topic Re: visionOS 3D-Tiles in Swift Maps SDK Questions

visionOS 3D-Tiles

DavidCredo — Fri, 06 Dec 2024 15:01:08 GMT

Hi!

Really exciting that you're supporting visionOS now, thank you for your great work!
My understanding is that there's currently no support for 3D-Tiles on visionOS at the moment.
I am currently trying to build an application that leverages 3D tiles and i would really like to evaluate your framework for this.

So my question is:
Is it possible to utilise 3D-tiles on visionOS at the moment and if not, is this planned for a future release?

Thanks in advance,

David

Re: visionOS 3D-Tiles

ZacharyKline — Mon, 09 Dec 2024 20:26:24 GMT

Hello,

Thank you for sharing your excitement for visionOS support. We are excited too!

Can you clarify what type of experience you are expecting for your app? Currently, we do support 3D tiles in visionOS in a 2D window using the OGC3DTilesLayer. We have an example of that layer being used here. I can't say much about what to expect in the future other than we are looking into and hoping to expand our support beyond the 2D window.

Re: visionOS 3D-Tiles

DavidCredo — Tue, 10 Dec 2024 06:38:53 GMT

Hello,

Thank you for the quick response!
2D Window support is great, however, my use-case requires visionOS Volumes.
The Table-Top View most closely resembles what i am going for.
If there's any way i can help migrate existing code for visionOS, I'd be happy to help.

Kind regards
David

Re: visionOS 3D-Tiles

ZacharyKline — Tue, 10 Dec 2024 19:44:19 GMT

Thank you that makes sense. I'll keep note of it.

In the meantime, I know this doesn't fit your exact use case but it may be worth noting. You can create your own tabletop map for visionOS by attaching one of our views to a `RealityView` via this constructor. That won't provide the true 3D experience you are wanting with the layers but may be something worth checking out in the meantime.

Re: visionOS 3D-Tiles

youknowww — Tue, 06 Jan 2026 06:41:41 GMT

Thanks @ZacharyKline that was a great starting point. If anyone, like me, comes across this years later here's some starting code to get this going with attachments. You'll see I have a little drag handler component to move my map around in the RealityView like its a normal window. The advantage of using this instead of normal window is now I can manipulate it like an Entity and set up collision with other custom Entities. There's more to it but this is a cleaned up section because so much of my code is in prototyping at the moment.

Also found this code pretty helpful as a great starting point for simple attachments:

https://github.com/satoshi0212/visionOS_30Days/blob/main/Day28/Day28/View/ImmersiveView.swift

@State private var dragStartPos: SIMD3<Float>? @State private var mapDragStartPos: SIMD3<Float>? var body: some View { RealityView { content, attachments in if let mapView = attachments.entity(for: "map_viewer"), let handleView = attachments.entity(for: "drag_handle"){ // Position in front of user at eye level mapView.position = SIMD3(x: 0, y: 1.5, z: -1.5) // Position handle below the map (in local coordinates of map) handleView.position = SIMD3(x: 0, y: -0.31, z: 0.01) handleView.components.set(InputTargetComponent()) // try using collision component with filter to track activation gesture (pinch) var collision = CollisionComponent(shapes: [.generateBox(width: 0.2, height: 0.08, depth: 0.05)]) collision.filter = CollisionFilter(group: DragHandle.collisionGroup, mask: .all) handleView.components.set(collision) mapView.addChild(handleView) content.add(mapView) } } attachments: { Attachment(id: "map_viewer") { MapViewer() .frame(width: 1200, height: 800) .glassBackgroundEffect() } Attachment(id: "drag_handle") { DragHandleView() } } .gesture( DragGesture() .targetedToAnyEntity() // prevent direct touch .handActivationBehavior(.pinch) .onChanged { value in guard value.entity.components[CollisionComponent.self]?.filter.group == DragHandle.collisionGroup, let parent = value.entity.parent, let grandparent = parent.parent else { return } let currentPos = value.convert(value.location3D, from: .local, to: grandparent) // First frame: capture starting positions if dragStartPos == nil { dragStartPos = currentPos mapDragStartPos = parent.position } // Calculate delta from drag start and apply to original entity position if let dragStart = dragStartPos, let mapDragStart = mapDragStartPos { let delta = currentPos - dragStart parent.position = mapDragStart + delta } } .onEnded { _ in // Reset for next drag dragStartPos = nil mapDragStartPos = nil } ) }