26 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dlr5668 wrote:
RickyAstle98 wrote:
dlr5668 wrote:

You can edit source scripts in SVP folder. Not many have 240 hz OLED for it to be useful

The OLED? The BFI still benefits for any panel types, my chinese 180Hz IPS sad
How to edit source script to make every other frame black? thanks!

Even SVP dev cant do it in 1 try. Dump SVP 4\script into chatgpt/claude and start debugging until you get it right or bored. Thats how I would start

Actually that easy to implement as I said earlier vf-append/vf-fps or deinterlace filter, without feature primary target, put black PNG over and over, devs can do that too easy, I dont use AI debugger sorry!

Go to forum: Using SVP

27 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dlr5668 wrote:
RickyAstle98 wrote:

Any possibility to make black frame insertion like vf-append or deinterlace filter that doubles the FPS? I mean what IF any other black/grey frame for motion debluring?

You can edit source scripts in SVP folder. Not many have 240 hz OLED for it to be useful

The OLED? The BFI still benefits for any panel types, my chinese 180Hz IPS sad
How to edit source script to make every other frame black? thanks!

Go to forum: Using SVP

28 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

Any possibility to make black frame insertion like vf-append or deinterlace filter that doubles the FPS? I mean what IF any other black/grey frame for motion debluring?

Go to forum: Using SVP

29 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dlr5668 wrote:
RickyAstle98 wrote:

I forgot that I used 2x2 all the time

How do you do that? I use only v2 models (for better rewind speed) and wanna try. How does it affect model performance?

What do you mean?

Go to forum: Using SVP

30 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

Drakko01 wrote:
RickyAstle98 wrote:

Whats the default align/valign values for SVP4 Pro? I forgot that I used 2x2 all the time, maybe thats why all models starting 4.16 looks fragmentary smooth!

Fragmentary smooth - when a small parts of one large object can be interpolated differently, even being a same large object! For example: faces, 4.4 to 4.16 faces looks buttery smooth, from 4.17 to 4.26 small parts like lips, cheeks, ears and other objects can be separated from the frame! There were cases when the lips moved apart alone, never happens for 4.4 to 4.16 before, also happens when the main object in the frame is interpolated correctly, but all the background moving both axis, looks like one step forward-two steps back frame!

The 4.26 new model is worse than 4.15 lite in Matrix rooftop 3 agents opening the doors, pattern errors on door textures, no door errors in 4.15v2 (both mod32 and mod64 input)!

Too noticeable door errors 4.4 (worst)
Less noticeable door errors 4.9 (okay)
No door errors 4.11 to 4.16 (perfect)

For testing what you stated, can you provide or indicate the exact scene.thanks

Nevermind, just say default align/valign values, thanks! The rendered results will look different on different PCs!

Go to forum: Using SVP

31 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dawkinscm wrote:
RickyAstle98 wrote:

Whats the default align/valign values for SVP4 Pro? I forgot that I used 2x2 all the time, maybe thats why all models starting 4.16 looks fragmentary smooth!

Fragmentary smooth - when a small parts of one large object can be interpolated differently, even being a same large object! For example: faces, 4.4 to 4.16 faces looks buttery smooth, from 4.17 to 4.26 small parts like lips, cheeks, ears and other objects can be separated from the frame! There were cases when the lips moved apart alone, never happens for 4.4 to 4.16 before, also happens when the main object in the frame is interpolated correctly, but all the background moving both axis, looks like one step forward-two steps back frame!

The 4.26 new model is worse than 4.15 lite in Matrix rooftop 3 agents opening the doors, pattern errors on door textures, no door errors in 4.15v2 (both mod32 and mod64 input)!

Too noticeable door errors 4.4 (worst)
Less noticeable door errors 4.9 (okay)
No door errors 4.11 to 4.16 (perfect)

Are you talking about 4.26 lite only because the lite models do behave differently to the normal ones.

The 4.26 lite? I dont see 4.26 lite model, where is it? Youre wrong!

