Path: utzoo!attcan!uunet!husc6!mailrus!purdue!decwrl!labrea!polya!kaufman
From: kaufman@polya.Stanford.EDU (Marc T. Kaufman)
Newsgroups: comp.graphics
Subject: Re: Film Recorder Wanted
Message-ID: <4981@polya.Stanford.EDU>
Date: 11 Nov 88 07:18:11 GMT
References: <4071@psuvax1.cs.psu.edu> <325@ur-cc.UUCP>
Reply-To: kaufman@polya.Stanford.EDU (Marc T. Kaufman)
Organization: Stanford University
Lines: 37

In article <325@ur-cc.UUCP> haake@.cvs.rochester.edu (Bill Haake) writes:

>The Mirus is based on an electrostatic imaging screen which
>is supposedly very high quality. It also uses a color filter
>wheel and has 24 bits per pixel. It comes with a 35 mm film back
>but no optional backs are yet available. The addressable resolution is 
>advertised as "547 to 8191 lines". I think this means both horizontal
>and vertical but I am not sure. This is billed as an "all digital" 
>film recorder because the exposure level is determined by the
>length of time the screen is lit rather than the intensity of the 
>screen.  Unfortunately this results in very long exposure times.
>I went for a demo and we killed an 8000 line slide after about 30 
>minutes and it hadn't even finished the first pass of a three pass 
>RGB image. 15 min. for a 2000 line RGB image. SCSI interface. 
>List price $5995.

>Anyone have any comments? 

The choice of electrostatic or magnetic deflection is significant only to the
extent that the sweep is non-linear.  The Mirus generates nice rectangular
images.  Addressable resolution is 500-8000 points horizontally, 2/3 of that
vertically (if you want square pixels).  "normal" resolution is 2000 x 1333.
This compares with film resolution of ~40 line pairs /mm, which is 2800 lines
(not pairs) in 35 mm. (this for something like Ektachrome 100HC film).  It is
not particularly useful to attempt to image at resolutions higher than the film
can support.  Color tables are supplied for each film that provide gamma
correction (linearization of the perceived intensity steps).  Yes, it is slower
than some, but that is the normal complexity tradeoff -- which hopefully should
be reflected in the price you pay.

The basic interface is SCSI, with RGB raster images.  The Mac II driver
does the rasterization with Quickdraw.  There is not currently a SUN driver.
There is also a PC version with a Centronics type parallel interface.

DISCLAIMER:  I am biased.  I wrote the FilmPrinter firmware and the Mac Driver.

Marc Kaufman (kaufman@polya.stanford.edu)