|
Post by George D on Jan 3, 2018 10:22:26 GMT
|
|
|
Post by Hugh Pearson on Jan 3, 2018 14:38:33 GMT
I might be wrong on this, but I don't think that 16mm film necessarily runs at 16 fps. Nigeria appears to be on the PAL or PAL/SECAM system - therefore, the frame rate would have been 25 fps. Just being pedantic
|
|
RWels
Member
Posts: 2,857
|
Post by RWels on Jan 3, 2018 15:51:13 GMT
Wel quite, I don't know of any TV system that ever ran at 16 fps!!
|
|
Simon Collis
Member
I have started to dream of lost things
Posts: 536
|
Post by Simon Collis on Jan 3, 2018 23:11:58 GMT
I thought the point of the original mention of 16fps was that roughly every other frame was usable, that's the level of motion that would result? Or did I misread it?
|
|
Simon Collis
Member
I have started to dream of lost things
Posts: 536
|
Post by Simon Collis on Jan 3, 2018 23:21:48 GMT
Wel quite, I don't know of any TV system that ever ran at 16 fps!! Just out of curiosity I googled this. Switzerland used a 16.6fps 30 line mechanical system in 1932, will that do?
|
|
RWels
Member
Posts: 2,857
|
Post by RWels on Jan 3, 2018 23:59:20 GMT
Wel quite, I don't know of any TV system that ever ran at 16 fps!! Just out of curiosity I googled this. Switzerland used a 16.6fps 30 line mechanical system in 1932, will that do? Well I never! On the other hand, it was not an unusual speed for film in the early 20th century.
|
|
|
Post by Sue Butcher on Jan 4, 2018 6:18:14 GMT
16fps is about the minimum film frame rate you can use and still have images move smoothly when projected. I think it was the usual standard for silent films, but films went to 24fps when optical sound came in because that required a higher speed for good audio bandwidth.
|
|
|
Post by brianfretwell on Jan 7, 2018 16:20:34 GMT
I think I saw on one site that sometimes BBC film recordings were made at 16.67 fps (I think at Lime Grove), but for only viewing in-house not broadcast.
|
|
|
Post by brianfretwell on Jan 7, 2018 16:22:28 GMT
Equally - at that frame rate you may be able to create some form of computer programme (if it doesn't exist already) to build an low res image to fill the gaps and help rejoin where frames were cut etc. This should be more achievable for realism, seeing as it's greyscale rather than colour and help the image blend in more easily at 16fps. Another use for the Vidfire process perhaps.
|
|
|
Post by Sue Butcher on Jan 8, 2018 4:47:42 GMT
Computer inbetweening was good enough to fill in that missing second in War Machines, and it may be better now. My guess is you could get by with only ten good frames per second in a damaged film, maybe less if you do some manual tweaking of fast hand and face movements. One of the problems with the M&W film is the geometric distortion of the image where the emulsion has sagged. It changes from frame to frame, so it's going to hard to correct.
|
|
|
Post by martinjwills on Jan 8, 2018 10:20:07 GMT
when you convert 29.97fps to 25fps. the dropped frames show on fast movement but not slow movement, Pal Speedup works on the 23.98fps to 25fps by altering the audio so it does not copy [add] or drop [cut] frames. Some of the damage on this print could be repaired by the damaged bit of the picture replaced from the frame each side. Some software takes frames 1 and 3 and will produce an intermediate between the 2 frames replacing the middle frame, so this would be near to the missing original frame 2. That should cope with faster movement scenes without the jurking. Im sure if the difference between frames 1 and 3 is 0%-2% a copy could be added which would speed up the production of frames.
Many years ago Digipaint on the Amiga computer, had a rub through option, which was very useful, as you could have 2 pictures loaded and rub the damaged bit with the pointer, and the back image would replace the bit on the front image, so you would be able to rub the damage off Erics face in close detail with a small pointer, without touching the other parts of the frame.
If all of the M&W film has been scanned and stored, and put into some order, it could be worked on over future years, and software produced to work on sections, similar to the early chroma dot recovery in the 1990s
|
|
|
Post by John Wall on Jan 8, 2018 10:29:03 GMT
A programme of that type would have been a series of sketches - I suspect there will nee more images for some than others.
|
|
|
Post by George D on Jan 11, 2018 10:05:59 GMT
In the article it says. "All 37 scan files were processed and over 5000 image files were produced. "
While it may appear as a lot, for a 50 minute program,unfortunately this is apx. 100 pictures a minute or less than 2 frames per second. Assuming we get 3 pictures out of a image file. That would bring it up to a frame rate of 6 frames per second max. Great that we got this.. however, we should have minimal expectations of the output result.
|
|
RWels
Member
Posts: 2,857
|
Post by RWels on Jan 11, 2018 13:39:00 GMT
In the article it says. "All 37 scan files were processed and over 5000 image files were produced. " While it may appear as a lot, for a 50 minute program,unfortunately this is apx. 100 pictures a minute or less than 2 frames per second. Assuming we get 3 pictures out of a image file. That would bring it up to a frame rate of 6 frames per second max. Great that we got this.. however, we should have minimal expectations of the output result. It's a 25 minute show, and these pictures apparently are all from 8 minutes. 625 p/m or 10 p/s - on average; still not a huge frame rate, but there'd be some illusion of movement at least.
|
|
|
Post by George D on Jan 11, 2018 13:44:33 GMT
So basically it is believed approximately 8 minutes of the film can have some sort of reconstruction and not the whole episode?
|
|