[Tig] Dolby PRM-4200

Richard Kirk richard at filmlight.ltd.uk
Tue Aug 9 11:58:22 BST 2011

Carl Skaff wrote:
> So... what if you are starting a new company with brand new equipment 
> and DON'T have an old CRT that you want to use as a reference when 
> calibrating the new and fancy Dolby/Sony/Eizo/Barco/Cinetal/DreamColor 
> etc etc....
The 'right' thing might be to have a new standard. DCI was an attempt at 
this, but it has a gamma curve that goes down to zero, and real 
projectors give some gray level for zero RGB. The DCI standard also 
means that a lot of the DCI X'Y'Z' space is filled with impossible 
colours, and the real colours are confined to a small lozenge of space 
up the diagonal. There is an informal standard that uses the DCI gamma 
with the P3 primaries, which is a bit more video-like, and makes 
interpolation easier. People have used both of these spaces with 
success, but I feel we could do better. However, there are a lot of new 
display technologies, and things are changing rapidly, so this isn't a 
great time for writing new standards.
If you want to approximate to the old standard, and you have Truelight 
software, and a projector probe, then you can summon the ghost of a Sony 
CRT BVM. Fist calibrate your display. Set displayLab{on} in the profile. 
You can then switch between the new calibration and the internal 
'SonyHD' calibration (now also called 'Video HD' - it's the same data).

This calibration uses the standard phosphor matrix, and a tone curve 
that was taken to be the average of all the larger Sony BVM monitors we 
knew that were set up using a PLUGE test chart and a Philips probe. The 
average was skewed towards two sites that had more than one of these 
monitors in separate rooms, so we knew the set-up person could achieve 
the same result more than once. There should be no systematic difference 
between the RGB tone curves, so each monitor gave us three traces to 
average. For very low light levels, we chose a simple function that was 
smooth, but not as steep as the very new monitors which very often had 
quite different tone curve shapes, and did not flatten out like the 
older monitors. It has a contrast ratio of about 4000:1. If you stuck to 
the legal monitor range, then the RGB primaries were independent: if you 
went all the way to the maximum white then the beam currents were 
limited, so whites were dimmer then the addition of RGB would predict.

If you do not see a significant difference when you switch between the 
standard 'SonyHD'  calibration and your display, then your display is 
probably close to an average well-maintained Sony BVM in about 2004, 
provided the white point is D65 and the luminance somewhere between 
about 35 and 45 ft-L for video (this varied between sites), or 16 ft-L 
for film simulation.

If you haven't got Truelight, we could convert the SonyHD calibration to 
an ICC profile. However, once we get into ICC-land it is much harder to 
know exactly how the profile is being used.

Disclaimer: I work on Truelight.

Richard Kirk

FilmLight Ltd.  	Tel: +44 (0)20 7292 0400 or 0409 224 (direct)
Artists House,		Fax: +44 (0)20 7292 0401
14-15 Manette Street    
London W1D 4AP

More information about the Tig mailing list