Hi there, I'm developing a visionOS app that is using the anchor points and mesh from SceneReconstructionProvider anchor updates. I load an ImmersiveSpace using a RealityView and apply a ShaderGraphMaterial (from a Shader Graph in Reality Composer Pro) to the mesh and use calls to setParameter to dynamically update the material on very rapid frequency. The mesh is locked (no more updates) before the calls to setParameter. This process works for a few minutes but then eventually I get the following error in the console: assertion failure: Index out of range (operator[]:line 789) index = 13662, max = 1 With the following stack trace: Thread 1 Queue : com.apple.main-thread (serial) #0 0x00000002880f90d0 in __abort_with_payload () #1 0x000000028812a6dc in abort_with_payload_wrapper_internal () #2 0x000000028812a710 in abort_with_payload () #3 0x0000000288003f40 in _os_crash_msg () #4 0x00000001dc9ff624 in re::ecs2::ComponentBucketsBase::addComponent () #5 0x00000001dc9ffadc in re::ecs2::ComponentBucketsBase::moveComponent () #6 0x00000001dc8b0278 in re::ecs2::MaterialParameterBlockArrayComponentStateImpl::processPreparingComponents () #7 0x00000001dc8b05e4 in re::ecs2::MaterialParameterBlockArraySystem::update () #8 0x00000001dd008744 in re::Scheduler::executePhase () #9 0x00000001dc032ec4 in re::Engine::executePhase () #10 0x0000000248121898 in RCPSharedSimulationExecuteUpdate () #11 0x00000002264e488c in __59-[MRUISharedSimulation _doJoinWithConnectionContext:error:]_block_invoke.44 () #12 0x0000000268c5fe9c in _UIUpdateSequenceRunNext () #13 0x00000002696ea540 in schedulerStepScheduledMainSectionContinue () #14 0x000000026af8d284 in UC::DriverCore::continueProcessing () #15 0x00000001a1bd4e6c in CFRUNLOOP_IS_CALLING_OUT_TO_A_SOURCE0_PERFORM_FUNCTION () #16 0x00000001a1bd4db0 in __CFRunLoopDoSource0 () #17 0x00000001a1bd44f0 in __CFRunLoopDoSources0 () #18 0x00000001a1bd3640 in __CFRunLoopRun () #19 0x00000001a1bce284 in _CFRunLoopRunSpecificWithOptions () #20 0x00000001eff12d2c in GSEventRunModal () #21 0x00000002697de878 in -[UIApplication _run] () #22 0x00000002697e33c0 in UIApplicationMain () #23 0x00000001b56651e4 in closure #1 (Swift.UnsafeMutablePointer<Swift.Optional<Swift.UnsafeMutablePointer<Swift.Int8>>>) -> Swift.Never in SwiftUI.KitRendererCommon(Swift.AnyObject.Type) -> Swift.Never () #24 0x00000001b5664f08 in SwiftUI.runApp<τ_0_0 where τ_0_0: SwiftUI.App>(τ_0_0) -> Swift.Never () #25 0x00000001b53ad570 in static SwiftUI.App.main() -> () () #26 0x0000000101bc7b9c in static MetalRendererApp.$main() () #27 0x0000000101bc7bdc in main () #28 0x0000000197fd0284 in start () Any advice on how to solve this or prevent the error? Thanks!

I have been using ARKit to get hand tracking data on a continuous loop by implementing the AnchorUpdateSequence. I want to try out the .predicted hand tracking, but it seems as though using ARKit session and HandTrackingProvider do not allow me to enable this feature?

