Log in

View Full Version : HAvsFunc


Pages : 1 2 3 [4] 5 6 7 8 9 10 11 12 13

mastrboy
5th October 2015, 17:46
Update r19.


Delete functions ediaa, maa, SharpAAMCmod, GammaToLinear, LinearToGamma.


Any reason why you decided to remove those functions ?

HolyWu
6th October 2015, 15:21
Any reason why you decided to remove those functions ?
The main AA functionality of the first three become duplicate of santiag, thus get removed. For masking or pre/postprocessing, just do it yourself. It's not that hard.

The latter two are obsolete since we already have fmtc.transfer.

HolyWu
6th October 2015, 16:19
Could you consider porting SRestore to Vapoursynth? It's a monster of a function but quite slow in Avisynth as it doesn't like multithreading much.
I encounter a difficulty when I step into that monster. We don't have PlaneMax/PlaneMin function to do the things like PlaneAverage!

Boulder
6th October 2015, 16:44
I encounter a difficulty when I step into that monster. We don't have PlaneMax/PlaneMin function to do the things like PlaneAverage!OK, thanks for checking it out anyway :) I'll just need to split the intermediate lossless encode to multiple parts in Avisynth to compensate, and hope that it won't crash:devil:

Myrsloik
6th October 2015, 16:55
I encounter a difficulty when I step into that monster. We don't have PlaneMax/PlaneMin function to do the things like PlaneAverage!

Are any other ones from avisynth (or in general) useful? It's trivial to calculate all of these values at the same time so maybe combining it into one more generalized function would make more sense. Thoughts? Ideas? Other far too complicated scripts you want to make sure they can be ported later on?

Or I'll simply add PlaneMinMax if no other useful ones are suggested. Did anyone actually ever use the median functionality in avisynth?

HolyWu
6th October 2015, 18:12
Currently there is no other script I really have interest in, so can't say what other ones may be needed. But indeed those runtime functions are very rarely used though.

bebs
6th October 2015, 21:46
Is it also possible to port chubby rain ? I find it efficient to remove rainbow on VHS records.

http://avisynth.nl/index.php/ChubbyRain

mawen1250
7th October 2015, 10:38
Are any other ones from avisynth (or in general) useful? It's trivial to calculate all of these values at the same time so maybe combining it into one more generalized function would make more sense. Thoughts? Ideas? Other far too complicated scripts you want to make sure they can be ported later on?

Or I'll simply add PlaneMinMax if no other useful ones are suggested. Did anyone actually ever use the median functionality in avisynth?

Something like calculating std, PSNR, etc?

aegisofrime
11th October 2015, 09:29
Hi all,

So I'm revisiting Vapoursynth after a long hiatus, and I'm confused about how to get everything working again.

Consider the following simple script:

import vapoursynth as vs
import havsfunc as haf
core = vs.get_core()
core.avs.LoadPlugin(r"C:\Program Files (x86)\AviSynth+\plugins\DGDecode.dll")
DFTTest = core.dfttest
#core.std.LoadPlugin(path=r"C:\Program Files (x86)\VapourSynth\plugins32\DFTTest.dll")
clip = core.avs.MPEG2Source(d2v =r"K:\Test\Test.d2v",idct=5)
clip = haf.QTGMC(clip, Preset='Very Slow', TFF=True)
clip.set_output()

I get the following error:

AttributeError: No attribute with the name fft3dfilter exists. Did you mistype a plugin namespace?


This is despite the fact that I do have FFT3DFilter.dll in my plugins32 directory. I thought that it was supposed to be loaded automatically? In any case, loading it manually doesn't change anything.

So I thought that I might as well use DFTTest as the denoiser since that is supported apparently, so I changed my QTGMC arguments to the following:

clip = haf.QTGMC(clip, Preset='Very Slow', TFF=True, Denoiser= DFTTest)

And now I get the following:

NameError: name 'DFTTest' is not defined

Again, I have downloaded DFTTest.dll (r3) and placed it in my plugins32 folder.

