Motion Smoothing in OXR is not only not what OXR is meant for,
Notice the word “style” and think the implications of it. I didn’t mean that the implemention should be implemented in OpenXR.
The current fragmented implementions of motion smoothing isn’t helping the industry as evident by this thread. For example there’s a lot of HMDs that can’t tap on the 3 major vendor VR compositors and thus being stuck on a very basic implementation of motion smoothing. It’s all software side so why not give everyone the same benefits?
There is some speculation happening here, but we’re wondering why Steam motion smoothing appears to be more robust than the WMR implementation. Steam and Oculus are both using this for motion smoothing:
NVIDIA’s Turing GPUs introduced a new hardware functionality for computing optical flow between images with very high performance.
This is essentially hardware accelerated motion smoothing. The next sentence:
The Optical Flow SDK 1.0 enables developers to tap into the new optical flow functionality.
You ask what I need to be satisfied like I’m being an insufferable princess, but all that’s being suggested here is that WMR is lagging behind because they’re quite literally leaving performance on the table. Perhaps 20% or more.
@Altsak if you check out the article linked above, it appears we already have this!
Are you though? Seems pretty on-track to me. I think you’re just more diplomatic than I am
Yes, I really have no idea either. But I did a bunch of practice carrier landings in the F-14 last night and so far so good. I’ve had a couple of moments where the tracking gets a little shaky, but it seems to correct itself quickly.
Just for supersampling reasons, I’m rendering at 3k * 3k pixels. The Index isn’t able to show all these pixels by a long shot. Means I can keep settings the same and get a big bump in sharpness just by changing HMDs
