• Thanks: 33

Topic: What about SVP *after* madVR?

While playing with AviSynth to optimize videos, here's what I found.

At least for low quality videos, I'm getting MUCH better results if I first upscale/process/denoise the video, and THEN increase the frame rate to 60 at the end.

SVP, however, does the opposite. It first increases to 60fps and then sends it to madVR.

What if we were to do the opposite?

First, SVP would have higher-quality frames to work with.

Second, madVR would have more time to process each frame = higher quality (41ms per frame on a 24fps video instead of 16ms with 60fps)

Why are you doing SVP first?

  • Thanks: 37

Re: What about SVP *after* madVR?

It would be redundant, because in the end you still need the video renderer again after SVP.

If madVR first instead of SVP, it would be like this madVR->SVP->renderer(such as Overlay Mixer, EVR, madVR, etc). CMIIW

Re: What about SVP *after* madVR?

ffdshow has a lot of tools for "first upscale/process/denoise the video"

  • Thanks: 33

Re: What about SVP *after* madVR?

so... would that be possible or not? and would it be a good idea or not?