Can anyone point me in the right direction please?

Edit: I'm on Vapoursynth R28!

MonoS
11th October 2015, 10:49
clip = haf.QTGMC(clip, Preset='Very Slow', TFF=True, Denoiser= DFTTest)

DFTTest should go between "quotes"

jackoneill
11th October 2015, 12:38
d2vsource can replace DGDecode. (Unless it wasn't working for you?)

Did you remember to put the (correct) fftw3 DLL in the same folder as FFT3DFilter.dll? Or maybe it's supposed to go in system32 (or whatever it's called in 64 bit Windows).

aegisofrime
11th October 2015, 13:20
DFTTest should go between "quotes"

Damn, that's all it took. Doh! Thanks a bunch! :)

Now to go set up a 64-bit installation.

d2vsource can replace DGDecode. (Unless it wasn't working for you?)

Did you remember to put the (correct) fftw3 DLL in the same folder as FFT3DFilter.dll? Or maybe it's supposed to go in system32 (or whatever it's called in 64 bit Windows).

I reused my old script which I knew was working, which had d2vsource. I'm using DGDecode now which works thanks.

I did indeed have the libfftw3 dll in my plugins folder. I fixed it by downloading a copy of vsFFT3Dfilter from VFRManiac's site :) The included FFT3DFilter.dll wasn't working for some reason.

AzraelNewtype
11th October 2015, 21:51
I reused my old script which I knew was working, which had d2vsource. I'm using DGDecode now which works thanks.


d2vsource is much, much faster, so you may want to go back.

YamashitaRen
13th October 2015, 19:38
Are any other ones from avisynth (or in general) useful? It's trivial to calculate all of these values at the same time so maybe combining it into one more generalized function would make more sense. Thoughts? Ideas? Other far too complicated scripts you want to make sure they can be ported later on?

Just remembered that every-time I bump into some script with mt_lutxy, I wonder how I can port it to Vapoursynth...

edit : ergh forget what I said, everytime there is a mt_lutxy, I see a mt_lutspa instead ~~
edit 2 : maybe I should just let this here : https://gist.github.com/tp7/1e39044e1b660ef0a02c

HolyWu
20th October 2015, 18:42
Update r20.


YAHR: Use native filter of AWarpSharp2.
QTGMC: Fix SOvs argument is not scaled to the bit depth of input clip.
LSFmod: Fix overshoot(2) and undershoot(2) arguments are not scaled to the bit depth of input clip.
MinBlur: For r=2 or 3, improve the speed when the bit depth of input clip is 16.

polus
21st November 2015, 17:26
I am just trying out HAvsfunc, and in particular QTGMC.
If I run this very simple script in vsedit:

import vapoursynth as vs
core = vs.get_core()
import havsfunc as haf
sourceMovie = core.ffms2.Source(source='myFile.vob')
clip = haf.QTGMC(sourceMovie[1000:2000], Preset='Slow', TFF=True)
clip.set_output()


I get a walkback, with Vapoursynth complaining of extra argument 'patsize' in the bitdepth function call from the Resize function of HAvsfunc. Here is the python trace

Failed to evaluate the script:
Python exception: bitdepth: Function does not take argument(s) named patsize
Traceback (most recent call last):
File "src/cython/vapoursynth.pyx", line 1484, in vapoursynth.vpy_evaluateScript (src/cython/vapoursynth.c:26808)
File "/home/stefano/Documents/Projects/LinuxEncodingTutorial/tests/scripts/QTGMC-test.vpy", line 15, in <module>
clip = haf.QTGMC(sourceMovie[1000:2000], Preset='Slow', TFF=True)
File "/usr/lib/python3.5/site-packages/havsfunc.py", line 919, in QTGMC
spatialBlur = Resize(core.rgvs.RemoveGrain(repair0, [12] if isGray else [12, CMrg]), w, h, 0, 0, w + epsilon, h + epsilon, kernel='gauss', a1=2)
File "/usr/lib/python3.5/site-packages/havsfunc.py", line 3766, in Resize
return core.fmtc.bitdepth(last, bits=bits, fulls=fulls, fulld=fulld, dmode=dmode, ampo=ampo, ampn=ampn, dyn=dyn, staticnoise=staticnoise, patsize=patsize)
File "src/cython/vapoursynth.pyx", line 1349, in vapoursynth.Function.__call__ (src/cython/vapoursynth.c:24738)
vapoursynth.Error: bitdepth: Function does not take argument(s) named patsize

