[Tig] Seeing Red (was Re: Why Whites May Not Match)

Richard Kirk richard at filmlight.ltd.uk
Fri Feb 3 09:40:28 GMT 2012


Owen Williams <owen at ywwg.com> said...
> I'm beginning to think that I don't see colors the way other people do,
> in a way that's somewhat significant.  Specifically, there are times
> when I see red in an image that other people think looks neutral.  This
> has been a problem most of all in the Dreamcolor monitor series.  I see
> the calibrated white point as extremely red even when we've calibrated
> it (and especially compared to the plasma).  I had a couple other people
> in the office compare the two and what they think is a match looks
> definitely wrong to me (dreamcolor red, plasma neutral).  And what I see
> as a match they see as incorrect (dreamcolor blue/green, plasma
> neutral).
>
> So my question is, how can I figure out if what I'm seeing is based in
> some sort of physical reality, or if I'm just a disabled colorist?
Hi, Owen.

The various eye tests on the web will only tell you if you have 
defective vision (missing primaries) or anomalous vision (weak primaries 
so you are effectively missing one primary over some of the color 
space). I am guessing that you do not have defective vision as such or 
you would not have chosen work with color images, but we have to check 
the obvious first. I suppose it is just possible you have been 
protanomalous all your life and never noticed, but I expect you will 
pass these tests.

So, you suspect your eyes are seeing three primaries, but slightly 
different ones to some of the other people you know. There are some 
tests for this. There is a thing called an anomaloscope. The Nagel 
anomaloscope test for red-green vision: the subject mixes various 
amounts of red and green to match a target yellow. Regrettably, people 
usually are anomaloscopes to test whether the subject can find a 
consistent match (which shows you that you can distinguish red from 
green in yellow mixtures) and any match within a broad band is accepted 
as a pass. What we need is not measure lots of people with some standard 
anomaloscope and see what the scatter of matches is.

You can probably get some useful measurements if you get yourself and 
the people who disagreed with you measured with the same anomaloscope. 
Just getting yourself measured may not tell you much. Explain to the 
optometrist what you are trying to do, and they may give you the raw 
anomaloscope data.

If you find a difference, it does not necessarily matter. Everyone's 
eyes adapt. If you are looking at one display in a darkened room, you 
will get the same stimulus as everyone else. I suspect people have 
always disagreed on matches like this. This is particularly true these 
days because there are several display technologies with different 
primaries, where 10 years ago any decent image was always viewed on a CRT.

A lot of people have studied the shift of color vision as a sign of 
health. heart disease, diabetes, alchohol use, Viagra use, and other 
things all affect visualcolor balance. There is also a lot of random 
variation in color balance between people, perhaps depending on...
  genetics (R,G primaries only)
  macular dye distribution
  distribution of R and G cones (the ratio can vary from 4:1 to 4:1 in 
people with normal vision)
  yellowing of the lens (probably slight)
  long term adaption (maybe)

What is the solution? Stick the displays in different rooms. You 
shouldn't have two displays showing the same image. No, this isn't a 
monumental cop-out: it's the right thing to do, with the sole exception 
of comparing different technologies - which is not part of a production 
process. If you are comparing different technologies, then stick them in 
the same room, but everyone is allowed to tweak the white point differently.

Hope this makes sense.

Regards
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