I know it's late in the Western Hemisphere, so I ran 4 quick producing runs using the Best Matched Format from the Profile Analyzer.
There weren't any obviously brighter clips to my eyes, but when I laid them all on the timeline and masked off and staggered the leftmost portion of each clip, it became obvious that hardware encoding did in fact noticeably brighten the produced video:
The SVRT clip is (and should be) identical to the source, and the CPU-produced clip also matches as far as brightness is concerned.
However, both the QuickSync and nVidia-produced clips are clearly brighter, so whatever is causing the issue seems to be with how PD hands off the encoding to the GPU, rather than there being an issue with the actual hardware or its associated driver. If that's the case, I imagine videos produced with AMD GPUs would also be too bright.
I've packed the project with all my test clips, and you can download everything from this OneDrive
folder.
I think this should be reported to Cyberlink, and I don't know if it's related to the latest patch or not. Maybe someone with the previous version can produce a GPU-encoded version and see.
Cyberlinks22490, please contact CL tech support
here, and be sure to include a link to this thread so they can see what's happened and also download the relevant clips.
I'm puzzled as to why AshWilliams' suggestion didn't make a difference, though, so a workaround might be to use the NewBlue Gamma Corrector FX in combination with the Fix/Enhance tool.
I can exactly match everything except the black of the boat on a GPU-produced clip to the original if I set the Fix/Enhance Exposure setting to 86, and then set the NBGC FX Expand setting at 136 and Compress to 0. You may want to play around to get just the right look.
YouTube/optodata
DS365 | Win11 Pro | Ryzen 9 3950X | RTX 4070 Ti | 32GB RAM | 10TB SSDs | 5K+4K HDR monitors
Canon Vixia GX10 (4K 60p) | HF G30 (HD 60p) | Yi Action+ 4K | 360Fly 4K 360°