The offending line is:
return core.fmtc.bitdepth(last, bits=bits, fulls=fulls, fulld=fulld, dmode=dmode, ampo=ampo, ampn=ampn, dyn=dyn, staticnoise=staticnoise, patsize=patsize)


If (just for the fun of it, I don't really know what I'm doing) I edit the line and take out the last argument:
return core.fmtc.bitdepth(last, bits=bits, fulls=fulls, fulld=fulld, dmode=dmode, ampo=ampo, ampn=ampn, dyn=dyn, staticnoise=staticnoise) #, patsize=patsize)


The error disappears. Can anyone enlighten me on what is going on? Perhaps a version mismatch betwene VS and HAvsfunc?

I am running VapourSynth's git version and HAvsfunc r20-2, all installed with Archlinux packager (including all of HAvsfunc's dependent plugins).

Thanks for the help.

HolyWu
21st November 2015, 17:47
It means your built of fmtconv is somewhat old. The patsize parameter was added in fmtconv r17, which was released on 2015/07/09.

polus
21st November 2015, 21:31
Thanks for the quick reply. I suspected it was a version mismatched but did know where to look. Works fine now.

HolyWu
9th January 2016, 19:00
Update r21.


QTGMC: The frame property _FieldBased is set to 0 after deinterlacing now. Fix the dct argument wasn't passed to mv.Recalculate.
SMDegrain: Fix the dct argument wasn't passed to mv.Recalculate.
Clamp: Fix a wrong expression bug introduced in r20, causing the results of QTGMC's SLMode=2/4 and LSFmod's limit to be incorrect.

an3k
16th January 2016, 10:57
The core plugins are MVTools2, MaskTools v2, RemoveGrain and TemporalSoften (https://github.com/dubhater/vapoursynth-temporalsoften), they must be loaded. Others depending on your settings. If you don't use them, they don't have to be loaded.

Are these the correct links for the plugins?

MVTools2 (https://github.com/dubhater/vapoursynth-mvtools), MaskTools v2, RemoveGrain (http://www.vapoursynth.com/doc/plugins/rgvs.html#rgvs) and TemporalSoften (https://github.com/dubhater/vapoursynth-temporalsoften)

Where do I get MaskTools v2? Haven't found anything except MSmoosh (https://github.com/dubhater/vapoursynth-msmoosh)

Boulder
16th January 2016, 10:59
The functions of MaskTools are included in the Vapoursynth core.

an3k
17th January 2016, 03:55
Just a list of some Plugins one may need for HAvsFunc (required) / QTGMC (optional)

