I found related resources in WWDC.
According to https://developer.apple.com/wwdc16/606?time=869 and https://developer.apple.com/wwdc20/10631?time=1481, they are using simdgroup_barrier() if thread group fits in a single SIMD group because threadgroup_barrier() is more expensive than simdgroup_barrier().
simdgroup size is normally 32 (afaik 64 in some AMD gpus) so I think simdgroup_barrier() can be used if thread group size is smaller than 32 or 64.
Prin_E
Users receive reputation points by contributing quality content to the forums.
Posts include post and replies published on the forums.
Post authors can mark replies as Accepted to indicate successful solutions.
Apple Developer Forums admins can mark replies as Apple Recommended to indicate an approved solution
Last seen
User for
Post
Replies
Boosts
Views
Activity
The sample runs on Monterey beta 7&8! Thanks!!
Oh, I figure it out there's development team property in Build Settings - Signing. I set it to my team and the sample works! ☺️
There seem to be fixed in Big Sur beta3 (20A5323l). The noisy lines are disappeared in my machine. (UHD 630)
I added color texture (R16) in the shadow pass and used color textures in the shading pass. It works well but it has some extra performance cost...I asked Apple through TSI 6 days ago but I haven't received any response yet! 😠
I'm very looking forward to Apple's feedback! 😁Please let me know!
Hi,I have troubles running ModernRendering on my MacBook Pro.When I run on Radeon Pro 5500M, the sample shows nothing excepts lighting effects and when I capture GPU frames, the system hangs and finally get kernel panic :-(It runs pretty well on Intel Graphics 630. But I have to make some tweaks to resolve some texture loading issues.MacBook Pro 16 inch 2019, macOS 10.15.2