Go to forum: Using SVP

32 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

Whats the default align/valign values for SVP4 Pro? I forgot that I used 2x2 all the time, maybe thats why all models starting 4.16 looks fragmentary smooth!

Fragmentary smooth - when a small parts of one large object can be interpolated differently, even being a same large object! For example: faces, 4.4 to 4.16 faces looks buttery smooth, from 4.17 to 4.26 small parts like lips, cheeks, ears and other objects can be separated from the frame! There were cases when the lips moved apart alone, never happens for 4.4 to 4.16 before, also happens when the main object in the frame is interpolated correctly, but all the background moving both axis, looks like one step forward-two steps back frame!

The 4.26 new model is worse than 4.15 lite in Matrix rooftop 3 agents opening the doors, pattern errors on door textures, no door errors in 4.15v2 (both mod32 and mod64 input)!

Too noticeable door errors 4.4 (worst)
Less noticeable door errors 4.9 (okay)
No door errors 4.11 to 4.16 (perfect)

Go to forum: Using SVP

33 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

oriento wrote:

4.22 lite is out !

Smoother than any other lite models, except 4.15 (for me atleast)
TRT 10.3 / opt_level 2 / engine built average time - 26 seconds (720p)
Different behavior with different aspect ratios and containers!

Go to forum: Using SVP

34 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

frc.threads - concurrent frames
mpv_add - buffered frames

Go to forum: Using SVP

35 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dawkinscm wrote:
RickyAstle98 wrote:
dawkinscm wrote:

As he said above, on average he got around an 8 fps improvement. But this stat should be taken in the context of dev notes that suggest there might be be specific situations where the performance might be slightly worse.

My actual scenarios
v14test3 320fps / v15.2 310fps (4.4v2 with 1.78 dynamic shape and 320 fps with 2.35 dynamic shape)
Starting models 4.16 everything gets worse, just a performance drops variables from 6 to 12% respectively, the movie dynamic shapes always faster (even when the targeted resolution is slightly higher)

So it works better with Performance Boost disabled? Am I understanding this correctly?

Yes and no! Easier explanation - depending on the aspect ratio of the sources, depending on the resolution and the model used, performance can be either increased to 2% or degraded to 12% with or without TRT boost option! Seek performance is night and day with boost enabled! I did not regret that I switched to a new library!
So - with 15.2 you gets maximum 12% performance degradation, but practically is just 6% of that!

Go to forum: Using SVP

36 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dawkinscm wrote:
flowreen91 wrote:
aloola wrote:

v15.2 gives me a tiny FPS boost
v14test3 ~150fps vs ~ v15.2 ~158fps, also fixed dynamic shape (Performance boost: off) bug

15.2 changelog:

    Upgraded to TensorRT 10.3.0.

    Fixed performance regression of RIFE and SAFA models starting with vs-mlrt v14.test4. This version may still be slightly slower than vs-mlrt v14.test3 under some conditions, however.

Looking forward to @aloola benchmarks for the new TensorRT like he did before:
https://www.svp-team.com/forum/viewtopi … 324#p84324

As he said above, on average he got around an 8 fps improvement. But this stat should be taken in the context of dev notes that suggest there might be be specific situations where the performance might be slightly worse.

My actual scenarios
v14test3 320fps / v15.2 310fps (4.4v2 with 1.78 dynamic shape and 320 fps with 2.35 dynamic shape)
Starting models 4.16 everything gets worse, just a performance drops variables from 6 to 12% respectively, the movie dynamic shapes always faster (even when the targeted resolution is slightly higher)

Go to forum: Using SVP

37 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

Xenocyde wrote:
RickyAstle98 wrote:

New TensorRT library!
https://github.com/AmusementClub/vs-mlr … a.v15.2.7z

Worth installing over 9.2?

According to my realtime tests and encoding tests, the difference is starting to increase from 4.16 models!
The old models like 4.4 (v1/v2) encoding speed with 2% performance difference and maximum to 12% for newest models!
Fixed my issues? Nope, but now models look smoother a little bit! Seek performance? Variable!
Worth? If you want, you can test, for me new models looks smoother, what happens to you - who knows?!

