tntwise / rife-ncnn-vulkan Goto Github PK

Thank you for your upstream version
Fix Build from Source README.md section

Hi, sorry for bothering you but it's kinda weird that the Lite version of 4.14 model is slower than the normal one.

I am using a RTX 4060, and all other Lite models are quicker than the normal models, except 4.14-lite

Is this a bug? Or just how things should work?

Thank you

From ./build/main.cpp code, where should I modify to earn 16 bit depth option for png or webp?

This issue only happens when using .EXE after the 7976202 commit Release 20240524 & Release 20240515

Swap the .EXE back to the older version Release 20240326 will fix the issue, so this swap submodule update could be the cause of this issue

Thank you for your hard work in maintaining this open-source project. I apologize for any inconvenience caused.

Hello, I use a plugin "RIFE.dll" from vapoursynth
from "HomeOfVapourSynthEvolution/VapourSynth-RIFE-ncnn-Vulkan"
But this project has been discontinued from github.
it even supports the "rife-v4.6" model, from then on, no other works. the error "layer MemoryData not exists or registered" appears

and "librife.dll" from "styler00dollar/VapourSynth-RIFE-ncnn-Vulkan"
It doesn't work on my PC, the problem could be that my CPU or GPU are incompatible with "librife.dll" I already spoke to him but I couldn't resolve it.

Your application works perfectly on my PC, all models work.

But in the vapoursynth plugin "RIFE.dll", the models from "rife-v4.7" below do not always work and it appears "layer MemoryData not exists or registered"

I've been waiting all this time for "nihui/rife-ncnn-vulkan" to have a new version but I'm still waiting.

If it were possible, I would like you to adapt the "rife-v4.8" model so that the error "layer MemoryData not exists or registered" does not appear
If it is not possible, please let me know

Grateful for your help. On (my) Android device, CPU typically runs rife-ncnn-vulkan faster than GPU, so it's meaningful to fix the bug.

I tested with this command

rife-ncnn-vulkan -0 0.png -1 1.png -o output-cpu.png -m rife-v4.6 -g -1 -v

• OS: Termux (Android), with LD_PRELOAD=/system/lib64
• Hardware: Samsung Z Fold5, Snapdragon 8 Gen 2, Adreno 740

If you are happy to run try on your Android phone (or emulator), you can install the Termux package builds from my actions:

• https://github.com/knyipab/tur/actions/runs/9099228085
• https://github.com/knyipab/tur/actions/runs/9099226500

TNTwise/rife-ncnn-vulkan

Failed, log:

[0 Adreno (TM) 740]  queueC=0[3]  queueG=0[3]  queueT=0[3]
[0 Adreno (TM) 740]  bugsbn1=1  bugbilz=0  bugcopc=0  bugihfa=0
[0 Adreno (TM) 740]  fp16-p/s/u/a=1/1/0/1  int8-p/s/u/a=1/1/0/1
[0 Adreno (TM) 740]  subgroup=64  basic/vote/ballot/shuffle=1/1/1/1
[0 Adreno (TM) 740]  fp16-8x8x16/16x8x8/16x8x16/16x16x16=0/0/0/0
Segmentation fault

nihui/rife-ncnn-vulkan

Succeeded, log:

[0 Adreno (TM) 740]  queueC=0[3]  queueG=0[3]  queueT=0[3]
[0 Adreno (TM) 740]  bugsbn1=1  bugbilz=0  bugcopc=0  bugihfa=0
[0 Adreno (TM) 740]  fp16-p/s/a=1/1/1  int8-p/s/a=1/1/1
[0 Adreno (TM) 740]  subgroup=64  basic=1  vote=1  ballot=1  shuffle=1
0.png 1.png 0.500000 -> output-cpu.png done