I have the exact same annoying messages...

And finally, it seems that CGImageRef cgImage = imageRep.CGImage; destroys the original representation, as CGImageGetBitsPerComponent(cgImage) returns 8 (initially 10) and CGImageGetBitsPerPixel(cgImage) 32 (initially 40).

And a direct construction of the CGImage via: CGImageSourceRef imageSource = CGImageSourceCreateWithURL((CFURLRef)url, nil); CGImageRef image = CGImageSourceCreateImageAtIndex(imageSource, 0, NULL); contains 8bits components too...

Sorry, I meant wantsExtendedDynamicRangeContent and not maximumExtendedDynamicRangeColorComponentValue

MTKView has a colorspace property. And to paraphrase Apple documentation : The default value is nil, indicating that the rendered content isn’t color-matched. If you set this to a different color space, Core Animation performs any necessary color transformations when compositing the view’s contents.

I have exactly the same error on a M1 Max (MacBook Pro 16") : Execution of the command buffer was aborted due to an error during execution. Invalid Resource (00000009:kIOGPUCommandBufferCallbackErrorInvalidResource) There is no crash, only a slowdown when the log is displayed.

Thanks Ceylo, I'll give it a try, the next time I suspect such a situation! If applicable, in the case I detect any remaining activity, could it be attached to a killed application? An in this case, would it not be occuring below any monitoring?

You can use __fp16 or _Float16 data types. __fp16 is not an arithmetic data type and should be used for storage purposes. It allows conversion from and to floats. _Float16 is an arithmetic data type and have supports for standard operations. _Float16 is more modern and a better choice. You can set a half constant with the f16 suffix following your real number.

I suggest you to read the comprehensive article "Optimizing Parallel Reduction in Metal for Apple M1" from Matthew Kieber-Emmons : https://kieber-emmons.medium.com/optimizing-parallel-reduction-in-metal-for-apple-m1-8e8677b49b01

Oh, I see! Thanks, noah-w! That's a nice change... But why not do the same thing with colorAttachments?

OK... I see... Thanks!

That makes sense... Thanks a lot !