Go to forum: Using SVP

38 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

New TensorRT library!
https://github.com/AmusementClub/vs-mlr … a.v15.2.7z

Go to forum: Using SVP

39 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

Insindro wrote:
RickyAstle98 wrote:
Blackfyre wrote:

Massive difference in size! wow hopefully I can run 4K 2x with a 3090, but I highly doubt it. Usually I have to wait for the Lite v2, but even then, not all of them can run.

Idk why but everything after 4.16 v2 always not smooth, now even container fix doesnt affect 4.20v1v2 models! RTX 4070 whats wrong?

Container fix?

For example I have some videos that not smooth 4.10+ models with mkv container (this issue affects all mkv containers), mp4 container fixes this issue sometimes!

Go to forum: Using SVP

40 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

Blackfyre wrote:
dawkinscm wrote:

Rife v4.20 has been released. I didn't do a large amount of testing but I did check the few remaining artefacts I know of and there doesn't seem to be a clear  difference to v4.15/v4.18 on my system.

Update: Apparently v4.20 is great when using 16x interpolation with Anime. I would be interested to hear if that is helpful for anyone.

Massive difference in size! wow hopefully I can run 4K 2x with a 3090, but I highly doubt it. Usually I have to wait for the Lite v2, but even then, not all of them can run.

Idk why but everything after 4.16 v2 always not smooth, now even container fix doesnt affect 4.20v1v2 models! RTX 4070 whats wrong?

Go to forum: Using SVP

41 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

donnieyeen wrote:

Did anybody try to manually update the TRT libraries to v15.1: latest TensorRT library? Any benefits over the standard TRT install from SVP?

TRT library 9.2 (4070) : 4.15v2 lite = 720p@24 x7 without drops
TRT library 15 : 4.15v2 lite = 720p@24 x5 with drops

Go to forum: Using SVP

42 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

Xenocyde wrote:

Guys, do I need to disable hwdec=auto-copy in config? MPV got updated and it reset my config file.

Actually no, auto-copy better for RIFE procedures, for me for example (RTX 4070 user)

Go to forum: Using SVP

43 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dawkinscm wrote:

OK so after more testing my current settings are:

TensorRT: 9.2
SC: Image Detect
SCT: 25%
AdjFra: Blend

Disabled similar to when previously set to SCT: 100%
SVP motion vectors work nicely except for Hugo.
NVOP was the worse in previous testing and nothing has changed.

Idk how you test, everything works as intended, check my previous comment, everything over 15/25% is similiar to look!

Go to forum: Using SVP

44 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dawkinscm wrote:
Chainik wrote:

SVP 4.6.0.272

1. added motion vectors-based scene change detection for RIFE
2. configurable "decrease to..." function (on the "Frame size" page) - selectable target resolution and downscaling filter

Thanks for the update. I decided to completely reinstall everything to remove any changes I made. The results so far are that
- SVOP and Image Comparison and disabled give very similar results.
- NVOF might be a little worse overall but still similar.
- This is the same as before except that all of them feel a little less smooth than the previous setup with SC set to 100.
- Setting SC to 100 with this new software seems to be similar to the others as well. Although when I add back the MVP interpolation code that helps. I also put back the TensorRT9.2 files.

Let's see what others say.

Everything works when SC threshold lower than 15% (yes), NVOF handles fast paced scenes a lot better, SVP vectors just duplicate frames, Image Comparison almost the same with dump method (over 15%)
Over 25% every methods handles frame almost the same (as dumb/comparison method)

Go to forum: Using SVP

45 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

StanTheMan wrote:

Hey, guys!
Users of 4070 Super and 4080 Super, please respond, how's the RIFE doing?
Currently i have 3060 Ti and plan to upgrade. Ultimately, i want to upgrade to 4080 Super, but the most real scenario that i will upgrade to 4070 Super (not even Ti), because of budget. What about fps that 4070 Super or 4080 Super can provide? Is it handle 4K or not? Also, what about GPU Super Resolution upscale in the Media Player Classic (MPC Render), can it handle? I use Super Res in the level 1 (of 4), it's great performance with picture quality 1080->4K, but i want the full 4 levels. I watch only in 60 fps.

