particles per second. Defaults to 0.0.
let rate = TimeInterval(CGFloat(1.0) / Settings.particleBirthRate)
let sequence = SKAction.sequence([SKAction.wait(forDuration: rate), block])
let repeatAction = SKAction.repeatForever(sequence)
run(repeatAction, withKey: "emitting")
}
}
}
Orange dots are added just for debugging purposes and you can remove that part if you like.
Personally I would say that you are overthinking this, but I might be wrong because there is no clear description of what you are trying to make and how to use it. Keep in mind that SpriteKit can render a bunch of sprites in a single draw call in very performant way. Same goes with SKEmitterNode if used sparingly. Also, don't underestimate SKEmitterNode... It is very configurable actually.
Here is the setup of Particle Emitter Editor:
Anyways, here is the final result:
Note that nodes count comes from an orange SKSpriteNodes used for debugging. If you remove them, you will see that there is only one node added to the scene (emitter node).
What you want is completely possible, probably even in real time. Unfortunately to do such a thing the way you describe with moving particles as being a particle for each pixel would be best done with a pixel shader. I don't know of a clean method that would allow you to draw on top of the scene with a pixel shader otherwise all you would need is a pixel shader that takes the pixels and moves them out from the center. I personally wouldn't try to do this unless I built the game with my own custom game engine in place of spritekit.
That being said I'm not sure a pixel per pixel diffusion is the best thing in most cases. Expecially if you have cartoony art. Many popular games will actually make sprites for fragments of the object they expect to shader. So like if it's an airplane you might have a sprite for the wings with perhaps even wires hanging out of this. Then when it is time to shatter the plane, remove it from the scene and replace the area with the pieces in the same shape of the plane... Sorta like a puzzle. This will likely take some tweaking. Then you can add skphysicsbodies to all of these pieces and have a force push them out in all directions. Also this doesn't mean that each pixel gets a node. I would suggest creatively breaking it into under 10 pieces.
And as whirlwind said you could all ways get things looking "like" it actually disintegrated by using an emitter node. Just make the spawn area bigger and try to emulate the color as much as possible. To make the ship dissappear you could do a fade perhaps? Or Mabye an explosion sprite over it? Often with real time special effects and physics, or with vfx it is more about making it look like reality then actually simulating reality. Sometimes you have to use trickery to get things to look good and run real-time.
If you want to see how this might look I would recommend looking at games like jetpac joyride.
Good luck!
I am building a ios game with swift and I have run into a bit of a problem. I am trying to spawn balls from the top of the screen and have them come down towards the ground. They are supposed to have random x values and go down at random rates but instead of spawning on the screen the nodes spawn on an x value which is not encompassed by the screen. Please help me as I think I have done everything right.
Here is the code for my addball function...
func addBall(){
//create ball sprite
var ball = SKSpriteNode(imageNamed: "ball.png")
//create physics for ball
ball.physicsBody = SKPhysicsBody(rectangleOfSize: ball.size) // 1
ball.physicsBody?.dynamic = true // 2
ball.physicsBody?.categoryBitMask = PhysicsCategory.Ball // 3
ball.physicsBody?.contactTestBitMask = PhysicsCategory.Person & PhysicsCategory.Ground
ball.physicsBody?.collisionBitMask = PhysicsCategory.None // 5
//generate random postion along x axis for ball to spawn
let actualX = random(min:ball.frame.size.width/2+1, max: self.frame.size.width - ball.frame.size.width/2-1)
println(actualX)
//set balls positon
ball.position = CGPoint(x: actualX, y: size.height - ball.size.width/2)
//add ball to scene
addChild(ball)
//determine speed of ball
let actualDuration = random(min: CGFloat(3.0), max: CGFloat(5.0))
//create movement actions
let actionMove = SKAction.moveTo(CGPoint(x:actualX, y: -ball.size.width/2), duration: NSTimeInterval(actualDuration))
let actionMoveDone = SKAction.removeFromParent()
ball.runAction(SKAction.sequence([actionMove, actionMoveDone]), withKey: "action")
}
here is the code for my random functions
func random() -> CGFloat {
return CGFloat(Float(arc4random()) / 0xFFFFFFFF)
}
func random(#min: CGFloat, max: CGFloat) -> CGFloat {
return random() * (max - min) + min
}
The problem here is that your SKScene likely takes up much more space than the screen of your device. Thus, when you calculate a random value using the whole scene, some of the time the ball will spawn in the area of the scene not visible to you.
The two main properties that control the scene's size are its size and scaleMode properties. The scaleMode property relates to how the scene is mapped. Unless you initialized and presented this scene yourself, you can check the scaleMode in your view controller. It will likely be set to aspectFill, which according to Apple means:
The scaling factor of each dimension is calculated and the larger of the two is chosen. Each axis of the scene is scaled by the same scaling factor. This guarantees that the entire area of the view is filled but may cause parts of the scene to be cropped.
If you don't like this, there are other scaleModes. However, in most cases this mode would actually be preferable since SpriteKit's internal scaling is able to make universal apps. If this is fine for you, then the easiest thing to do is set hardcoded values for something like the spawn locations for your ball node.