[Tig] Printerlights

Dan Tatut dtatut at marquise-tech.com
Fri Dec 10 17:21:44 GMT 2010

On 10.12.2010 04:10, Carl Skaff wrote:
> Sohonet http://www.sohonet.co.uk sponsors the TIG.
> Digital Vision is patron of the TIG.
> Colorist Directory at http://tig.colorist.org/wiki/Category:Colorist
> ====
> In the old days the labs used an Analyzer to report to the DoP how the
> exposure was on the neg.
> They had a setting they called 25/25/25 that was used for correctly
> exposed footage. And if they needed
> to turn it up to 28/28/28 that ment it was underexposed by one stop.
> (Well, you know were I'm heading. I could very well we wrong about the
> details but you probably understand what o mean)
> How is this done nowadays? If you are using a scanner and a DI pipeline.
> Any comments appreciated.
> /Carl
> (Sent from mobile device)
> Carl Skaff
> ___________________________
> Head of Telecine
> Stockholm Postproduction
> www.stopp.se
> phone: +46 8 50 70 35 00
> fax: +46 8 32 77 22
> _______________________________________________
> http://reels.colorist.org
> http://tig.colorist.org/wiki3

Once the image has been scanned as a DPX file, it goes in the DI 
software. If the scan was log, then the DI software needs to use a 
display LUT in order to view digitally captured densities as intensities 
(log to lin conversion for viewing). The image can be processed by 
adding/subtracting values to the DPX image before viewing it through the 
LUT. Since the scan was log, adding/subtracting values in the DPX image 
data will be viewed as increasing/decreasing the exposure, once the 
image is viewed through the proper LUT (log to lin) of course.

with the DI process the lab gets calibrated to be as neutral as possible 
in order not to influence and digital image processing. To simulate the 
printerlights in the DI world is a pretty simple process... either you 
can use the formulaes to compute the density change or (which is often 
done) measure the changes in density on a patch (RGB) and give the DI 
software the average density delta values when increasing/decreasing by 
one point or half a point.


Dan Tatut
VP Business Strategy&  Development
Rue des Avouillons 4
1196 Gland - Switzerland
Tel +41 22 364 54 71

More information about the Tig mailing list