Any possibility of repairing this video? (screenshots in link)

[Darth Moore]

The earlier script gets rid of the strong, thin, red lines in those areas. But not the weaker reddish blotches. I tried several things but they damaged the rest of the picture too much. I may try a few more things over the next several days...

Good day to you ! May I ask if you had any chance to had a final look at it in hope you may can do some miracle ? BIG thank you for your time and kind help in advance!!

[jagabo]

Sorry. I played around with a few more things but didn't find anything that would remove the red blotches without also screwing up other parts of the frame. But the earlier script does remove some of the worst thin red lines.

[Darth Moore]

Sorry. I played around with a few more things but didn't find anything that would remove the red blotches without also screwing up other parts of the frame. But the earlier script does remove some of the worst thin red lines.

Fully understand and BIG thanks for trying. Allow me to ask a last thing:

After I had opened your AVS file and the video loaded with your correction script into VirtualDub2, then: to ensure quality loss and produce exactly same parameters MOV file as original including the corrections of course, I just have to : "Save video" and wait until its save it, or Export or ?

Sorry for the dummies' question.

[jagabo]

After I had opened your AVS file and the video loaded with your correction script into VirtualDub2, then: to ensure quality loss and produce exactly same parameters MOV file as original including the corrections of course, I just have to : "Save video" and wait until its save it, or Export or ?

Yes, you can use VirtualDub2 to save the video. But you will incur some quality loss if you use a lossy codec (no matter what program you use). Only with a lossless codec can you get a new video with no further losses. But lossless codecs deliver much bigger files. On the order o 10 times bigger.

[Darth Moore]

So there is no way to save the movie exactly as:
MOV file
3840x2160
Data rate: 27601 kbs
Total bitrate: 27601 kbs
Frame rate: 29,97 frames/second
H264 (baseline @L4.0)

Exactly as the source file BUT with your script correction ?

If yes, could you be so kind to help me step by step how ? If not possible, how can i get the closest possible ?

And again, many many thanks for your kind help and patience!

[jagabo]

So there is no way to save the movie exactly as:
MOV file
3840x2160
Data rate: 27601 kbs
Total bitrate: 27601 kbs
Frame rate: 29,97 frames/second
H264 (baseline @L4.0)

Exactly as the source file BUT with your script correction ?

You can SAVE it like that but it will be lower quality than the original video. This is always the case when using lossy codecs.

If yes, could you be so kind to help me step by step how ? If not possible, how can i get the closest possible ?

1) Open the script
2) Select Video -> Fast Recompress
3) Select Video -> Compression... Select the x264 8-bit encoder.
4) Press the Configure button. Configure the encoder the way you want. I recommend 2-pass VBR.
5) Back at the main VirtualDub2 page select File -> Save Video... Select "Save as type" to the container you want. Provide a file name for the new file. Save.

[Darth Moore]

Big thanks, followed it step by step, but it says when wanting to save it:


[Attachment 80967 - Click to enlarge]ű

Any idea why ?

[jagabo]

What option did you select in step 4? If you're doing a multipass encode you must first save with Multipass 1st Pass (fast) (or the non-fast version), the save again with Multipass Nth Pass. During the first pass the codec examines the video so it knows what every frame/scene looks like. Then in the second pass it performs the actual compression, distributing bits as determined by the first pass.

If you still have problems post a screen shot of the configuration dialog.

This video may indicate you have an invalid frame width or height, or an invalid color format:

https://www.youtube.com/watch?v=do2eycUKUxg

[Darth Moore]

You were again right, THANK YOU, the problem was at first, we selected multipass. As soon as we selected 1 pass, it started the process. We have the following settings:


[Attachment 80978 - Click to enlarge]

What still not clear (due to our zero knowledge in this subject, which we apologize for) whether these settings will do deliver to the closest to lossless (with your script correction of course) keeping these parameters:
MOV file
3840x2160
Data rate: 27601 kbs
Total bitrate: 27601 kbs
Frame rate: 29,97 frames/second
H264 (baseline @L4.0)

or not, and we should use different settings ?

And if we are pre-calculating well, the estimated video size will be around 30 gb, while the original is 54 gb. Does it mean quality loss, or just means more efficient compression ?

(forgive us for stupid questions, but we are missing fundamentals here about video framerate/compression/etc, as we never did something like this, and this is the only video we are trying to fix)

Finally, when these process finish, we doesnt do load/change anything else, just go straight to compression, set multipass and again save video, correct ?

Many many many many many thanks for your time and help !!!

[jagabo]

Encoding at exactly the same bitrate and other settings will not deliver the same quality. Lossy encoding with AVC will always deliver lower quality. The higher the bitrate and other settings you use the less losses you will incur. But if you go too high with the settings you will find the resulting video will not play properly on some devices.

The Placebo preset will definitely use settings incompatible with a lot of hardware players. And it is very very slow. I recommend using Medium and see what you think. You probably won't notice the difference on casual viewing. I would set the Profile and Level to Auto.

Another way you can encode is "Single Pass - ratefactor-based (CRF)" in VirtualDub2, just called crf or rf in some other programs. This is a "quality" based mode that works in a single pass. The lower you set the CRF value the better the quality will be, reaching lossless at CRF=0. Around CRF=18 you will get something that is very close to the source at normal playback speed. At around CRF=12 it's very hard to see differences even zooming in and comparing still frames. The latter will likely end up much bigger than your source though.

Keep in mind that the filtering causes some damage to the video. It removes most of the strong thin red lines but causes some of the red artifacts to smear over a larger area.