Search options (Page 5 of 9)
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!
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)
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!
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)!
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!
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!
Chainik wrote:> instead finding rife_sc all the time?
just don't close the window 
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?
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
I did already, I fix this issue, tooks about 5 minutes to find out, anyway thanks!
Update : I also noticed seek performance drop, about 2x times with NVOF vectors used!
pensioner600 wrote:I also noticed that ALL versions of V2 give the entire picture a shake, which is very clearly visible on a static scene on small light objects on a dark background. I noticed a long time ago that something was wrong, but I didn’t have time to watch movies. Versions NOT V2 and lite no jitter. I’m definitely switching to 15lite or 16lite, even if it’s x3, not x4. V2 is complete nonsense. Today I devoted half a day to tests, a lot became clear))
Picture shake happens depending on the model and fps targets, if you let RIFE profile do 60fps exact, picture shake may occur, also different TRT behave differently!
Example : 4.4v2 fixed 60 fps 25>60 (picture shake) / 4.4v2 2.5x 25>62.5 (picture shake fixed) / 4.9v2 (no picture shakes)
I have no idea if you see picture shake on 2x 3x 4x factors or something, I dont see picture shake, I use TRT 9.2 library!
RAGEdemon wrote:RickyAstle98 wrote:RAGEdemon wrote:Bro, you can easily get 4K RIFE: use 4.4 to double your fps and sync to 48Hz refresh (for 24fps source). Honestly, it looks butter smooth. I would be hard-pressed to notice this vs 60/120. The magic is ensuring the output FPS matches your refresh.
I know about my 4070 capabilities, but in 4.4 too much artefacts (not critical but noticeable), secondly I dont need sync anything, GSync matches movies automatically! No magic at all!
My standard solutions for movies >
1) 4.4v2 1280x720 24>144
2) 4.9v2 1280x720 24>144
3) 4.15v2l 1280x720 24>144
Each iteration after 4.4 causes a massive performance hit @~16% for minimal artefact improvement. Above 4.9 is trained mostly on Anime.
What you say doesn't make sense to me. You say you don't use 4.4 because "too much artefacts (not critical but noticeable)", but then you are happy to force downscale all 4K and 1080p content to 1280x720p?! And you don't notice the huge degradation in quality?
I am guessing you don't have a 4K display, or else you wouldn't be saying something so seemingly bizarre 
To each their own, I guess...
Who said my monitor is 4K? I dont watch 4K at all, my monitor is standard chinese Full HD 165Hz GSync with builtin upscalers, I just test 4K source for performance, hehe! I dont see massive quality degradation only because of monitor upscalers, but my solutions shown are exclusively for 6x movie fps targets, obviously I watch 1080p with 3x for sure!
RAGEdemon wrote:RickyAstle98 wrote:Blackfyre wrote:All I done was disable hardware-accelerated GPU scheduling. Restarted the PC, and it built after.
Didn't help though, none of 4.15 versions or 4.16 work at 4K without dropped frames on my RTX 3090
Something is wrong, because 4.16 lite working at 4K with my RTX 4070 without dropped frames
But 4.15v2 (nonlite) is very strong, I down 2160p to 1800p - works!
Bro, you can easily get 4K RIFE: use 4.4 to double your fps and sync to 48Hz refresh (for 24fps source). Honestly, it looks butter smooth. I would be hard-pressed to notice this vs 60/120. The magic is ensuring the output FPS matches your refresh.
I know about my 4070 capabilities, but in 4.4 too much artefacts (not critical but noticeable), secondly I dont need sync anything, GSync matches movies automatically! No magic at all!
My standard solutions for movies >
1) 4.4v2 1280x720 24>144
2) 4.9v2 1280x720 24>144
3) 4.15v2l 1280x720 24>144
Blackfyre wrote:Drakko01 wrote:I honestly dont remenber what I changed in the old vs-mlrt. Because I run into this problem before
All I done was disable hardware-accelerated GPU scheduling. Restarted the PC, and it built after.
Didn't help though, none of 4.15 versions or 4.16 work at 4K without dropped frames on my RTX 3090
Something is wrong, because 4.16 lite working at 4K with my RTX 4070 without dropped frames
But 4.15v2 (nonlite) is very strong, I down 2160p to 1800p - works!
dawkinscm wrote:RickyAstle98 wrote:dawkinscm wrote:I wouldn't recommend ever using a denoiser unless you have poor quality sources. You say "without quality loss" but with any denoiser there is always a risk of losing high frequency detail.
This type of denoisers is without quality loss...
Which one is it? The Open Intel Denoiser?
Yes, propably, I need denoiser for 720p sources, I dont prefer adjust sharpness level while watching 24>168 4.15v2lite!
dawkinscm wrote:RickyAstle98 wrote:Anybody has mpv shader like Intel denoiser (which makes all videos looks pretty without quality loss)?
I wouldn't recommend ever using a denoiser unless you have poor quality sources. You say "without quality loss" but with any denoiser there is always a risk of losing high frequency detail.
This type of denoisers is without quality loss...
Anybody has mpv shader like Intel denoiser (which makes all videos looks pretty without quality loss)?
Any possibility to RTX SR into mpv?
dawkinscm wrote:Now that I'm back on TRT8.5.1 I am able to run Rife v4.15v2 at 72fps which uses about 75-90% GPU. The lite version uses about 60-80% GPU. For me I suppose that it goes without saying that:
4.15v2/72 > 4.15v2 lite/72 > 4.15v2/60 > 4.15v2/60 lite > everything else.
FHD or 4K?
dawkinscm wrote:RickyAstle98 wrote:You didnt understand my post, its only realtime playback bug, the models are normal, just a MPV bug, which I can fix just reencode source files! Thats because I dont give rendered example, models frame handling broken only with realtime playback! Anyway I found a fix, maybe someone facing this bug, just reencode files, voila! And how strange its sounds, I record realtime playback, and there I didnt see issue, thats why I dont give examples, even recording ontop!
So maybe I also misunderstood. I thought the video would show the artefacts but I see none. Are you saying that I have to play the clip on PC using Rife?
1) Shared video isnt test video
2) No you can play clip on anything
3) This video showing exactly how issue looks like
4) Nevermind
Posts found: 101 to 125 of 213