Linux:
Required
MVTools2: Download (https://github.com/dubhater/vapoursynth-mvtools/releases) / Git (https://github.com/dubhater/vapoursynth-mvtools) / Forum (http://forum.doom9.org/showthread.php?t=171207)
MaskTools v2: included in VapourSynth core
RemoveGrain: included in VapourSynth core
TemporalSoften (or use the version from SceneChange, see this post (http://forum.doom9.org/showthread.php?p=1753645#post1753645)): Download (https://github.com/dubhater/vapoursynth-temporalsoften/releases) / Git (https://github.com/dubhater/vapoursynth-temporalsoften)
Optional
fmtconv: Download (https://github.com/EleonoreMizo/fmtconv/releases) / Git (https://github.com/EleonoreMizo/fmtconv) / Forum (http://forum.doom9.org/showthread.php?t=166504)
FFTW3: Download (http://www.fftw.org/download.html) (preferred) / Download (https://github.com/FFTW/fftw3/releases) / Git (https://github.com/FFTW/fftw3)
SceneChange: Forum (http://forum.doom9.org/showthread.php?t=166769)
dfttest: Download (https://github.com/HomeOfVapourSynthEvolution/VapourSynth-DFTTest/releases) / Git (https://github.com/HomeOfVapourSynthEvolution/VapourSynth-DFTTest)
NNEDI3: Download (https://github.com/dubhater/vapoursynth-nnedi3/releases) / Git (https://github.com/dubhater/vapoursynth-nnedi3) / Forum (http://forum.doom9.org/showthread.php?t=166434)Windows:

See http://forum.doom9.org/showthread.php?t=156028

sl1pkn07
17th January 2016, 05:15
my list

addgrain: for QTGMC function (optional)
awarpsharp2: for YAHR function
bilateral: for logoNR function
ctmf: for HQDeringmod function
dctfilter: for Deblock_QED function
deblock: for Deblock_QED function
dfttest: for QTGMC function (optional)
eedi2: for santiag function
fft3dfilter: for QTGMC function (optional)
flash3kyuu_deband: for SmoothLevels function
fluxsmooth: for logoNR function
fmtconv: for QTGMC function
genericfilters: for HQDering function
knlmeanscl: for QTGMC function (optional)
mvtools: for QTGMC, GrainStabilizeMC and LSFmod function
nnedi3: for QTGMC function
sangnommod: for santiag function
scenechange: for QTGMC function
temporalsoften: for TemporalSoften function

speedyrazor
24th January 2016, 09:33
Hi, I am trying to get QTGMC working with VapourSynth 64bit portable, but whenever I try to run my script it crashes VSPipe of VapourSynth Editor straight away.
VapourSynth is up and running as I can run other scripts fine.
I have these plugins (which are all the 64bit versions) in the plugin folder:

DFTTest.dll
Yadifmod.dll
fmtconv.dll
libfftw3f-3.dll
libmvtools.dll
libnnedi3.dll
nnedi3_weights.bin
scenechange.dll
temporalsoften.dll
vslsmashsource.dll

And here's the contents of the coreplugins folder:

AvsCompat.dll
EEDI3.dll
RemoveGrainVS.dll
VIVTC.dll
Vinverse.dll
avisource.dll
genericfilters.dll
libassvapour.dll
libhistogram.dll
libtemporalsoften.dll

And here is my script:

import vapoursynth as vs
import havsfunc as haf
core = vs.get_core()
ret = core.lsmas.LibavSMASHSource(source=r"F:/Interlaced_Test.mov")
ret = haf.QTGMC(ret, Preset='Medium', TFF=True)
ret = core.fmtc.resample (clip=ret, w=720, h=576, css="444", kernel="spline36")
ret = core.fmtc.matrix (clip=ret, mats="709", matd="601")
ret = core.fmtc.resample (clip=ret, css="422")
ret = core.fmtc.bitdepth (clip=ret, bits=10)
ret.set_output()

My source is a 1920x1080 Prores file.

sl1pkn07
24th January 2016, 09:36
and the error is?

speedyrazor
24th January 2016, 09:39
and the error is?

"VSPip.exe has stopped working"

and

"VapuorSynth Editor has stopped working"

sl1pkn07
24th January 2016, 09:41
no more?

please add a sample of the video

EDIT:
tested with http://samples.mplayerhq.hu/V-codecs/HCPA/AppleProRes422.mov

this is your error?

