[argyllcms] Re: A bit of fun with a Datacolor 1005
- From: Philip Pemberton <philpem@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Wed, 20 Mar 2013 20:30:27 +0000
On 20/03/13 01:45, Ben Goren wrote:
> I should think a great deal will depend on why you're profiling the film
> and what you're planning on doing with it.
I'm actually profiling the combination of film recorder and film...
The film recorder has two configurable variables:
* Luminant - how bright the CRT is. Mainly used to correct for loss in
the colour filters and optics (it's constant for each film type). The
idea is to keep this high enough to keep exposure times down while
keeping blooming to a minimum.
* Exposure time - self explanatory. This is represented as a
positive-delta-encoded 256x3 table. In other words, you start at zero,
then each successive value is added onto the last to get the time for
that pixel value. This is what allows the film response to be
compensated for.
The goal is to get the thing as close to sRGB as is reasonably possible.
The catch is that every single film type Polaroid created a profile for
is now discontinued and the filmtable file format isn't documented.
Thankfully I've figured out enough to make a linear film table (or
indeed any table I need)...
It's been a pretty big reverse-engineering exercise, I've wasted several
packs of Fuji FP100C instant film, but I'm close! Just need to get the
calibration "good enough" to get a recognisable image, then profile it
like a printer and fix up the last few errors.
I'm not expecting 1:1 repeatability, just "good enough to use".
Thanks,
--
Phil.
philpem@xxxxxxxxxxxxx
http://www.philpem.me.uk/
Other related posts:
