Hybrid(Windows/Linux/Mac): Input -> x264/x265/Xvid/VP8/VP9

[Platos]

Ok, thank you. I tried it.

So far it looks the same as ususl. So it seems it works with 60FPS in the speedchange tab.

And yes, i used always 0.2 scene changes. But i will try 0.15.

[Platos]

I found some issue:

When i want to save "current setting as a configuration" in Vapoursynth (there left from the script preview, click on the "+" and save as "test"), it wont work. I get endless errors which all looks like:

FilteringModel get-> found no field 'VsBalanceBordersSobelFastEdgeMaskBinarizeTreshold'

While the stuff in the 'xxx' is different in each error.

I Can't make a debug file right? because it has nothing to do with encoding itself ? Or what would you need?

It happens btw. too, when i start Hybrid (with my normal setting i always used now for 1-2 weeks) and save the current settings again as "test" (so with absolutly no changes).

I had to end hybrid by taskMGR because i surely clicked away 50 error messages. So there are alot.

When i start Hybrid i see in the Logfile: "Resetting Vapoursynth filter order. Adjust&Save the new FilteringModel defaults."

I'm not sure, but i did not notice this message before today in Hybrid. But maybe i just never noticed it.

[Selur]

Debug output has nothing to do with encoding.

"Resetting Vapoursynth filter order. Adjust&Save the new FilteringModel defaults."

You should adjust and save the new filtering model defaults.
1. "Filtering->Vapoursynth->Misc->Filter Order/Queue->Reset"
2. "Defaults->Save All"
3. Restart Hybrid should help with that.


Regarding saving the configuration: I can reproduce the problem and will look into it.

FilteringModel get-> found no field 'VsBalanceBordersSobelFastEdgeMaskBinarizeTreshold '

I made a typo somewhere 'VsBalanceBordersSobelFastEdgeMaskBinarizeTreshold ' should be 'vsBalanceBordersSobelFastEdgeMaskBinarizeTreshold'.


Cu Selur

[Selur]

@Platos: send you a link to a dev version which should fix the typo.

[Platos]

Thanks, it works now. Now i can save new Profiles.

The filter Order is also solved.

When i was in this Tab filterorder, i clicked on the "use filter Queue" box (to look what happens). Then it appeared this Errors:

vsUseQueueCheckBox: Couldn't find vsmCleanAddFilterPushButton

And 3 other errors which it could not find "vsmCleanRemovedFilterPushButton" and "vssRestoreAddFilterPushButton" and "vssRestoreRemovedFilterPushButton".

It happens after i use the new dev version (probably before too) and also after i restore the filter order as you said (and also before, i test both).

[Selur]

Argh,... should be "vsMCleanAddFilterPushButton" -> will look into it in ~ 2hrs, afk atm.

[Selur]

send you a new link

[Platos]

Used the newest dev and it works now.

How can i make, that the Profile under Vapoursynth is automaticly loaded on startup? I have always to load/apply the profile, that it works.

I dont know, hoe i did that last time, because the Rest in my Profile is loaded directly on start. Is the Option under "config-profiles-auto save defaults" the only option to auto-apply the profile under vapoursynth?

[Selur]

There is no way to autoload profiles, but maybe [INFO] About profiles and saving settings,.... helps you find a solution, like auto-saving the current settings as defaults,...

Cu Selur

[Platos]

Ah ok, then i try it with the default settings.

Question: When i minimize Hybrid, it does not go to tray. It dissappears completly (but it is still there, because i can see it in the TaskMGR). But therefore it is not in the tray i can not use it anymore.

Is there a way to solve that? Do i have to set it in Setting somewhere or why does this happen ?

[Selur]

Argh, that seems to be sometthing that changed.
-> I'm looking into it

[Selur]

No clue whats causing this. atm. I suspect it's related to Qt.6.5.
As soon as I once hide the systray icon, I can't make it visible again.

[Platos]

Hmm strange.

I can only say i noticed it then (probably you did not see it): https://forum.videohelp.com/threads/345169-Hybrid(Windows-Linux-Mac)-Input-x264-x265-X...52#post2690567

So but i can't say absolutly 100% that it worked before. But im pretty sure it worked before. But i can't be 100% sure.

But maybe this helps you.

I thought it's probably already solved, but i re-noticed it today, so i said you again for let you know.

[Selur]

I'm trying some workarounds,..

[Selur]

Send you a link to a new dev for testing.

Cu Selur

[Platos]

Thanks!

It works now with the newest dev. Nice!

Btw. i tried some new models in vsmlrt and i think the time to create the .engine file is waaaay shorter on compact-models.

I believe, the time to create the .engine file is longer, when the upscaling-speed on the used model is slower. So at least it was the case for me. Slow Upsclaing = slow .egine creating so far.

i needed over 30 minutes for one

for compact-models i need normally 2-5 mins or so.

Are the created .engine files actually saved forever now ? For me it seems like that (what i would like, because of that time consumption for creating .engine files).

[Selur]

Yes, compact models are usually faster to process and regarding the .engine files: Hybrid doesn't delete them, so in theory they should stay 'forever'. Didn't have time to check what criteria all go into the models. (I suspect it's 'just' the resolution, so you might be fine to first create xy .engine files for all the resolutions you use.)