Failed to evaluate the script:
Python exception: bitdepth: Function does not take argument(s) named patsize
Traceback (most recent call last):
File "src/cython/vapoursynth.pyx", line 1489, in vapoursynth.vpy_evaluateScript (src/cython/vapoursynth.c:26885)
File "/home/sl1pkn07/aplicaciones/vapoursynth-test/qtmc-test.vpy", line 5, in <module>
ret = haf.QTGMC(ret, Preset='Medium', TFF=True)
File "/usr/lib/python3.5/site-packages/havsfunc.py", line 918, in QTGMC
spatialBlur = Resize(core.rgvs.RemoveGrain(repair0, [12] if isGray else [12, CMrg]), w, h, 0, 0, w + epsilon, h + epsilon, kernel='gauss', a1=2)
File "/usr/lib/python3.5/site-packages/havsfunc.py", line 3774, in Resize
return core.fmtc.bitdepth(last, bits=bits, fulls=fulls, fulld=fulld, dmode=dmode, ampo=ampo, ampn=ampn, dyn=dyn, staticnoise=staticnoise, patsize=patsize)
File "src/cython/vapoursynth.pyx", line 1354, in vapoursynth.Function.__call__ (src/cython/vapoursynth.c:24815)
vapoursynth.Error: bitdepth: Function does not take argument(s) named patsize

speedyrazor
24th January 2016, 09:51
no more?

please add a sample of the video

Thats it, they both just crash with "stopped working".

I have attached a sample which has the same effect. It must be my setup, plugins, etc?

speedyrazor
24th January 2016, 09:56
no more?

please add a sample of the video

EDIT:
tested with http://samples.mplayerhq.hu/V-codecs/HCPA/AppleProRes422.mov

this is your error?

Failed to evaluate the script:
Python exception: bitdepth: Function does not take argument(s) named patsize
Traceback (most recent call last):
File "src/cython/vapoursynth.pyx", line 1489, in vapoursynth.vpy_evaluateScript (src/cython/vapoursynth.c:26885)
File "/home/sl1pkn07/aplicaciones/vapoursynth-test/qtmc-test.vpy", line 5, in <module>
ret = haf.QTGMC(ret, Preset='Medium', TFF=True)
File "/usr/lib/python3.5/site-packages/havsfunc.py", line 918, in QTGMC
spatialBlur = Resize(core.rgvs.RemoveGrain(repair0, [12] if isGray else [12, CMrg]), w, h, 0, 0, w + epsilon, h + epsilon, kernel='gauss', a1=2)
File "/usr/lib/python3.5/site-packages/havsfunc.py", line 3774, in Resize
return core.fmtc.bitdepth(last, bits=bits, fulls=fulls, fulld=fulld, dmode=dmode, ampo=ampo, ampn=ampn, dyn=dyn, staticnoise=staticnoise, patsize=patsize)
File "src/cython/vapoursynth.pyx", line 1354, in vapoursynth.Function.__call__ (src/cython/vapoursynth.c:24815)
vapoursynth.Error: bitdepth: Function does not take argument(s) named patsize


How can I see this error message?
So what does this actually mean please?

feisty2
24th January 2016, 10:08
vapoursynth.Error: bitdepth: Function does not take argument(s) named patsize

sl1pkn07
24th January 2016, 10:21
this means i need update my fmtconv (http://forum.doom9.org/showpost.php?p=1747378&postcount=167)

but i get segfault XD

PID: 11849 (vspipe)
UID: 1000 (sl1pkn07)
GID: 100 (users)
Signal: 11 (SEGV)
Timestamp: dom 2016-01-24 10:30:17 CET (13s ago)
Command Line: vspipe -i qtgmc-test.vpy -
Executable: /usr/bin/vspipe
Control Group: /user.slice/user-1000.slice/session-c1.scope
Unit: session-c1.scope
Slice: user-1000.slice
Session: c1
Owner UID: 1000 (sl1pkn07)
Boot ID: a6e84e9c964048a48a6df7c9ff78e494
Machine ID: c20ee0c57658685bfedf50384b0e3ec0
Hostname: sL1pKn07
Coredump: /var/lib/systemd/coredump/core.vspipe.1000.a6e84e9c964048a48a6df7c9ff78e494.11849.1453627817000000.lz4
Message: Process 11849 (vspipe) of user 1000 dumped core.

