Search options (Page 3 of 7)

Could be one of few reasons:

1. SVP not actually being applied.
2. Frame drops (SVP Index not stable --> Weak hardware or settings cranked too high).
3. You have strongest artifact masking, or other weird settings maybe interfering?

Other reasons maybe. We need more details to determine the issue.

cemaydnlar wrote:

I have no idea if it works or not. It says estimated fps 48. When i try to do 60fps or above it gets laggy. I couldn't see a difference between x2 and normal video playback. x2 is just a little bit smoother. I get 0 arifacts with x2.

If at x2 is a little bit smoother it means that interpolation is working and this should be the case for video with 8-bit colour depth. At the moment only one person with a GeForce 30 series graphics card has done a thorough test on a 1080p file: https://www.svp-team.com/forum/viewtopi … 477#p80477 and based on the fps we can assume that at the moment x2 for 1080p for your card is the maximum. If you want more: x3 or x4 then 720p files are left on the ground.

Perhaps there will be some optimisation of RIFE in the future. In any case, I estimate that I will need 116% more powerful graphics card than the ZOTAC GAMING GeForce RTX 3070 Ti AMP Extreme Holo to interpolate x5 1080p video files: https://www.svp-team.com/forum/viewtopi … 480#p80480 I suppose this will be more than possible for NVIDIA GeForce RTX 4090 graphics cards.

Oh well... artificial intelligence comes at a price.

In the image you posted above:

Pixel Format: p010

Try to use 8-bit colour depth video.

cemaydnlar wrote:

Ok i'll give rife a try. Is a rtx 3070 enough for x2 ?

Yes, as long as it is video with no more than 1080p resolution, 8-bit colour depth without HDR

cemaydnlar wrote:

I am selecting rife interpolation but my fps stays at 24 ? What am i doing wrong ? Can you help me use rife?

Read this first:
https://www.svp-team.com/wiki/RIFE_AI_interpolation

Set the following values:

AI model: Generic (v4)
GPU threads: 2
GPU device: https://www.svp-team.com/forum/viewtopi … 170#p80170
RIFE via CUDA: Off

If you continue to have problems read what problems others had and how they dealt with them:

whole threads:
https://www.svp-team.com/forum/viewtopic.php?id=6580
https://www.svp-team.com/forum/viewtopic.php?id=6553

also read my thread from Chainik's post of April 28 to the end, everything should work from then on:
https://www.svp-team.com/forum/viewtopi … 360#p80360

skip it altogether: == RIFE / PyTorch installation == it's too slow now
https://www.svp-team.com/forum/viewtopi … 372#p80372

You asked the same question almost 3 years ago and got plenty of answers: https://www.svp-team.com/forum/viewtopic.php?id=5500

You spoke up when others asked about similar problems: https://www.svp-team.com/forum/viewtopic.php?id=6139

Others have asked before you and someone even stated that he preferred FrameRateConverter over SVP for this reason: https://www.svp-team.com/forum/viewtopic.php?id=4382

I just showed You that RIFE beats both mvtools2 and FrameRateConverter in what you asked about.

It can be seen that You care a lot about this and over the years you have not found a solution that satisfies you.

Do you think there is some magical way to make the base SVP algorithm capable of approaching what artificial intelligence can achieve? 

Base SVP interpolation uses only a tiny fraction of the computing power that RIFE requires.

You can spend hours testing every possible setting in SVP yourself, or you can use that time to do extra work that will allow you to upgrade your GPU. You can also take a shortcut, although I don't recommend it: https://www.svp-team.com/forum/viewtopi … 388#p80388

If you want to get a really good quality with the least amount of artifacts possible use this: https://www.svp-team.com/wiki/RIFE_AI_interpolation

About halo effect around moving objects read here:
http://forum.doom9.net/showpost.php?p=1 … stcount=17
http://forum.doom9.net/showpost.php?p=1 … stcount=24

About anime read this: https://www.svp-team.com/forum/viewtopi … 308#p80308

Fast paced scenes are difficult for any model. Specifically created model for this is here:  https://github.com/google-research/frame-interpolation

If you want the best possible quality for anime, look for AI models with the best score in the ATD12K test.

If you want the best possible quality for fast paced scenes, look for AI models with the best score in the Xiph-2k test.

You can search for scores and compare them here among others:
https://arxiv.org/pdf/2204.03513.pdf
https://arxiv.org/pdf/2104.02495.pdf
https://arxiv.org/pdf/2202.04901.pdf

If you additionally want reasonably good speed and ease of use, then only SVP + RIFE filter for VapourSynth (ncnn Vulkan)

cemaydnlar wrote:

KnightRiderKARR wrote:

Changing block.w and block.h to 16 is good for 720p or lower - at 1080p i see more artifacts.
With ""smooth.mask.area"" and ""smooth.mask.area_sharp"" does reduce artifacts little better than "Artifacts Masking" from SVP Control Panel

After days of trying these are the best settings you will get for max smoothness min artifact.

Artifact masking disabled. Svp shader 21. Frame rate conversion to screen.

Invalid script.... I tried to update components but still invalid.....

If i remove those scripts then it works... :
analyse.refine[0].thsad = 2000;
analyse.refine[0].search.type = 4;
analyse.refine[0].search.distance = 2;
analyse.refine[0].search.satd = false;
analyse.refine[0].penalty.lambda = 10.0;
analyse.refine[0].penalty.lsad = 8000;
analyse.refine[0].penalty.pnew = 50;

I dont know why all analyse.refine[0] scripts not working on me......

levels.gpu = 0;     does good job on the edges but in few scenes the smoothness is not very good and CPU usage is very high

Overall is better with these custom scripts that we posted here -- i dont know if we can have the very smoothness like Optical Flow from Davinci Resolve but i think its very hard to do...

cemaydnlar wrote:

KnightRiderKARR wrote:

Yes "analyse.block.overlap =1" will decrease detail smoothness (i mean less accuracy on small objects like text when vinyl record is rotating) but if you disable the artifacts masking then you have overall better smoothness but little more artifacts.

Ok so here are my thoughts. Biggest difference happens when i select weakest artifact masking. So in order to get max smoothness you have to select disabled no matter what. With analyse.block.overlap setting and masking disabled it is good for artifacts but in my opinion you lose a lot of smoothness. When you don't change analyse.block.overlap then it starts to produce a lot of artifacts when you watch movies. Basically you have max smoothness but it is unwatchable. In the end i have no idea how you can combine standart analyse.block.overlap setting with disabled masking and still get as little artifacts as you get with analyse.block.overlap =1 setting. I will write it down if i find a solution for it.

Try add these:

smooth.mask.area            = 16;    (i use 8)
smooth.mask.area_sharp        = 0.70;

(its alternative artifacts masking)

This help to reduce artifacts but im not sure if does good job overall - for now i have better results with analyse.block.overlap = 2

cemaydnlar wrote:

Can we make it a little better? Did you find better values than those above?

analyse.main.search.coarse.distance  = -24;   (a little bit better but it needs more CPU - especially if scene is on forest)
analyse.block.h    = 16;  (better smooth but more artifacts)

For now i dont have better values....

I personally use "Artifacts masking = disabled"  and  "analyse.block.overlap    = 1"  because i have 144Hz monitor and i interpolate to 144fps (if you have 144Hz or more its better to interpolate as screen refresh rate)

> breath like sounds
Usually PSU sounds like breath if it works at high load.
So, check your PSU and change it to more powerful.

> setting that makes svp as smooth
Use this settings:

Frames interpolation mode: Uniform
Artifacts masking: Disabled
Processing of scene changes: Blend adjacent frames

All other settings are at your discretion.