v14 runtime + 4.14 v2 is perfection. Really nice update

Chainik wrote:

how exactly "a library that leverages NVIDIA RTX architecture for accelerating large language models (LLMs)" is connected with us?

My brain only saw "NVIDIA RTX architecture for accelerating" sorry

Drakko01 wrote:

For those who have more technical knowledge, this it's good for us, right?
https://www.google.com/amp/s/www.techpo … face%3famp

I saw this one too. It should come this month

4.9 v2 works just fine here

playzoffs wrote:

I thought we wont be able to watch HD real time before new RTX 6XXX series. I ll try with my 3070ti soon

this one can do 2k 24-48. If you watch 4k just use downscale in svp

Мелкий баг, mpv работает, а mpc-hc нет. Решено добавлением mpv в path

2060 эсс не купить? Вначале года за 15к брал на казахстан озоне. Отлично тянет rife 1080p x2. Всетаки райф почти без артефактов

4к можно немного даунскелануть (до2к) тогда тоже тянет норм

33

(2 replies, posted in Using SVP)

Are you watching integer multiplier? I like x2, x3 is too smooth for me and 24->60 fps is not smooth

durf19851 wrote:

Я слабо представляю выкачивание сериала например Наруто, который 800 серий. Часто сериал смотрится фоном, т.е. включил и он на сайте крутится на втором или третьем мониторе.
Как вариант можно попробовать подгружать в ваше приложение не потоковое видео, а торрент файл, который передаст список серий в ваш плагин и при выборе какой-то конкретной серии сам подгрузит его из пиров и начнет воспроизведение, когда подгрузится нужное количество в буфер, как будто вы открыли видео в ютубе и включили автовоспроизведение.  Это было бы более удобно.

Это торрсервер. Можно поставить на ноуте, телефоне, впске (120р в месяц) или локально на ПК. Ставите кеш 1-2 гб, скармливаете ему торент и смотрите с свп

Recently "upgraded" my HTPC to 2060S ($190 new). Works pretty well with 2k content x2 fps. Using V2 since its below 4K

36

(32 replies, posted in Using SVP)

cemaydnlar wrote:
AutoClickers wrote:
cemaydnlar wrote:

Is there a movie version of anime4k for 1080p that works the same without upscaling ?

MadVR or realEsgran

I downloaded realEsgran but i don't know how to use it. Do i need to delete anime4k or do they work seperatly ? Can you give another quick guide big_smile

Not possible. Its either svp or that

37

(32 replies, posted in Using SVP)

DragonicPrime wrote:

I just use Anime4k. Might not be as good as full upscalers, but it gets the job done for me and makes anime looks significantly better imo. here's the link in case anyone wants it. https://github.com/bloc97/Anime4K

Its pretty good. Just press ctrl+1 for 1080p anime and ctrl+2 or 3 for low quality old one

38

(32 replies, posted in Using SVP)

Just use nvidia super res. Its close enough

mpv can be too wild to config imo. Can you retest with MPC-HC ?

v2 doesnt work with performance boost = disabled. But aside from this its fine

41

(11 replies, posted in Using SVP)

New dev rife build will also reduce lag on rewind to zero

42

(11 replies, posted in Using SVP)

Rife kinda killed it (for video interpolation)

dawkinscm wrote:

Just updated to the latest pre-release TensorRT library. Start up time is now about 5 seconds and the seek is just shy of real time but very choppy so not worth the upgrade to the pre-release yet.

What do you mean start time? Once cached its below 2s for me

I thinks its worth including in SVP. Performance boost = false should call V1 model (for now) and true V2 model

https://i.imgur.com/5yQiUhR.mp4

zerosoul9901 wrote:
dlr5668 wrote:
Chainik wrote:

> while v2 models significantly reduces the playback seek time, improve performance

dunno, I can't see any improvements
not "significantly", not even slightly

----
unpack "rife_v2" into SVP 4\rife\models (i.e. there must be SVP 4\rife\models\rife_v2\rife_v4.6.onnx)
replace SVP 4\rife\vsmlrt.py

switch back to "v1" models - rename "rife_v2" folder to any other name, like "_rife_v2"

v2 takes 7.5/8 and crashes at seek, v1 stays at 3.5/8 as it should

Don't use dynamic shapes, it doesn't seem to work well with v2 models and requires more VRAM usage compared to v1, use static shapes instead, and use flag force_fp16 instead of fp16 during inference, reducing memory usage and engine build time during engine build, you can also add _implementation=2 flag to easily switch v2 models.

Thanks works great. Same VRAM requirements and much faster almost instant rewind

https://i.imgur.com/NM5g5Zk.png

Chainik wrote:

> while v2 models significantly reduces the playback seek time, improve performance

dunno, I can't see any improvements
not "significantly", not even slightly

----
unpack "rife_v2" into SVP 4\rife\models (i.e. there must be SVP 4\rife\models\rife_v2\rife_v4.6.onnx)
replace SVP 4\rife\vsmlrt.py

switch back to "v1" models - rename "rife_v2" folder to any other name, like "_rife_v2"

v2 takes 7.5/8 and crashes at seek, v1 stays at 3.5/8 as it should

Isnt ensemble twice as slow?

48

(3 replies, posted in Using SVP)

mark007 wrote:

Thanks a lot for that. Are those functions available within SVP (ie some BFI option) or would I need to create my own vapoursynth script?

C:\Program Files (x86)\SVP 4\script
look here

Имхо бред использовать такую старую версию. Проще держать копию mpv и смотреть аниме контент там
https://github.com/bloc97/Anime4K/blob/ … uctions.md

Xenocyde wrote:
dlr5668 wrote:
dawkinscm wrote:

I've been trying for a while now to get RIFE running smoothly with my 3080 with MPC-BE. I've given up on that so instead I tried using MPV and there was an instant improvement. It still needed a bit of work and I had to learn about MPV config files so that I could configure it to match the scaling used in my MadVR config. Now it looks great and runs very smoothly.

Nvidia should release soon https://www.youtube.com/watch?v=XA-tQpQqD7U Perfect match for Rife

Video Super Resolution support introduced today with the 531.18 drivers.

and its out. Here is mpc render for it https://github.com/emoose/VideoRenderer … ag/rtx-1.0

Not bad https://imgsli.com/MTU4Mzgx