Stack trace of thread 11953:
#0 0x0000000000000001 n/a (n/a)


i think this is the real problem

speedyrazor
24th January 2016, 11:03
vapoursynth.Error: bitdepth: Function does not take argument(s) named patsize
Does this mean I cant use QTGMC with a 10 bit source , or something else?
Is there anyway to to fix this, so I can use QTGMC?

feisty2
24th January 2016, 11:10
Does this mean I cant use QTGMC with a 10 bit source , or something else?
Is there anyway to to fix this, so I can use QTGMC?

it tells you to update fmtconv.dll

speedyrazor
24th January 2016, 11:18
it tells you to update fmtconv.dll

Updated to fmtconv-r17, still have exatclt the same issue. ".... has stopped working"

feisty2
24th January 2016, 11:25
Updated to fmtconv-r17, still have exatclt the same issue. ".... has stopped working"

sweetheart, "... has stopped working" is not a diagnosable error, that's not the way how troubleshooting works..

speedyrazor
24th January 2016, 11:27
sweetheart, "... has stopped working" is not a diagnosable error, that's not the way how troubleshooting works..

Well, could you tell me how to get more info then please, I would love to report more, but currently thats all I have.

sl1pkn07
24th January 2016, 11:27
sounds a segfault (see my latest post ;_; )

feisty2
24th January 2016, 11:31
http://i.imgur.com/OKVw36i.png
try troubleshooting with vseditor

sl1pkn07
24th January 2016, 11:33
i get segfault in both. vspipe and vsedit

but the coredump is wreid, ony point to http://forum.doom9.org/showpost.php?p=1754591&postcount=181

speedyrazor
24th January 2016, 11:36
http://i.imgur.com/OKVw36i.png
try troubleshooting with vseditor
I have tried that, if I choose 'Check Script' again.... All I get is "VapuorSynth Editor has stopped working", as I keep saying, I dont get anything else. See attached screen grab.

sl1pkn07
24th January 2016, 11:45
please use imgur or something

foxyshadis
24th January 2016, 12:01
Probably something triggering a vsFatal, which annoyingly crashes instead of letting the user do anything about it. It'd be nice if vsFatal at least popped up a simple messagebox detailing the error before crashing, if it's not going to incorporate any kind of crash-reporting code.

Boulder
24th January 2016, 12:07
Does the source filter return normalized framerates? That kind of a crash occurs when it does not.

jackoneill
24th January 2016, 12:12
Debugging a crash in Linux:

gdb --args <your command here>
run
[crashes]
bt full

You probably don't have debugging symbols at this point, so this only tells you what plugin is crashing. Recompile that with

-O0 -g

in CFLAGS and CXXFLAGS and use gdb again. Then post the backtrace.

foxyshadis: In Linux at least you get the error message before the crash, and it's not a segfault (signal 11).

sl1pkn07
24th January 2016, 12:29
http://sl1pkn07.wtf/paste/view/cffc1dfc

using this sample: https://www.arri.com/DE/camera/alexa/learn/alexa_sample_footage/
(ftp://ftp-footage.arri.de/ALEXA%20XT/03_ProRes_16-9_2048x1152/30fps/J001C062_140110_R6MS.mov)

jackoneill
24th January 2016, 13:20
Sorry about that. Please update nnedi3: https://github.com/dubhater/vapoursynth-nnedi3/releases/tag/v8

sl1pkn07
24th January 2016, 13:29
yep. now works for me

tnx bro!

speedyrazor
24th January 2016, 13:39
Updated with V8, unfortunately I am getting:

