Closed NenelFX closed 7 months ago
gpu_thread=16
... It made me think that you had 8*RTX4080. Try 1
first.
gpu_thread=16
... It made me think that you had 8*RTX4080. Try
1
first.
did not work.
I don't know much about this code "gpu_thread=16" in the other version of "HomeOfVapourSynthEvolution/VapourSynth-RIFE-ncnn-Vulkan" I used "gpu_thread=2 and 16" and it didn't make a difference in performance but there were no errors in all rife models but this version of "styler00dollar/VapourSynth-RIFE-ncnn-Vulkan" None of the models have worked for me yet always gives the error pipe:: Invalid data found when processing input if the error is because of the vulkan version of my gpu I'll have to wait for a version of RIFE v4.7 and 4.8 to come out from "nihui/rife-ncnn-vulkan"did not work.
I don't know much about this code "gpu_thread=16" in the other version of "HomeOfVapourSynthEvolution/VapourSynth-RIFE-ncnn-Vulkan" I used "gpu_thread=2 and 16" and it didn't make a difference in performance but there were no errors in all rife models but this version of "styler00dollar/VapourSynth-RIFE-ncnn-Vulkan" None of the models have worked for me yet always gives the error pipe:: Invalid data found when processing input if the error is because of the vulkan version of my gpu I'll have to wait for a version of RIFE v4.7 and 4.8 to come out from "nihui/rife-ncnn-vulkan"
pipe:: Invalid data found when processing input
is a generic message with no meaning. Can be anything. The error is either in another place in the console or in the piped data. Either look for the real error or look directly in the piped output by piping data into console instead of ffmpeg to see if something is written there. Sometimes error messages get piped. I assume your vapoursynth script is faulty and it is not even related to the plugin.
Previously I ran his simplified script here and worked well.
sup = core.mv.Super(clip)
bw = core.mv.Analyse(sup,isb=True)
clip = core.mv.SCDetection(clip,bw,thscd1=600,thscd2=130)
clip = core.resize.Bicubic(clip,format=vs.RGBS,matrix_in=1)
clip = core.resize.Bicubic(clip=clip, format=vs.RGBS, width=1280, height=720)
clip=core.rife.RIFE(clip, model=9, gpu_thread=2, tta=False, uhd=False, sc=True, factor_num=5, factor_den=2)
clip=core.resize.Bicubic(clip=clip, format=vs.YUV420P10, matrix_s="470bg")
clip=core.fmtc.bitdepth(clip, bits=8)
I would like to know exactly what the requirement for "VapourSynth-RIFE-ncnn-Vulkan" is
because the error is probably the Vulkan version of my GPU
If it would be vulkan related you would see something like vk...
in the error log. Read what I wrote and debug why your script is faulty.
I will close this due to lack of updates and assume it was user error.
I did some testing on this today, and for some reason, the max gpu thread count is 4. My GPU is not being utilized fully, and the vram is sitting at 1.5gb. my only guess is ncnn version, as compiling regular rife-ncnn-vulkan with newer ncnn versions causes a segfault in any scenario. I'll test later for known working versions of ncnn and see if that fixes it. Could be Linux specific tho, haven't tested windows or Mac.
tested and nothing, i can only assume a problem with vapoursynth, as the full error reads. Script evaluation failed: Python exception: 'utf-8' codec can't decode byte 0xb3 in position 6: invalid start byte
Traceback (most recent call last):
File "src/cython/vapoursynth.pyx", line 3124, in vapoursynth._vpy_evaluate
File "src/cython/vapoursynth.pyx", line 3125, in vapoursynth._vpy_evaluate
File "render.py", line 19, in
Failed to recognize file format. Exiting... (Errors when loading file)
I tried to compile older vapoursynth versions, but they dont compile on linux due to fexceptions not being enabled somewhere, I will still look into this, but i am not that good with c++ or vapoursynth in general.
Tried vapoursynth-git-R65.32.gcac1a7a3-1
, ncnn version 1.0.20240122 and gcc 13.2.1 on arch and it works for me. 2 threads result in close to 100% gpu usage on my 4090. This error happens if you select a cpu device and try to use multiple threads, but it works with one thread. With a gpu it should work without problems.
Hello, would you be able to do a build with this file? meson.build.txt meson.build.2.txt
It's giving me an error on my PC
pipe:: Invalid data found when processing input
It's all versions that give this error
the version of nihui/rife-ncnn-vulkan works all models
Will it be my GPU? Device Name GeForce 920M 1gb vram Vulkan 1.1.95
Or could it be that some dll?
I used this code
"clip=core.rife.RIFE(clip, model=9, gpu_thread=16, tta=False, uhd=False, sc=True, factor_num=5, factor_den=2)"
[ import vaporsynth as vs from vaporsynth import core core.num_threads=4 core.max_cache_size=4096 clip = core.raws.Source(r"-") sup = core.mv.Super(clip) bw = core.mv.Analyse(sup,isb=True) clip = core.mv.SCDetection(clip,bw,thscd1=600,thscd2=130) clip = core.resize.Bicubic(clip,format=vs.RGBS,matrix_in=1) clip = core.resize.Bicubic(clip=clip, format=vs.RGBS, width=1280, height=720) clip=core.rife.RIFE(clip, model=9, gpu_thread=16, tta=False, uhd=False, sc=True, factor_num=5, factor_den=2) clip=core.resize.Bicubic(clip=clip, format=vs.YUV420P10, matrix_s="470bg") clip=core.fmtc.bitdepth(clip, bits=8)
sup = core.mv.Super(clip)
fw = core.mv.Analyse(sup)
bw = core.mv.Analyse(sup,isb=True)
clip = core.mv.FlowBlur(clip,sup,bw,fw,blur=50.0)
sup = core.mv.Super(clip)
fw = core.mv.Analyse(sup)
bw = core.mv.Analyse(sup,isb=True)
clip = core.mv.FlowFPS(clip,sup,bw,fw,60,1,blend=False)
clip.set_output() ]