Color Question

[Andrews, A.S.]

There is an item on the E-Prime web site regarding the different
implementations of 16-bit colour, but I doubt this is your problem.
Perhaps you have a physical problem with your display hardware. You
should also experiment with setting your display to 32 bit colour to get
the best representation of the RGB value.

It is worth noting that when assessing the appearance of a colour, your
perception will be heavily influenced by the surrounding colours. If you
implement the grid proposed below you will see something different if
you allow some black space between each swatch (an 'adaptation field').
This famous example makes the point rather well:
http://www.grand-illusions.com/chequerboard.htm

Colour calibration is a huge field, which I was involved in some time
ago. Depending on your requirements it can be a simple as using an
application with a swatch card (supplied with many monitors and graphics
cards) through to using instruments that cost thousands and are
traceable to National Standards. Here's a couple of links that were
passed to me - I have no direct affiliation or necessarily any
experience of their particular products: http://www.eldim.fr/,
http://www.photoresearch.com/,
http://www.minolta.co.uk/frameset.php?page=http://www.konicaminoltaeurop
e.com/products/industrial_products/overview/color_measurement/

Best Regards,

Tony Andrews
Principal Computer Officer
School of Psychology
University of Leicester

-----Original Message-----
From: eprime at mail.talkbank.org [mailto:eprime at mail.talkbank.org] On
Behalf Of James Intriligator
Sent: 14 September 2005 18:38
To: Richard P. Heitz; Katz, Lena B.
Cc: E-Prime Listserv
Subject: Re: [Eprime] Color Question

Lena -->  one thing worth keeping in mind is that different
monitors/computers will display (255,255,0) very differently.
  An easy way to see this is to try displaying the same thing on:
  - a CRT monitor
  - a CRT monitor that has the brightness/contrast turned way down
  - an LCD monitor seen "straight on"
  - an LCD monitor seen "from the side"
You may find that each one will appear very different.
For many experiments this will not matter too much.

BUT, if you are worried about it, you have 3 options:
  (1) trial and error:  just try different values and find the one that
looks like what you want.
  (2) What I have done in the past is to have the program draw/label
lots of different variants at the same time.  For example:  have it draw
a 10 by 10 grid of coloured squares where changes on the x-axis are
changes in red (235, 237, 239,...255) and changes on the y-axis are
changes in green (235,
237, 239,...255).   So the bottom-left square would be:  (235,235,0) and
the
top-right would be (255,255,0). You should be able to find one of these
100 choices that looks like what you want.
  (3) you could get very nerdy and calibrate your monitor.  This may
require you to make measurements, estimate gammas, create lookup tables,
and other complex things.

  **** I wonder if anyone in e-prime world has addressed colour
calibration stuff?  Anyone willing to share?

  -james


On 14/9/05 06:18, "Richard P. Heitz" <gte016z at mail.gatech.edu> wrote:

> Yes, that's an RGB value.  And, I just checked to be sure - 255, 255, 
> 0 appears to be a pretty bright yellow.
> 
> Katz, Lena B. wrote:
>> I'm playing around with Ccolor, and when I put in (255,255,0), i 
>> don't see a true yellow.  Does anyone know if this is in RGB?  Or am 
>> I perhaps using the wrong scale entirely?
>>  
>> Thanks!
>>  
>> Lena

------------------------------------------------
James Intriligator, PhD
Senior Lecturer

School of Psychology - Experimental Consumer Psychology University of
Wales - Bangor Brigantia Building - Room 328 Bangor, Gwynedd LL57 2AS
Wales, UK

Tel: +44 (0)1248 383630
Fax: +44 (0)1248 382599
email: j.intriligator at bangor.ac.uk
------------------------------------------------