[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Sony telecine

Hans Lehmann asked why I was concerned about area array telecines and

A line array CCD telecine (i.e. virtually all CCD machines from the dawn of
history to the Spirit) uses a detector which is a single television line
high, building up the full television picture by moving the film past a
slit. A decent servo system can ensure that this happens at a constant
speed, giving good vertical stability.

An area array is like a CCD video camera: it grabs the whole frame in one
instant. That is not a problem in a video camera, when what the CCD is
looking at is moving anyway. The telecine, of course, does not do that: it
aims to capture a succession of still pictures. The film frame will have to
(appear to be) perfectly stationery in front of the area array for the
length of time it takes the CCD to scan it.

As far as my simple mind can see, there are two ways of achieving this.
Either there is a continuous film transport and a very fast shutter or
there has to be a non-continuous motion, i.e. something like a pin gate,
which advances the film one frame, exposes it to the CCD, then moves on to
the next.

Given sufficient light through the film, the exposure time can be very
short indeed, but both solutions seem to me to be asking a lot. In the case
of the shutter, it has to be precisely synchronised to something in the
two-thousandth of a second range (i.e. less than a video line at high
definition rates).

For the step motion solution, well, there has already been plenty of
discussion in this forum about the problems of pin registration, and that
is without trying to do it at real time. Avoiding damage to the film
becomes an issue here, too.

So, as I said, I look forward to see how those undeniably clever people at
Sony have solved this problem.