Platform: visionOS 2.6 Framework: RealityKit, SwiftUIComponent: ImagePresentationComponent I’m working with the new ImagePresentationComponent from visionOS 26 and hitting a rendering limitation when switching to .spatialStereoImmersive viewing mode within a WindowGroup context. This is what I’m seeing: Pure immersive space: ImagePresentationComponent with .spatialStereoImmersive mode works perfectly in a standalone ImmersiveSpace Mode switching API: All mode transitions work correctly (logs confirm the component updates) Spatial content: .spatialStereo mode renders correctly in both window and immersive contexts. This is where it’s breaking for me: Window context: When the same RealityView + ImagePresentationComponent is placed inside a WindowGroup (even when that window is floating in a mixed immersive space), switching to .spatialStereoImmersive mode shows no visual change The API calls succeed, state updates correctly, but the immersive content doesn’t render. Apple’s Spatial Gallery demonstrates exactly what I’m trying to achieve: Spatial photos displayed in a window with what feels like horizontal scroll view using system window control bar, etc. Tapping a spatial photo smoothly transitions it to immersive mode in-place. The immersive content appears to “grow” from the original window position by just changing IPC viewing modes. This proves the functionality should be possible, but I can’t determine the correct configuration. So, my question to is: Is there a specific RealityView or WindowGroup configuration required to enable immersive content rendering from window contexts that you know of? Are there bounds/clipping settings that need to be configured to allow immersive content to “break out” of window constraints? Does .spatialStereoImmersive require a specific rendering context that’s not available in windowed RealityView instances? How do you think Apple’s SG app achieves this functionality? For a little more context: All viewing modes are available: [.mono, .spatialStereo, .spatialStereoImmersive] The spatial photos are valid and work correctly in pure immersive space Mixed immersive space is active when testing window context No errors or warnings in console beyond the successful mode switching logs I’m getting Any insights into the proper configuration for window-hosted immersive content

Any way to extend the video recording time in Reality Composer Pro from 3:00 to any longer value, such as editing preferences in Terminal or other workaround? Is there any way to use the strap and a USB-C cable as a live video stream input source that would mirror to Quicktime or some other video capture tool? I am assuming there is no online documentation or user manual for the strap, but please correct me if I'm wrong. Thank you.

hi, I'm trying to create a virtual movie theater, but after running computeDiffuseReflectionUVs.py and applying attenuation map, I noticed the light falloff effect just covers over the objects. I used apple provided attenuation map (did not specify the attenuation map name on python script) with sample size of 6000. I thought the python script would calculate vertices and create shadow for, say, back of the chairs. Am I understanding this wrong?

One of the most common ways to provide a window size in visionOS is to use the defaultSize scene modifier. WindowGroup(id: "someID") { SomeView() } .defaultSize(CGSize(width: 600, height: 600)) Starting in visionOS 26, using this has a side effect. visionOS 26 will restore windows that have been locked in place or snapped to surfaces. If a user has manually adjusted the size of a locked/snapped window, the users size is only restore in some cases. Manual resize respected Leaving a room and returning later Taking the headset off and putting it back on later Manual resize NOT respected Device restart. In this case, the window is reopened where it was locked, but the size is set back to the values passed to defaultSize. The manual resizing adjustments the user has made are lost. This is counter to how all other windows and widgets work. I reported this last month (FB18429638), but haven't heard back if this is a bug or intended behavior. Questions What is the best way to provide a default window size that will only be used when opening new windows–and not used during scene restoration? Should we try to keep track of window size after users adjust them and save that somewhere? If this is intended behavior, can someone please update the docs accordingly?

I'm seeing this error while attempting to compile my VisionOS app under Xcode 26. My existing code looks like: let (naturalSize, formatDescriptions, mediaCharacteristics) = try? await videoTrack.load(.naturalSize, .formatDescriptions, .mediaCharacteristics) This is now giving a compiler error: Type of expression is ambiguous without a type annotation I don't see that anything that was changed or deprecated in the latest version. Also loading the properties individually seems to work fine i.e.: let naturalSize = try? await videoTrack.load(.naturalSize) let formatDescriptions = try? await videoTrack.load(.formatDescriptions) let mediaCharacteristics = try? await videoTrack.load(.mediaCharacteristics)

As in the title: openImmersiveSpace works as expected. The ImmersiveSpace in the ID opens normally but the function never resolves a Result. Here is a workaround I used to make user it was on the UI thread and scenePhase was active: ` @MainActor func openSpaceWithStateCheck() async { if scenePhase == .active { Task { switch await openImmersiveSpace(id: "RoomCaptureInteraction") { case .opened: isCapturingImagery = true break case .error: print("!! An error occurred when trying to open the immersive space captureRoomImagery") case .userCancelled: print("!! The user declined opening immersive space captureRoomImagery") @unknown default: print("!! unknown default result of opening space") break } } } else { print("Scene not active, deferring immersive space opening") } } I'm on visionOS 2.4 and SDK 2.2. I have tried uninstalling the app and rebuilding. Tried simply opening an empty ImmersiveSpace. The consistency of the ImmersiveSpace opening at least means I can work around it. Even dismissImmersiveSpace works normally and closes the immersive space. But a workaround seems hamfisted.

