Hi there, I've been using SVP for years and really appreciate the transcoding feature.
I have a pretty high-end system but even it can't handle 4k video without toning some of the settings down, so I've resorted to transcoding them instead. I also want to export high-fps videos to view on my phone and tablet. I've done a lot of reading online but I was hoping for some advice.
Does transcoding use the exact same motion interpolation as on-the-fly upscaling, or does it use a more demanding approach?
Also, are there any hidden options/ovcopts arguments/overrides to boost motion interpolation and prediction, even if they're completely impractical and end up taking twice as long to transcode? Upscaling 4k video to double or triple the frame rate takes ages anyway so I don't mind the extra time.
Given that I'm transcoding, should I max everything out, or are there some settings that are pointless or will just increase artifacting in 4k videos.
Is there any way to implement some of MadVR's own intensive upscaling (like chroma upscaling, noise/banding reduction) in the encode itself, or are there any other ways to implement these enhancements in the encode itself?
Should I always upscale to an exact multiple of the frame rate, like from 24 to 48fps, or do uneven multiples not matter?
Finally, is there any point upscaling all the way up to 144/165hz or should I just aim for 48/60?
Sorry for the barrage of questions. Any input would be appreciated.