4 ms·

Why do I feel like there's a guy here that could take film restoration to the next level. I use a personal favorite movie The Boy in the Plastic Bubble as an e

Why do I feel like there's a guy here that could take film restoration to the next level. I use a personal favorite movie The Boy in the Plastic Bubble as an example. I don't know why I like this movie as much, but I do. I just looked it up. It was filmed, or video taped, in 1976. It got four Emmy nominations.) It was shot on video I believe? As a child, I recall it being a movie of a week film. It looked crystal clear to me, but then again my perception might have been different? For years, I've seen the movie on PBS type channels. I think it's due to the dedgration of the original video, or film. It looked terrible. It was fuzzy, and barely watchable. I don't believe the remastered versions, if they exist, are good enough quality for the media companies? Someone did a better job at restoration about two years ago, but it's not nearly as sharp, and vivid, as I recall, and it's only watchable in a small format. Every time I watch it I get interested in video restoration. My point is film/video restoration seems like it has so much room for research, and innovation?

dylan604 5y ago
> My point is film/video restoration seems like it has so much room for research, and innovation? From someone that has spent significant amounts of time in film/video restoration, innovation has been made. The ability to remove scratches/hair, color restoration, etc has come a long way (to almost be laughable at how easy it is with the right tools). That being said, the restoration can only do so much. Each recording medium/format has its own "quirks" or "characteristics". For example, 16mm has a very noticeable grain pattern compared to 35mm. Also, each type of film stock has its own color/exposure traits. Silver nitrate film has a different look than other film stocks, plus has the added bonus of being able to burn your facility down if handled improperly! Then we get to video. Ugh! If this program that you are remembering was originated on video and is old enough, then it's only going to be a shiny turd after restoration. Watching old pre-90s SD content on a small screen hides a lot of the imperfections. Trying to uprez that content to HD or scaled to fill a larger modern screen will reveal all of those imperfections. The old SD video formats just did not have a lot of data to begin with, so making something out of nothing has only really been possible recently with some of the ML packages. If you really want to go down the rabbit hole, look up the specs on the old video formats like 3/4" Umatic, 1" Type C, Betacam, Betacam SP, D1, D2, Digibeta, etc. That list is from older->newer, and by the time we got to Digibeta, SD was looking as good as it was going to get. That's still a far cry from the data we can now generate. Add to those limitations that every time you made a copy of it, you made that image slightly worse. BetacamSP and newer could handle dubs much better than the others, and Digibeta/D1/D2 could make actual lossless clones.
- zrail 5y ago
  Iirc this is why there will never be an HD version of Star Trek DS9. They filmed on actual film for much of it but the masters of the special effects are all on video tape.
  - WalterBright 5y ago
    Just redo the special effects, like they did for ST TOS.
    - dylan604 5y ago
      I love the use of the word "just" like this. One of the prroducers I worked with would double of the quote for any client that uses the word "just" when trying to downplay the amount of effort. The producer(s) of ST TOS would have to know that the money earned from doing that work would be worth the effort. It could easily cost more than they'd make.
      - zrail 5y ago
        Yep. 79 vs 176 episodes. Much more complicated effects. Smaller fan base. There's no way it would be financially viable.
pfranz 5y ago
https://www.imdb.com/title/tt0074236/technical?ref_=tt_dt_spec https://www.imdb.com/title/tt0074236/technical?ref_=tt_dt_sp... imdb claims it was shot on 35mm film. It also says "printed film format: 35mm" so there's probably a fairly high quality master out there assuming it still exists. Like you've noticed, the studio often gives whatever copy is handy instead of digging up the highest quality version closest to the source. Heck, in this article they were doing a 4k restore on a movie from the 70s but didn't get a release print until the director mentioned an audio effect was missing. > it's only watchable in a small format. By that do you mean 4:3, the almost squarish format? If so, that's how it was shot and how it will likely always be presented.

dylan604 5y ago

> My point is film/video restoration seems like it has so much room for research, and innovation? From someone that has spent significant amounts of time in film/video restoration, innovation has been made. The ability to remove scratches/hair, color restoration, etc has come a long way (to almost be laughable at how easy it is with the right tools). That being said, the restoration can only do so much. Each recording medium/format has its own "quirks" or "characteristics". For example, 16mm has a very noticeable grain pattern compared to 35mm. Also, each type of film stock has its own color/exposure traits. Silver nitrate film has a different look than other film stocks, plus has the added bonus of being able to burn your facility down if handled improperly! Then we get to video. Ugh! If this program that you are remembering was originated on video and is old enough, then it's only going to be a shiny turd after restoration. Watching old pre-90s SD content on a small screen hides a lot of the imperfections. Trying to uprez that content to HD or scaled to fill a larger modern screen will reveal all of those imperfections. The old SD video formats just did not have a lot of data to begin with, so making something out of nothing has only really been possible recently with some of the ML packages. If you really want to go down the rabbit hole, look up the specs on the old video formats like 3/4" Umatic, 1" Type C, Betacam, Betacam SP, D1, D2, Digibeta, etc. That list is from older->newer, and by the time we got to Digibeta, SD was looking as good as it was going to get. That's still a far cry from the data we can now generate. Add to those limitations that every time you made a copy of it, you made that image slightly worse. BetacamSP and newer could handle dubs much better than the others, and Digibeta/D1/D2 could make actual lossless clones.

zrail 5y ago
Iirc this is why there will never be an HD version of Star Trek DS9. They filmed on actual film for much of it but the masters of the special effects are all on video tape.
- WalterBright 5y ago
  Just redo the special effects, like they did for ST TOS.
  - dylan604 5y ago
    I love the use of the word "just" like this. One of the prroducers I worked with would double of the quote for any client that uses the word "just" when trying to downplay the amount of effort. The producer(s) of ST TOS would have to know that the money earned from doing that work would be worth the effort. It could easily cost more than they'd make.
    - zrail 5y ago
      Yep. 79 vs 176 episodes. Much more complicated effects. Smaller fan base. There's no way it would be financially viable.

pfranz 5y ago

https://www.imdb.com/title/tt0074236/technical?ref_=tt_dt_spec https://www.imdb.com/title/tt0074236/technical?ref_=tt_dt_sp... imdb claims it was shot on 35mm film. It also says "printed film format: 35mm" so there's probably a fairly high quality master out there assuming it still exists. Like you've noticed, the studio often gives whatever copy is handy instead of digging up the highest quality version closest to the source. Heck, in this article they were doing a 4k restore on a movie from the 70s but didn't get a release print until the director mentioned an audio effect was missing. > it's only watchable in a small format. By that do you mean 4:3, the almost squarish format? If so, that's how it was shot and how it will likely always be presented.