![]() |
Okay, you are not using Hybrid, you are using vsViewer on it's own.
Open the file with Hybrid, open the Vapoursynth Preview. |
Quote:
|
Yes, if you want to use that as source.
|
It's ok but the output file has kind of blurred kind of motion effect. I am trying to recreate manually setting the filters in Hybrid and there also is a blurry effect. Which filter can cause such problem ?
|
Probably, mclean
|
Quote:
|
Best simply apply the filters one by one check the Vapoursynth Preview to see what they do.
The point of Hybrids tons of option is to allow adjusting filters&co to your needs. Due to the input quality of your source, with conventional filtering, you will either have to live with blocking, artifacts, noise, or some blurring to hide those. -- merged -- Personally, I would apply different filters per scene. i.e. using RediuceFlicker and SmallDeflicker can help a lot in the zoomed out view, but if one applies them to the close-ups you end up with artifacts, like this: https://www.digitalfaq.com/forum/error.gif Cu Selur -- merged -- Here's something to try:
Now, try different filters under 'Filtering->Vapoursynth->DeGrain' and 'Filtering->Vapoursynth->DeNoise' and play with their settings. |
Quote:
Failed to evaluate the script: Python exception: ReplaceFramesSimple: Index out of bounds in mappings at line 1, column 6 Traceback (most recent call last): File "src/cython/vapoursynth.pyx", line 3378, in vapoursynth._vpy_evaluate File "src/cython/vapoursynth.pyx", line 3379, in vapoursynth._vpy_evaluate File "C:\Users\tesar\AppData\Local\Temp\tempPreviewVapo ursynthFile13_51_06_827.vpy", line 69, in clip = core.remap.Rfs(baseclip=clip, sourceclip=core.rdfl.ReduceFlicker(clip=clip, strength=3, aggressive=1), mappings="[30- 535]") ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^ File "src/cython/vapoursynth.pyx", line 3114, in vapoursynth.Function.__call__ vapoursynth.Error: ReplaceFramesSimple: Index out of bounds in mappings at line 1, column 6 |
Sorry, forgot to mention:
Enable 'Filtering->(De-)Interlace/Telecine->QTGMC Vapoursynth->Bob' without this the frame count is halved and thus the range would need to be "[15-267]" :) Cu Selur |
2 Attachment(s)
Quote:
|
1 Attachment(s)
Use what ever you like, with filtering it's always down to what one personally prefers.
Chroma shift is not correct for sample 2. |
Quote:
|
try without it,..
|
1 Attachment(s)
Quote:
|
Hmm.
Video: - Temporal NR seems too aggressive. - Something with the interlace seems off. NR can harm interlace. Resize by anything other than 4-pixel increments can cause wrong field swapping. - The luma/chroma are acceptable. - I don't like all the in-image wiggles. Not sure if this is from lack of line TBC, nth gen VHS, broadcast, NR processing, or what. It's distracting, like the video was invaded by amoeba, and it breathes. Audio: - Perhaps less muffled, which isn't really that hard to do restore-wise. I've not participated much in this thread. selur was guiding you through Hybrid, so I let the master do his work. :) |
Quote:
If this was video, not audio, the darker picture details would now be wiped out. Listen for not only the removed sound but also the sound we are trying to preserve. We're trying to do as little damage as possible to the wanted audio. The proper way is to filter out only these tones and preserve as much possible the rest of the audio. This means using filters that are very selective. Sometimes that's possible. Sometimes it isnt. As one example Izotope RX Elements contains a DeHum tool. Here's a YT demo https://www.youtube.com/watch?v=iDRQXJmC6Eg&t=12s It's important to use the right tools but also to know how to use them, including knowing their limitations. Having said that, most likely the original tape playback was poor. I suspect there's much clearer audio on that tape, waiting to be captured well. Audio capture is one of my specialties. Feel free to send me the original tape and I will capture the linear audio track at near to its best, for the cost of postage. PM me if interested. |
Quote:
|
Quote:
|
Quote:
|
1 Attachment(s)
Haven't tried anything, but Spotless with a radius of three and a luma mask to only filter stuff 150+ (expanded by one pixel) might work,...
=> argh, won't work since the ball would be filtered too and disappear in some cases :) Try:
Cu Selur Ps.: Attached a sample encode. PPs.: Yes, one can do more to filter this clip, but that is open for personal preferences. |
Site design, images and content © 2002-2026 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2026 Jelsoft Enterprises Ltd.