Failed to evaluate the script:
Python exception: nnedi3: Couldn't open file 'G:/ffmpeg/VapourSynth/vapoursynth64/plugins/nnedi3_weights.bin'. Error message: No such file or directory
Traceback (most recent call last):
File "src\cython\vapoursynth.pyx", line 1489, in vapoursynth.vpy_evaluateScript (src\cython\vapoursynth.c:26885)
File "", line 5, in <module>
File "G:\ffmpeg\VapourSynth\havsfunc.py", line 1045, in QTGMC
edi1 = QTGMC_Interpolate(ediInput, InputType, EdiMode, NNSize, NNeurons, EdiQual, EdiMaxD, bobbed, ChromaEdi, TFF)
File "G:\ffmpeg\VapourSynth\havsfunc.py", line 1326, in QTGMC_Interpolate
interp = core.nnedi3.nnedi3(Input, field=field, U=CEed, V=CEed, nsize=NNSize, nns=NNeurons, qual=EdiQual)
File "src\cython\vapoursynth.pyx", line 1381, in vapoursynth.Function.__call__ (src\cython\vapoursynth.c:25200)
vapoursynth.Error: nnedi3: Couldn't open file 'G:/ffmpeg/VapourSynth/vapoursynth64/plugins/nnedi3_weights.bin'. Error message: No such file or directory

And I checked and doubdle checked, I do have the file nnedi3_weights.bin in G:/ffmpeg/VapourSynth/vapoursynth64/plugins folder. I am using the portable version.

jackoneill
24th January 2016, 13:53
Updated with V8, unfortunately I am getting:

Failed to evaluate the script:
Python exception: nnedi3: Couldn't open file 'G:/ffmpeg/VapourSynth/vapoursynth64/plugins/nnedi3_weights.bin'. Error message: No such file or directory
Traceback (most recent call last):
File "src\cython\vapoursynth.pyx", line 1489, in vapoursynth.vpy_evaluateScript (src\cython\vapoursynth.c:26885)
File "", line 5, in <module>
File "G:\ffmpeg\VapourSynth\havsfunc.py", line 1045, in QTGMC
edi1 = QTGMC_Interpolate(ediInput, InputType, EdiMode, NNSize, NNeurons, EdiQual, EdiMaxD, bobbed, ChromaEdi, TFF)
File "G:\ffmpeg\VapourSynth\havsfunc.py", line 1326, in QTGMC_Interpolate
interp = core.nnedi3.nnedi3(Input, field=field, U=CEed, V=CEed, nsize=NNSize, nns=NNeurons, qual=EdiQual)
File "src\cython\vapoursynth.pyx", line 1381, in vapoursynth.Function.__call__ (src\cython\vapoursynth.c:25200)
vapoursynth.Error: nnedi3: Couldn't open file 'G:/ffmpeg/VapourSynth/vapoursynth64/plugins/nnedi3_weights.bin'. Error message: No such file or directory

And I checked and doubdle checked, I do have the file nnedi3_weights.bin in G:/ffmpeg/VapourSynth/vapoursynth64/plugins folder. I am using the portable version.

This is strange. Which version did you have earlier? Is libnnedi3.dll in that folder too? Is that where the previous version was? Did you simply overwrite the old libnnedi3.dll?

Edit: disregard the above. It was my compiler's fault. Please download v8 again.

For those interested in such things, GCC 5.3.0 broke

std::wstring_convert<std::codecvt_utf8_utf16<wchar_t>, wchar_t>::from_bytes()

which I use to convert the char* path returned by getPluginPath to wchar_t*, for _wfopen, which opens nnedi3_weights.bin.

speedyrazor
24th January 2016, 14:03
Just deleted out both nnedi3_weights.bin and libnnedi3.dll and downloaded them, put them in the plugins folder, but still get the same issue.

I have attached 2 screen grabs, one of the plugins folder and one of the core plugins folder.