SharePlay objects are not placed in the same place in the same space. I hope they can be placed in the same place. （Vision Pro）

I would like to translate info in a three.js based web app as a 3D model in a volumetric window. Is it possible to do this in a similar manner as loading a web page in a WKWebView?

We got very excited when we saw support for the PSVR2 on WWDC! Particularly interesting is WebXR to us, so we got the controllers to give it a try. Unfortunately they only register as gamepads in the navigator but not as XRInputDevice's We went through the experimental flags and didn't find something that is directly related. Is there a flag we missed? If not, when do you have PSVR2 support planned for WebXR?

Has anyone had success with MeshInstancesComponent? I tried to follow the sample code from What's New in RealityKit but it wouldn't compile. I was able to use one of the init overloads to get it to compile, but using it crashes both my device and the simulator. Even with one instance.

I want to display a model image in the windowGroup window. This image is not unique. To display the model image, how should I convert it into an image

Hello, I am developing a visionOS application and am interested in obtaining detailed data of users’ hands through ARKit, including but not limited to Transform and rotation angle. I have reviewed Happy Beem, but it appears to only introduce the method of identifying the user’s specific gestures. Could you please advise on how to obtain the Transform and rotation angle of the user’s hand? Thank you.

Hi all, I'm currently developing a real-time object reconstruction app using ARKit. The goal is to scan large objects using ARKit’s depth and transform data, and generate a point cloud. However, I’m facing a major challenge - Transform Drift / World Alignment Issues The localToWorld transform provided by ARKit frequently seems to drift or become unstable across frames. This results in misaligned point clouds even when the device is moved slowly or kept relatively still. In some cases, a static surface scanned over a few seconds results in clearly misaligned fragments. This makes it difficult to accurately stitch a multi-frame point cloud. I have experimented with various lighting conditions and object textures, but the issue persists in all cases. At times, the relative error between frames reaches up to 20 cm, while in other instances the error is minimal; however, the drift gradually accumulates over time, leading to an overall enlargement of the reconstructed object. I have attached images of both cases here. Questions: Are there specific conditions under which ARKit’s world transform is expected to drift? Is there a way to detect or recover from this drift during runtime? Any best practices for maintaining consistent tracking during scanning or measurement sessions?

Hi everyone, I'm developing a MR Vision Pro app where I’d like to anchor virtual objects (such as UI elements) around the user's arm. However, I’ve noticed that Vision Pro seems to mask out the area where the user’s real arm is, hiding virtual content in that region so that you see your real arm. Is there a way to render virtual elements on the user's arm—so that it looks like the object is placed directly on the arm despite the real-world passthrough? I was hoping there might be a way to adjust the depth or behavior of this masked-out region. Any insights or workarounds would be greatly appreciated! Thanks :)

Hi, I'm working on a VisionOS app and would like to integrate Background Assets to download large files after the app is installed. I'm wondering what would happen if the user takes off the headset while a background asset is being downloaded. Would it continue downloading or would the download be stopped/paused? I was looking for a way to download large assets while the user is not wearing the Vision Pro, is there any other alternative? Thanks in advance.

Hello everyone I would like to create my own spatial video on my Apple Vision Pro. According to all the documentation from Apple, this requires two camera angles that enhance the spatial perception. I have purchased the Enterprise license with main camera access for this purpose. However, this only gives me access to the left main camera of the glasses. Is there a way to access the right camera as well? Or is the one camera image enough to create a spatial video by splitting the image, for example? I am open to any help and ideas. My goal is to create the video with the cameras on the glasses, not externally.

In visionOS, I'm trying to create an immersive environment which would feature several spheres in which immersive movies are visible. I'm starting from a sample code which creates a sphere, sets an immersive movie as its material, and opens it as an immersive environment. This works fine. But if I create a sphere in an open immersive environment using Reality Composer Pro and sets its material to an immersive movie, I can see the movie on the sphere while I move outside of it but if I try to get inside the sphere, it disappears. What would be the right way of doing this ?

是对原本VisionPro每个App的内存限制做了扩展嘛？放宽了内存限制么？