As 4070 user almost 8 month, tested too many times, can it handle 4K? For example my GPU can run 2X@4K up to 4.9v2 models (struggles), Super almost the same, so yes, can handle (4.10+ models are more agressive)! Dont know too much about GPU Super Resolution, but it doesnt helps with target framerates imho! You watch only in 60FPS, 4K 24>60 is bad idea even for older models, its uses more resources than 3X factor! I downscale movies to HD and then 6X because of monitor! I think 4K>2K/2.5K will be enough for 24>60 procedures, without significant image quality drop!

Go to forum: Using SVP

46 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dawkinscm wrote:
flowreen91 wrote:
Xenocyde wrote:

Unfortunately I can still see microstuttering

I noticed that if i exit specific background apps like TeamViewer, i suddenly get 20-30 extra fps with SVP.
I believe TeamViewer reads the memory and causes a bottleneck even when it's minimized in taskbar.

Xenocyde try to close your 200 Chrome tabs and other background apps and see if u find the specific app that causes that microstuttering. It's good to know.
Try it with the 4.17 lite version cause it's easier on the GPU:
https://github.com/AmusementClub/vs-mlr … nal-models

The v4.17 lite version is an experiment and from what I see, it's worse than even v4.17.

As originally stated, the lite versions have lower accuracy of temporal calculations, so from some of my early tests I noticed artifacts where even older versions would have better!
For example : 4.16v2 still can produce wavy zigzag artefacts in specific pattern areas, while 4.9v2 artefact does no exist (only SPECIFIC areas)!

Go to forum: Using SVP

47 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

flowreen91 wrote:

67% GPU sounds good. Don't think it's a resource issue then.
V2 models give much better fps than V1 models. So u should see significant improvements from using them.

Xenocyde wrote:

Yes, vsync is always on.

Try to set rife_sc to 100 in SVP so it will not cause microstuttering from repeating frames when a scene change is detected.

Xenocyde, could vsync syncing cause microstuttering then? Cause every time i try to test with video-sync=display-resample it breaks the soap opera effect on panning scenes.
Try to test with it temporarily disabled by adding these in mpv config too:

# D3D11 renderer (default) is required for the HDR playback
gpu-api=d3d11
# disable VSync
video-sync=audio
d3d11-sync-interval=0


Removing vsync wait times from the equation should allow the RIFE generated frames to be displayed realtime. Even if u set it to 120 fps on a 60 fps capped monitor. xD

Sadly there is still no solution for microstuttering in general, its not produced by RIFE itself, its something else, happens!

Go to forum: Using SVP

48 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

dawkinscm wrote:
Xenocyde wrote:

I have an RTX 4080 and I only play 1080p TV series or 4K downscaled to 1080p @2.5X fps with RIFE 4.15. Maximum GPU utilization I see is around 67%. Microstutters are not too often, maybe once every 15 minutes at most.
...
There was a time when microstutters did not occur at all, but who even knows what version of RIFE or what drivers I was using back then.

The occasional micro-stutter isn't a major issue of course, but it's surprising that your GPU is hitting 67% with downscaling. For me it's around 30% with the same GPU. I use mpv player.

Propably yes, but sometimes overlays shows false % usage, 4.9v2 shows 88% for 3X when I can 6X without drops, 70% SVP RTX users use HAGS, thats why usage is higher!

Go to forum: Using SVP

49 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

Chainik wrote:

> instead finding rife_sc all the time?

just don't close the window big_smile

sad

Go to forum: Using SVP

50 Reply by RickyAstle98

(2,112 replies, posted in Using SVP)

Chainik wrote:

> Then how to make proper frame blending when RIFE_SC threshold over 15?

this is why you need a MV-based scene change detection instead of dumb image comparison

Last question, how to add custom scene change values into SVP UI instead finding rife_sc all the time?

Go to forum: Using SVP