h a l f b a k e r yNot the Happy Cuddle Club.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Services that convert movie film to video often either use a five-bladed telecine projector to project 18 or 24fps movies, or else they use a step-frame projector which projects movies showly so they may be captured a frame at a time.
Although it would be necessary, when going to VHS tape, to show
18/24 movies at 18 or 24 frames/second, it would seem that when recording to digital media this restriction is unnecessary. One should be able to get many of the advantages of a frame-step conversion if one used a projector that ran at either 15 or 30 frames/second, and then used digital software to duplicate frames as needed to yield a correct 18 or 24fps.
Even when using a 15fps projector, this approach would be faster than a frame-step conversion; if using a 30fps projector, it would be faster than even a normal "real-time" conversion (depending upon the film quality, however, a 30fps conversion might put more stress on the film and/or result in more jitter than a 15fps one).
Perhaps someone already does this, but given that everyplace seems to advertise that they either use a five-blade or step-frame system, I'm not certainly not aware of it.
[link]
|
| |
Sorry, I don't see it. In the tele-cine chain you want to be real-time for live transmission traditionally. At a post video house they may be copying at an 4 fps rate. It's all about the registration performance of the 'projector', where each frame settles into position because of the movement of the side rails. Under a microscope you would see how sloppy things are at a high fps rate. The rails heat up and go out of position as well, even over the length of the film. That is why it is nice to slow down to 4 fps and use four or two registration pins when copying motion picture film. |
|
| |
A lot of post houses are moving towards scanners like the Spirit DataCine, Spectre Virtual Datacine or the Cineon systems which can run at many different speeds, both real-time and non real-time. Pin-registered transfer does matter much more when you have shot your film with a pin-reg camera, like the Mitchell-Fries conversion and a very few others. |
|
| |
I recognize that copying films at 4fps is apt to be better than copying it at 15fps, but it's going to take longer and be more expensive. My thinking was that running the film at a smooth and exact submultiple of the video rate (or else 1x) and adjusting speed by digitally duplicating frames would be an improvement over approaches which try to do the 3-to-5 or 4-to-5 conversions in real time. |
|
| |
The exact submultiple is fractional. |
|
| |
I'm glad there's some pro's on line here. I have a related issue that I wish to tie to this with your permission and without starting another topic.
I have been experimenting for a few years with spinning LED arrays and currently the world patent holder of this technique and I are working toward a one metre diameter 3D OLED screen prototype. This design boasts the best resolution, best 3D ability and best price in the world. It can be ANY size.To tie it into this column, a projector can aim its image at this screen and the screen replicates this image, even brighter and with colour control.
However, his invention uses an XGA signal instead of a photo pick-up array next to the LED array concept.
Have you come across this before? |
|
| |
To understand better: The screen you are proposing has an integrated receptor array and it receives projected light, processes it and then emits it back towards the projection source? Is that right? |
|
| |
Sounds awfully complicated. What's the point? |
|
| |
My computer already converts films at about 14 fps--and that includes DivX compression at 1750 kbps. |
|
| |
//The exact submultiple is fractional.// |
|
| |
Which means that the time between moving each frame of film and shooting it with the camera will vary; this can cause jitter and/or flicker. If all frames are handled the same way, this issue is eliminated. |
|
| |
Yes, Bristolz. I don't mind sharing this concept as it has to get out there. You can make a simple active circuit yourself with a LED, 2222, phototransistor. Do you see the exciting concept here? Do you see these arrays in mid-air flying around, describing a 3D Sir Jagger on stage? |
|
| |
You should probably post it as an idea on its own lest the discussion fracture this idea. |
|
| |