World's First Color Moving Pictures Discovered 105
BoxRec writes "The BBC is reporting newly-discovered films made by pioneer Edward Raymond Turner from London, who patented his colour process on 22 March 1899." When Turner invented his process, though, existing projection systems weren't up to it; to see the discovered footage, British archivists digitized the film for computer playback. When you're used to old films being both black and white and jerky, it's amazing to see it in color and (relatively) smooth.
Forgotten Silver (Score:4, Funny)
1899? That'd be even earlier than Colin McKenzie's film, which I believe was 1911 ... I'd have to rewatch Forgotten Silver to confirm it, though.
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
Hurray color! Let's party like it's 1899!
And when you're used to modern video... (Score:5, Interesting)
Re: (Score:3)
Thought the same exact thing. Don't know whichidiot marked you as a troll. On an iOS device myself. Happens way too often.
Re: (Score:3, Funny)
Funny... works fine on my Linux box :P
Re: (Score:3)
Didn't realize this was a flash issue, 90% of the time it's youtube just doing the run around to please content providers (i.e. on music videos - plays on atomic web brower with different user agent). Not that it matters, found a better and playable video around here of this story.
Not sure I would call flash a "standard" though.
Re: (Score:3)
Flash is an abomination. Video doesn't need client side programming. It just needs a decent player like mplayer and a video file in a standard format (like Dirac [wikipedia.org], Theora [wikipedia.org], or VP8 [wikipedia.org]).
Re: (Score:3)
those three examples are no more standard than flash!
Re: (Score:2)
That's what you both get for using an iThing.
Re: (Score:2)
Incredible (Score:3, Insightful)
Re:Incredible (Score:5, Interesting)
Re: (Score:3)
rotating color wheel front on the projector similar to DLP projectors today
That was one tech they investigated when trying to invent the color TV back in the 1940s.
It wasn't just "investigated".... (Score:5, Informative)
It was actually ADOPTED as the official US color broadcast standard by the FCC from 1950-1953.
http://en.wikipedia.org/wiki/Field-sequential_color_system [wikipedia.org]
The main limitations of the CBS field-sequential system were the requirement for a rotating color filter wheel more than 2X the diameter of the picture tube. TV sets larger than 10" screen size or so became absolutely HUGE. The system was also incompatible with existing monochrome sets, which already had a substantial installed base by then.
Once RCA developed the all electronic system that eventually became "NTSC", the field sequential systems were relegated to niche applications such as the color cameras that flew to the moon on the Apollo landings. And yes, a similar system forms the heart of modern color DLP projectors.
Re: (Score:2)
information is only lost in the sense that information is lost when going from microcassette to 48khz audio. you're not losing anything that was meant to be there, or was even recorded.
you barely got 300 lines (columns in today's speak) in a composite NTSC signal. compare that to the 702 active (and 720 total) specified by D1 and all after. that's plenty of room.
Re: (Score:2)
You are corrrect, except the 300 lines. That's VCR resolution. NTSC is 540 lines. My ten year old 42 inch flatscreen CRT can do 720p, bty placing the scan lines closer together, leaving a black band at the top and bottom; smaller screen with better resolution. Unfortunetaly for me, it won't work with broadcast TV, only a DVD.
Re: (Score:2)
I believe field-sequential wheels continue to work with the proper kind of tube-driven black-and-white TV and an analog color NTSC signal.
Re: (Score:2)
I didn't know that, thank you for the education!
Re: (Score:2)
I agree, you can definitely see some sort of RGB separation on the moving areas.
It could easily be synced to the film transport. I don't see why sync would be a problem.
Re:Incredible (Score:5, Informative)
Sync is a problem because the objects are moving! The only way around it with B&W film is to have three simultaneous cameras shooting through color filters.
Re: (Score:3)
Or a color filter that rotates in some way. One way could be a strip of filter film in a loop (of some multiple of 3 frames) the same size as being shot (35mm ?) being rotated with a mechanism similar to the film transport mechanism. It just needs to syncronize the two film mechanisms together in the camera, and expose an extra marker somewhere to show where the start is. That or someone guesses the start later on through use of standard color chips in the slate. Once that is done, then it's just a matt
Re: (Score:2)
With a rotating color wheel, each color frame is exposed at a slight delay from each other by necessity.
Re: (Score:2)
But with a color filter loop film, the filter frame would have the same mechanism that holds the film still for each exposure. Just run the two like mechanisms locked together from the same crank or motor. It would require twice the force to do it.
Re: (Score:3)
And between the time that red is exposed and then green is exposed, fast moving objects have moved. You're getting a different picture. It's not about the pictures being lined up.
Re: (Score:2)
The only way around it with B&W film is to have three simultaneous cameras shooting through color filters.
Which is exactly what Technicolor [wikipedia.org] was, simultaneously photographing two consecutive frames of a black-and-white film behind red and green filters.
Re: (Score:3, Informative)
Really ingenious. This means you're not trying to do the same camera operations at the same time with three separate cameras
Re: (Score:2)
ingenius until you decide you want to shoot the Bourne trilogy (quadrilogy?) in Technicolor... those things were HEAVY.
people complain about how much a tricked out RED one weighs, but these things...
try shooting 6 simultaneous 70mm films for technicolor 3D!
Re: (Score:2)
Re: (Score:2)
Use the same mechanism that is used to transport the film, to transport color filters arranged in a film-like loop (multiple of 3 frames). Then interlock these mechanisms so you don't end up skipping or double shooting a color during shooting. Put color chips on the slate at the start and leave the camera rolling from slate to program. The film, of course, would be at the objective focal plane. The color filters would be out of focus just behind the lens (this would probably still handle an aperture up
Re: (Score:3)
nah, if you're able to build a pin-registered gate at all (and run it at 24 fps without destroying the film), you've already got the skills necessary to add a filter wheel at the same speed.
the engineering in old film gear is just phenomenal. awe inspiring that they could make all this stuff work together.
even the capstan-servo telecine machines of the 70s (and still today) are incredible. they could get 1200 feet of heavy film to move at _exactly_ the right constant speed to get exactly 576 lines per fil
Re: (Score:3)
Wikipedia says hand colored films began in 1895 with Thomas Edison. This isn't hand painted though. Anyone with photography knowledge have an explanation?
Explanation for what? None of those things contradict the other.
It looks to me (having not listened to the audio track) like it was shot through rotating red/green/blue filters, which results in some slightly psychadelic colour trails on moving objects but some remarkably clear full colour on still objects.
Re: (Score:2)
through rotating red/green/blue filters
Or cyan/magenta/yellow.
Re: (Score:2)
RGB : http://www.nationalmediamuseum.org.uk/PlanAVisit/Exhibitions/LeeAndTurner.aspx [nationalme...eum.org.uk]
Re: (Score:2, Informative)
Color mixing is different with pigments than with light. In pigments, the primaries are red, yellow, and blue. In light it's cyan, magenta, and yellow.
Re: (Score:3)
Color mixing is different with pigments than with light. In pigments, the primaries are red, yellow, and blue. In light it's cyan, magenta, and yellow.
Ah crap. You should really tell display manufacturers, they've been doing it wrong for decades!
Re: (Score:2)
They're using the right colors, just slightly naming them wrong. magenta is a purplish red, cyan is a greenish blue. (I learned this stuff in an undergrad physics class, one of the most interesting classes I took. The part with lasers and holograms was especially cool.)
Re: (Score:2)
They're using the right colors, just slightly naming them wrong. magenta is a purplish red, cyan is a greenish blue. (I learned this stuff in an undergrad physics class, one of the most interesting classes I took. The part with lasers and holograms was especially cool.)
I learned about this in primary school art class, and you still have it backwards.
Hold a magnifying glass to some white area on your display, and you will
see very nice red, green, and blue subpixels. Ok, this might be trickier with
high-density displays nowadays. Use a good magnifying glass...
You'll see nothing purplish about the red or greenish about the blue. There's a perfectly
fine green subpixel right next to it anyway, so making blue greenish would also be
stupidly reducing the display's gamut.
I
Re: (Score:2)
nope. other way round.
but the problem is we're shooting negative film...
however, light is light, and on it's way into the camera it has to go through RGB filters.
film is developed, and you end up with the negative (CMY).
film is _printed_ for viewing, and you're in RGB again.
Re: (Score:2)
Yes, that's how film works, but mix cadmium red with cadmium yellow (the reddest and yellowest pigments) and you get bright orange. Mix cadmium red and cobalt blue and you get a deep purple. Mix cobalt blue and cadmium yellow and you get bright green.
Re:Incredible (Score:4, Informative)
Re:Incredible (Score:4, Interesting)
If projecting three colors, even if in different time phases, you use red, green, and blue. Since those are primaries, this does give the best color saturation. But it also has the downside of reducing exposure more than secondaries (a problem that still exists even for today's digital camera through the tiny array of color filters).
Re: (Score:2)
you can squeeze extra dynamic range by reducing the density of the filter. you can correct for it with channel subtraction or in LAB space (so you can banish the noise to the chroma planes and keep a clean luma plane, where the detail is).
Re: (Score:1)
Re: (Score:2)
this is a way to automate the process to speed it up to film speed.
Re: (Score:3, Informative)
Why are you surprised? (Score:5, Insightful)
Re: (Score:1)
They have digitized it for the computer. They might have also fixed the transition and jerkiness
All they published is a still image. Easy to remove the jerkiness in that case...
Re: (Score:3)
Re: (Score:2)
They might have also fixed the transition and jerkiness.
They haven't. It's possibly slightly (though barely noticeably) smoother than the sort of pictures we're used to from that era, but I don't think it was worth commenting on if that happens to be the case.
Re: (Score:2)
They haven't.
They're trying to show the film in its original form, not cleaned up as much as possible. The film was taken one frame at a time through a color wheel, but the projector was supposed to show three RGB frames at a time through a color wheel. At each frame advance, three frames are shown, but they were not all taken at the same time. So the R, G, and B frames don't line up if there's any action. That's why the weird color jitter.
It's possible to do far more cleanup. See "Die Finanzen des Grossherzogs", by
Re: (Score:1)
And if that works, can they use the developed technique to fix the waving back and forth from every single release of the 1980 Heavy Metal movie?
Re: (Score:2)
linky?
if it's loud music and video, then it was the vidicon sensor physically shuddering as it resonated with the sound.
Re: (Score:2)
smoothness comes from being pin-registered. and being careful to deal with film shrinkage (while making sure it doesn't catch fire as it was undoubtedly Nitrate base film - which is probably why they copied it to bog standard 35mm first).
Much Better Video Available (Score:5, Informative)
YouTube has a much better video than the one linked in the article that contains the process they went through and talks about the capture and projection [youtu.be] intended by the inventor.
Re:Much Better Video Available (Score:4, Informative)
YouTube has a much better video than the one linked in the article that contains the process they went through and talks about the capture and projection [youtu.be] intended by the inventor.
I was going to provide the original link to the National Media Museum (which for the curious is here: http://www.nationalmediamuseum.org.uk/PlanAVisit/Exhibitions/LeeAndTurner.aspx [nationalme...eum.org.uk] ) ...but it's the same video anyway.
What intrigues me is that they apparently blew it to 35mm first instead of going straight to digital.
Re: (Score:2)
I thought the same thing. I would imagine they already had equipment to deal with 35mm film, and it was easier to transfer it to 35mm to feed that equipment rather than retrofitting the equipment to take a larger source.
I'm surprised they MANUALLY advanced each frame through the little shutter contraption. Don't any of these guys have a bag of Legos they could automate that process with???
Re: (Score:3)
If they had been able to scan the originals, that might yield a quality improvement - but there are so many things I'd love to try out with the raw materials.
I would be very interested in seeing what digital image processing might be possible to use - if one could mangle the three temporally separate frames into a luminance signal and a chrominance signal which interpolates using motion-compensation derived from luminance, that might temper the rainbow effect somewhat - and triple the temporal resolution!
Re: (Score:2)
modern print stock is about as good as it gets. even better if they can find black and white print stock.
but yeah, they could have done the scan with a fancy lightbox (integration sphere), that gate contraption and a really good digital camera and got more precision than the 10-bit you get from a film scanner (at most 16-bit linear with dual-flash scanning on a 12-bit sensor). they could even adapt the film scanner lenses to fit a DSLR...
but archivists make film based copies as a matter of procedure i thi
Re: (Score:3)
They explain that in the video, his film was not 35mm but 35mm equipment is standard and common. Thus it was cheaper and easier to transfer it to 35mm and then perform the restoration/digitization rather than building/adapting equipment and software for a one-off project.
Re: (Score:2)
Maybe, maybe not. If they were thinking inside the box and shooting frame by frame, then sure, they would benefit from doing a 35mm conversion, first. But, it might have been easier to just rig up a LINE SCANNER and pull the film across that scanner slowly. Then software can convert the line scan into frames and figure out the color sync. In later talkies, the software could also convert the optical audio side tracks, too. Full line scan would include everything out to the sprocket holes which would be
Re: (Score:2)
Oh, and the color film can also be the original color negative. Just get an extra LED at the color masking wavelength(s) and you'd have more accurate conversion of color to positive than even the original film printing process. Most motion picture films were shot on negative film with a process similar to photo films of the day.
Re: (Score:3)
I'm still looking for a genuine movie FILE, not some flashy thingy meant to frustrate people from saving it.
copyright? (Score:5, Insightful)
Wait its only been 113 years? Can I view that content without worrying about being sued by MPAA?
Re: (Score:1)
Disney's team of lawyers says no...
Re: (Score:2)
Worry more about the bots doing automated take downs now days.
The process was patented... (Score:4, Insightful)
if this had been in 2012, he wouldn't have patented a film process but instead followed Apple (and others) by patenting "The idea of colour moving pictures displayed to an audience" and his descendents would now be suing Hollywood for 15 gazillion dollars.
Re:The process was patented... (Score:5, Informative)
You know the reason Hollywood is in Hollywood? The film industry went as far away from Edison as they could in order to violate his motion picture patents.
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Glad to know I'm not the only one who noticed his last name, but the temporal color strobing would be tough to emulate if hand-coloring.
Re: (Score:2)
We can do this today with a synchronized bank of LEDs that flash in color rotation with the camera frames (black and white film or monochrome sensor).
"jerky" as a projection artifact (Score:2)
Plot Fail (Score:1)
"Hey, I got an idea: let's film some sugar-induced brats destroying table decorations!"
"Brilliant!"
Re: (Score:2)
Re: (Score:2)
The porn ones were stolen. This is what was left behind.
Re: (Score:1)
It is to Catholic priests (*ducks head*)
Re: (Score:2)
2. transfer it to 35 " film first before digitizing it.
angry reason: interneg loses detail.
Maybe, just maybe, these people know a little more about film restoration than you do.
More importantly though, where did you get this information from? It's not in the linked article, and it's not mentioned in the video. They only say they did it "digitally."
Know what makes me angry? People who don't capitalise the first word of a sentence, and fail to use apostrophes in words like "you're."
Second, actually (Score:4, Funny)
Misleading headline (Score:1)
"World's First Color Moving Pictures Discovered"
"...who patented his colour process on 22 March 1899..."
Moving pictures predated film by decades or millenia. The zoetrope was invented in 1833 according to Wikipedia -- http://en.wikipedia.org/wiki/Zoetrope [wikipedia.org] -- which also mentions a similar device in China in 180 AD.
Re: (Score:2)
You're off by 30,000 years:
http://news.discovery.com/history/prehistoric-movies-120608.html [discovery.com]
It's a recurring phenomenon (Score:2)