Cu Selur

[sophisticles]

I have run into an odd problem when trying to convert the Valkaama DVD to either x264 or x265 under Linux.

See the attached complete encodes.

Any idea what is going on?

[Selur]

Looks like the color space piped to the encoder isn't the color space the encoder expected, or the decoder outputs some unwanted data to the output.
Assuming you used Hybrid, I would need to look at a debug output to know the issue.

[sophisticles]

Yes, I used Hybrid. How do I create a debug output? Do I create it after an encode is done?

[Selur]

About creating a debug output, read: https://www.selur.de/support
Debug output is created while stuff is done.
In your case, the job creation, should be enough to see the cause.
So if you load your source, do your settings enable the debug output creation, trigger the job creation and then share the debug output I probably can tell you the cause of the issue.

Cu Selur

[Platos]

Hello,

I have found an Error.

In Filtering-Vapoursynth when i click on the "configuration apply" button (of a made profile), it gives me a error with

FilteringModel doSet->found no field: 'vsSCUNetNVFuser'

This is new. It once was without error with this profile (i have only this one).

[Selur]

Oh, thats some old parameter from the torchAddon, which is on the general skip list, so I'm surprised this happens.
Loading, saving and loading the profile again should fix the problem.

[Platos]

Ah, yes it worked, thanks.

I found another issue with BasicVSR++. I tried to use it, but it crashed. I used default setting in BasicVSR++.

Here is a debug file:

[Selur]

You are probably running out of VRAM with those settings.
Using the default settings and a 640x352 source you need ~7.2GB of VRAM when using BasicVSR++ with those settings.
With 4 times higher resolution, those settings will require around 29GB of VRAM. *gig*
Try:
a. using FP16
b. setting a tile size like 320x360.
=> doesn't look like a bug, but a problem with you not knowing what you are doing.
If the Vapoursynth Preview doesn't work, neither will encoding.

Cu Selur

Ps.: Also, your input and output resolution is the same, so you will lose a lot of the stuff the resizer did.

[Platos]

Hmm, i tried it, but i still get errors. I tried your settings and i even resized the video down to 640x360 and used this as input to lower vram usage. But it still dont work. And i also tried it with cpu cache enabled. I also tried it with tile size 640x720 and 160x180. Nothing worked.

My GPU has 12GB of vram. When you say, a 640x352 source should need 7.2GB, shouldn't it work ?

And i also tried all models in BasicVSR++. I Also tried different "length". So 1, 5, 9 and 15.

This says the vapoursynth preview:

# Imports
import vapoursynth as vs
# getting Vapoursynth core
core = vs.core
import site
import os
import ctypes
# Adding torch dependencies to PATH
path = site.getsitepackages()[0]+'/torch_dependencies/bin/'
ctypes.windll.kernel32.SetDllDirectoryW(path)
path = path.replace('\\', '/')
os.environ["PATH"] = path + os.pathsep + os.environ["PATH"]
os.environ["CUDA_MODULE_LOADING"] = "LAZY"
# Loading Plugins
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/fmtconv.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/SourceFilter/LSmashSource/vslsmashsource.dll")
# source: 'C:\Users\Gaming-Tower\Videos\Real Film Test\test_new.mkv'
# current color space: YUV420P10, bit depth: 10, resolution: 640x360, fps: 30, color matrix: 709, yuv luminance scale: limited, scanorder: progressive
# Loading C:\Users\Gaming-Tower\Videos\Real Film Test\test_new.mkv using LWLibavSource
clip = core.lsmas.LWLibavSource(source="C:/Users/Gaming-Tower/Videos/Real Film Test/test_new.mkv", format="YUV420P10", stream_index=0, cache=0, prefer_hw=0)
# Setting detected color matrix (709).
clip = core.std.SetFrameProps(clip, _Matrix=1)
# Setting color transfer info (709), when it is not set
clip = clip if not core.text.FrameProps(clip,'_Transfer') else core.std.SetFrameProps(clip, _Transfer=1)
# Setting color primaries info (), when it is not set
clip = clip if not core.text.FrameProps(clip,'_Primaries') else core.std.SetFrameProps(clip, _Primaries=5)
# Setting color range to TV (limited) range.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
# making sure frame rate is set to 30
clip = core.std.AssumeFPS(clip=clip, fpsnum=30, fpsden=1)
clip = core.std.SetFrameProp(clip=clip, prop="_FieldBased", intval=0) # progressive
# adjusting color space from YUV420P10 to RGBH for vsBasicVSRPP
clip = core.resize.Bicubic(clip=clip, format=vs.RGBH, matrix_in_s="709", range_s="limited")
# resizing using BasicVSRPP
from vsbasicvsrpp import basicvsrpp as BasicVSRPP
# target zoom factor: 1 adjusted to 4
clip = BasicVSRPP(clip=clip, model=1, tile_w=320, tile_h=360) # 2560x1440
# resizing 2560x1440 to 640x360
# adjusting resizing
clip = core.resize.Bicubic(clip=clip, format=vs.RGBS, range_s="limited")
clip = core.fmtc.resample(clip=clip, w=640, h=360, kernel="spline64", interlaced=False, interlacedd=False)
# adjusting output color from: RGBS to YUV420P10 for NVEncModel
clip = core.resize.Bicubic(clip=clip, format=vs.YUV420P10, matrix_s="709", range_s="limited", dither_type="error_diffusion")
# set output frame rate to 30fps (progressive)
clip = core.std.AssumeFPS(clip=clip, fpsnum=30, fpsden=1)
# Output
clip.set_output()

and this the log in the preview:

Failed to evaluate the script:
Python exception: 

IMPORTANT: PLEASE READ THIS FOR ADVICE ON HOW TO SOLVE THIS ISSUE!

Importing the numpy C-extensions failed. This error can happen for
many reasons, often due to issues with your setup or how NumPy was
installed.

We have compiled some common reasons and troubleshooting tips at:

https://numpy.org/devdocs/user/troubleshooting-importerror.html

Please note and check the following:

* The Python version is: Python3.11 from "C:\Program Files\Hybrid\64bit\Vapoursynth\vsViewer.exe"
* The NumPy version is: "1.24.2"

and make sure that they are the versions you expect.
Please carefully study the documentation linked above for further help.

Original error was: No module named 'numpy.core._multiarray_umath'


Traceback (most recent call last):
File "C:\Program Files\Hybrid\64bit\Vapoursynth\Lib\site-packages\numpy\core\__init__.py", line 23, in 
from . import multiarray
File "C:\Program Files\Hybrid\64bit\Vapoursynth\Lib\site-packages\numpy\core\multiarray.py", line 10, in 
from . import overrides
File "C:\Program Files\Hybrid\64bit\Vapoursynth\Lib\site-packages\numpy\core\overrides.py", line 6, in 
from numpy.core._multiarray_umath import (
ModuleNotFoundError: No module named 'numpy.core._multiarray_umath'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "src\cython\vapoursynth.pyx", line 3115, in vapoursynth._vpy_evaluate
File "src\cython\vapoursynth.pyx", line 3116, in vapoursynth._vpy_evaluate
File "C:\Users\Gaming-Tower\AppData\Local\Temp\tempPreviewVapoursynthFile23_38_37_623.vpy", line 35, in 
from vsbasicvsrpp import basicvsrpp as BasicVSRPP
File "C:\Program Files\Hybrid\64bit\Vapoursynth\Lib\site-packages\vsbasicvsrpp\__init__.py", line 7, in 
import numpy as np
File "C:\Program Files\Hybrid\64bit\Vapoursynth\Lib\site-packages\numpy\__init__.py", line 141, in 
from . import core
File "C:\Program Files\Hybrid\64bit\Vapoursynth\Lib\site-packages\numpy\core\__init__.py", line 49, in 
raise ImportError(msg)
ImportError: 

IMPORTANT: PLEASE READ THIS FOR ADVICE ON HOW TO SOLVE THIS ISSUE!

Importing the numpy C-extensions failed. This error can happen for
many reasons, often due to issues with your setup or how NumPy was
installed.

We have compiled some common reasons and troubleshooting tips at:

https://numpy.org/devdocs/user/troubleshooting-importerror.html

Please note and check the following:

* The Python version is: Python3.11 from "C:\Program Files\Hybrid\64bit\Vapoursynth\vsViewer.exe"
* The NumPy version is: "1.24.2"

and make sure that they are the versions you expect.
Please carefully study the documentation linked above for further help.

Original error was: No module named 'numpy.core._multiarray_umath'

It seems not to be a Hybrid issue, but i still did a debug:

[Selur]

No module named 'numpy.core._multiarray_umath'

Okay, seems like there is some issue with numpy.
Will try to look at it after work.

[hello_hello]

Selur,
I was playing with Hybrid today, now I finally have a PC it'll run on, and I just thought I'd let you know the 32 bit avsViewer won't open to preview video (should it default to the 64 bit version?)
The log file shows:
using 32bit AvisynthViewer: C:/Program Files/Hybrid\32bit\avsViewer.exe
Cheers.

[Selur]

No, 32bit Avisynth is the default, so 32bit avsViewer is also the default.
Just tested, here:

• started Hybrid
• (switched Avisynth to 32bit, since I use 64bit by default)
• loaded a source
• switched to Avisynth
• pressed 'Show Avisynth preview'

-> preview was shown without an issue.
If you can share a debug output of the problem (see: https://www.selur.de/support) and share it with me, I can look into it.

Cu Selur

[Platos]

I have an Issue since the last dev i guess.

Because now always all my videos have another aspect ratio. Doesnt matter what upscaler i use. I did not notice yesterday, because i did only a short test, if the processing did not crash (like before). But now i used a video i used already before (so i know it worked). And as i said now the aspect ratio does not stay original. It will be sth. else, doesnt matter what upscaler i use. I tried different models on vsmlrt and on VSGAN without vsmlrt. But as i see it happens also on BasicVSR++ (as i said did not notice yesterday on the other video).

Here is a debug: