h a l f b a k e r y"Bun is such a sad word, is it not?" -- Watt, "Waiting for Godot"
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,
|
|
|
Most 16mm and 35mm movie film projectors work by flashing light through each frame of film twice, moving the film very quickly by a full frame, then flashing twice again, moving again, etc. This starting and stopping can pose considerable stress on the film.
My proposal would be to design a projector
with a high-powered 48 flash/second Xenon strobe and a lens assembly that would rotate 24x/second. The lens assembly would rotate about a point that off-center by a quarter of the frame spacing, and would be carefully balanced to minimize vibration; it would include electronic servo controls to adjust focus, aspect ratio, etc.
In theory, the effect should be that each frame of film gets flashed twice on the screen with reasonably good registration, while the film itself moves smoothly through the machine without any starting or stopping. The Xenon tube would have to operate off high voltage to ensure adequate brightness with a short flash duration, but I believe such devices exist.
As to whether all this is practical... well, this is the Halfbakery...
DLP mirror lifetime white paper
http://www.dlp.com/...papers/135_Myth.pdf
DMD: DIgital Micromirror Device-TM [csea, Jan 27 2006]
[link]
|
| |
Larger format films (what's that format
which uses 72mm film but on it's side -
so the film spool is horizontal and it's
the height, not the width, of the film
that's 72mm?) have an exagerrated
form of this problem with the need for
even faster film transports where the
normal stop-go-stop-go action would
shred a film. It's solved with a
continually moving film and a rotating
square prism which tracks a film frame
as it moves past the gate and projects it
as a stationary image. |
|
| |
wear a flimsy mini-skirt to promote your image - (Minimize
film dress during projection) |
|
| |
[Hippo] - that would be IMAX. It originally used (and in most installations still uses) a "rolling loop" transport system which was much better than the gate and pawl system still used on 35mm and smaller formats. The rotating prism is even less damaging to the film. |
|
| |
As a side-effect, the rotating prism method allows transition from one frame to the next with no black time. |
|
| |
The rotating prism method was originally developed for use in ultra-high-speed film cameras (thousands of frames per second, needing half a reel just to get up to speed) that simply couldn't stop the film without tearing it apart. |
|
| |
//This starting and stopping can pose
considerable stress on the film.// The
projector transport system is designed
to keep the film moving continuously,
except where it moves by the gate. The
intermittent sprocket and geneva gear
is very efficient, a design feature
persisting 100 years and moving
millions of miles of unbroken celluloid
film. |
|
| |
//a projector with a high-powered 48
flash/second Xenon strobe// Cinema
projectors already use xenon bulbs:
large, fragile, expensive lamp units
operating at dangerous heat, pressure,
and electric levels. A Xenon flash
system with equal lumens output would
require a robust lamp and large
capacitor. |
|
| |
//a lens assembly that would rotate
24x/second// Cinema lens are large,
heavy, fragile, expensive components of
the film projection system and require
critical allignment at all times. There
are two projectors in a booth which may
have two to five lens each for different
film and screen formats. The idea of
increasing this number 24x is extremely
unpractical. |
|
| |
As [Freefall] describes, the rotating
prism system exists, common on
projectors used when mixing a movie
soundtracks where the film print is
''rock-and rolled" in forward and
reverse. The benifit of moving the film
faster in a projector was demonstrated
by Douglas Trumbell's Showscan system
which moved 65mm film 60fps. |
|
| |
I've been a 35mm projectionist for 25
years and was a student of Dr.
Harold Edgerton, inventor of the
xenon flash and a high speed
photgraphy pioneer. I could go on and
on on these subjects. stop me now.. |
|
| |
I'm aware that only a small length of film has to be started and stopped, but the accellerations required are pretty severe. Moving the film by 2.5cm in 5ms, assuming uniform accelleration, requires that it be accellerated to 25 meters/second in 2.5ms, and decellerated just as fast. That's 1000g by my figuring. Even if the length of film in question is fairly light, I'd still think that's a fair bit of stress. |
|
| |
As for the rotating prism concept, I have a film viewer that operates on that principle. It works, but I thought I'd suggest the moving lens assembly as an interesting alternative. At least on my viewer, the edges of the frame did not keep position perfectly as the prism rotated; professional ones might be better, but I don't know how much. If the lens assembly had good bearings, rotated smoothly, and were properly aligned, it should be able to project two images that differed in final screen position by half the film frame spacing; adjusting the lens to be slightly non-axial could reduce even that. |
|
| |
Again, probably not practical, but I'd still think it somewhat interesting as a notion anyway. |
|
| |
BTW, Cube, where were you getting the idea of increasing the number of lenses 24x? Just one lens, used twice for each frame. Since it would be rotated 180 degrees between uses, anamorphic characteristics should be the same both times. |
|
| |
Also, out of curiosity, how many cinemas still use two-projector booths? Most multiplexes use platter reels with a single projector, don't they? |
|
| |
There are always stresses on the film,
from the moment the projectionist pulls
down a leader till the reel is rewound
and back in the can. A century of
practice has shown Edison's machine
distributes the stresses nicely. If the
film is stressed to much, significant
stretch takes ages to accure, or past the
tolerence of the film base substrate
might tear or break. If the film did not
break, a transport accident might
damaged projector components.
Archive prints sometimes have enough
sprocket hole damage to be
unprojectable and this idea might be
useful in those cases. |
|
| |
Now that I think about it, if the light
was strobing, you would not need to
rotate the lens I misunderstood
thinking 24 lenses. Still, I have used
projectors with rotating lenses (flat and
anamorphic) and it is hard to imagine
the fragile assemblies spinning around. |
|
| |
Its true, multiplexes and cinemas
showing movies for several shows a day
use platter systems with one projector.
However, all high-end projection
facilities use two projectors with the
projectionist doing changeovers
between the reels every 20 minutes. |
|
| |
the prints don't need to last forever, most films are in and out of the theatre so quickly that it doesn't make sense to upgrade all the projectors to stop stress on films; the majority of which are destined for the ashcans of history. |
|
| |
Besides, there just crappy release prints. |
|
| |
Dinosaurs! ;) But interesting... |
|
| |
Some quibbles:
A 48Hz strobed Xenon lamp in a projection booth seems unlikely? Would the lifetime be the same or better than a constant-on lamp? Regular Xenon lamp strikes cause all sorts of RF and EMI problems in the audio chain, if not well shielded. Could be pretty noisy! |
|
| |
The duration of a Xenon discharge can be varied, but by the time it is over 10 ms, it's likely to need fairly complicated contol circuitry (not just capacitor discharge.) |
|
| |
Note: During film sound mixing, as many as 48 or more reels of magnetic film may be synchronixed, and run continuously in either direction at up to 10x 24fps in either direction. These "dubbers" have to be very gentle, and deal with splices well. And work 24/7/365 with little maintenance. Some of these old dinosaurs are still in business after 50 or more years, though many are being replaced by much smaller digital and solid-state versions. |
|
| |
Attack of the pesky pedant.... sound of irritating whine,
then a satisfying ping as the pedant lands and starts to
feast on bristolz's annotation - there = they're. |
|
| |
All this talk of wear-and-tear - does anyone have figures for the life-span of the hinges on the mirrors of the TI DLP chips used in digital projectors? |
|
| |
Oops. There! They're better. |
|
| |
I will post a [link] to an interesting white paper on the TI DLP mirrors - how's a trillion cycles +? |
|
| |
For new release prints that are going to be obsolete before they wear out, there's no need for any fancy projection equipment to minimize wear. And really precious archive prints shouldn't be exposed to the light levels required for projection. |
|
| |
But for prints that are somewhere between these extremes, it might be nice to be able to project them (even if not on a 40-foot theater screen) without the stresses imposed by the jerky film advance. This would be especially relevant for films that are already in poor condition. Indeed, it might even be possible to project films where portions of the sprockets have been torn off if the inside edges where the sprockets used to be are in good enough shape to allow accurate registration. A prism-based projector might be able to do that, but a jerk-advance projecter certainly could not. |
|
| |