So, we use ARFaceTrackingConfiguration and ARKit for a magic mirror like experience in our apps, augmenting users faces with digital content. On the iPad Pro 5gen customers are complaining that the camera image is too wide, I'm assuming that is because of the new wide-angle camera necessary for Apples center-stage Facetime calls? I have looked through Tracking and Visualizing Faces and the WWDC 2021 videos, but I must have missed any API's that allow us to disable the wide-angle feature on the new iPads programmatically?

Working with a M1 Macbook Air, macos 12.4. Anytime I open a new terminal window or just a new tab, it takes a really long time till I can type. I have commented out my entire ~/.zshrc and when run for i in $(seq 1 10); do /usr/bin/time $SHELL -i -c exit; done directly in an open terminal window it says it finished in 0.1s. So it must be something macos is doing before zsh is even starting. PS: In the activity monitor I can only see a spike in kernel_task cpu usage when opening a new terminal

So, I'm using some small python script that imports from pxr import Usd, Sdf, UsdGeom, Kind which used to work fine a few months ago. I got the USD tools from https://developer.apple.com/augmented-reality/tools/ and always had to run it in a Rosetta-Terminal (plus making sure the USD tools are in PATH and PYTHONPATH) but it worked. Now, whatever I do I always get   File "create_scene.py", line 2, in <module>     from pxr import Usd, Sdf, UsdGeom, Kind   File "/Applications/usdpython/USD/lib/python/pxr/Usd/__init__.py", line 24, in <module>     import _usd ModuleNotFoundError: No module named '_usd' Does anyone have any clues what this _usd is about?

I added a basic Hello World SwiftUI view to an existing UIKit project, yet I can not get the preview to work. Usual error is: MessageSendFailure: Message send failure for send render message to agent ================================== | RemoteHumanReadableError: Could not connect to agent | | Bootstrap timeout after 8.0s waiting for connection from 'Identity(pid: 30286, sceneIdentifier: Optional("XcodePreviews-30286-133-static"))' on service com.apple.dt.uv.agent-preview-service Neither this nor the generated report is very helpful. I also created a new Xcode project to see if the same View works in a new test project, which it does. If my project compiles without warnings and errors, but SwiftUI preview fails, what are the options I have left? (My deployment target is IOS14, my Xcode is the fresh Xcode 13.0)

Aloha Quick Lookers, I'm using the usdzconvert preview 0.64 to create *.usdz files, which I then edit in ascii *.usda format, and then recipe them as *.usdz. This way I was able to fix the scale (the original gltf's usually are in m=1, while the usdzconvert 0.64 always sets the result to m=0.01). Now I was trying to follow the docs to anchor my Glasses prim on the users face and whatever I try, it will only ever place it on my tables surface. If I import my *.usdz file into Reality Composer and export it to usdz It does get anchored to the face correctly. Now when I open the Reality-Composer-Export-usdz it really doesn't look so different from my manually edited usdz (it just wraps the Geometry in another layer, I assume because the import-export through Reality-Composer). What am I doing wrong? Here's the Reality Composer generated usda: #usda 1.0 ( autoPlay = false customLayerData = { string creator = "com.apple.RCFoundation Version 1.5 (171.5)" string identifier = "9AAF5C5D-68AB-4034-8037-9BBE6848D8E5" } defaultPrim = "Root" metersPerUnit = 1 timeCodesPerSecond = 60 upAxis = "Y" ) def Xform "Root" { def Scope "Scenes" ( kind = "sceneLibrary" ) { def Xform "Scene" ( customData = { bool preliminary_collidesWithEnvironment = 0 string sceneName = "Scene" } sceneName = "Scene" ) { token preliminary:anchoring:type = "face" quatf xformOp:orient = (0.70710677, 0.70710677, 0, 0) double3 xformOp:scale = (1, 1, 1) double3 xformOp:translate = (0, 0, 0) uniform token[] xformOpOrder = ["xformOp:translate", "xformOp:orient", "xformOp:scale"] // ... and here is my own , minimal, usdz with the same face-anchoring-token: #usda 1.0 ( autoPlay = false customLayerData = { string creator = "usdzconvert preview 0.64" } defaultPrim = "lupetto_local" metersPerUnit = 1 timeCodesPerSecond = 60 upAxis = "Y" ) def Xform "lupetto_local" ( assetInfo = { string name = "lupetto_local" } kind = "component" ) { def Scope "Geom" { def Xform "Glasses" { token preliminary:anchoring:type = "face" double3 xformOp:translate = (0, 0.01799999736249447, 0.04600000008940697) uniform token[] xformOpOrder = ["xformOp:translate"] // ... I'd add the full files, but they are 2Mb of size and the max file size is 200kb. I could create a minimal example file with less geometry in case the above code is not enough.

In our app, we're streaming short HLS streams to local AVPlayers. In the view we have an AVPlayerLayer which is connected to the AVPlayer we start, and usually we hear audio and see the video. Sometimes, instead of video, we'll see a grey screen, while still being able to hear the audio. The playerlayer will be a shade of grey, which is not a shade coming from our app. Also if for testing purposes we don't connect the player to the PlayerLayer, this grey color will not be there, so it's definitely coming from the AVPlayerLayer. If this is somehow related to an HLSStream becoming corrupted or something, what are the API's in AVFoundation we could use to debug this? So far, when this happens we see no peculiar catch blocks or errors being thrown in our debug logs.

In an app with multiple AVPlayers (think TikTok UX for example), where: • one video is currently streaming • other videos , that users will move to next, have already been loaded into an AVPlayer for prebuffering Is there an API to give the currently playing Stream higher priority than the pre-buffering ones that are not on screen? It's nice to have instant play while scrolling but we also don't want to starve the currently playing player of bandwidth.

So, I've been following along the video on "Distribute binary frameworks as Swift packages" - https://developer.apple.com/videos/play/wwdc2020/10147/, but now I'm sort of stuck. I have my Swift Package that works fine when used as a source package, which I want to ship as a binary now. The video says, I'm supposed to add a target: swift import PackageDescription let package = Package(     name: "Test",     defaultLocalization: "de",     platforms: [         .iOS(.v13)     ],     products: [         .library(             name: "Test",             targets: ["Test"]         ),     ],     dependencies: [     ],     targets: [ //        .target(name: "Test")         .binaryTarget(             name: "Test",             url: "https://static.looc.io/Test/Test-1.0.0.xcframework.zip",             checksum: "9848327892347324789432478923478"         )     ] ) but, the xcframework is what I am trying to build, don't have an xcframework yet? With the above, if I run: bash [konrad@iMac-2 Source]$ xcodebuild archive -workspace Test -scheme Test \archivePath "tmp/iOS" \ destination "generic/platform=iOS" \ SKIP_INSTALL=NO BUILD_LIBRARY_FOR_DISTRIBUTION=YES I get the error bash xcodebuild: error: Could not resolve package dependencies:   artifact of binary target 'Test' failed download: invalid status code 403