At 8:50 AM +0100 5/12/04, Shangara Singh wrote:
>It was 12/5/04 1:16 am, when Steve Upton wrote:
>
>> A monitor profile will describe which COLOURS are represented by those
>> numbers. Do not get fooled into thinking that RGB values = colours. RGB values
>> need to be sent to a device and observed by a human before they become
>> colours.
>>
>> for a short, light-hearted description, read my "Color of Toast" essay here:
>> <http://www.chromix.com/ColorSmarts/smartNote.cxsa?snid=1005&pid=1.pdg>
>
>Steve
>
>A brilliant article if a bit too scientific (I had to read it several times
>but I now, at least, understand color theory and the need for color
>management).

Thanks, yeah colour management can certainly be "heady" at times. Toast was about the 
simplest analogy I could come up with.

>However, I would like to take issue with the following: To have color you
>need 1) light, 2) an object and 3) an observer - for our purposes, a human
>observer. Without all these components you do not have color.
>
>This is an old worn, torn argument often used by color management gurus to
>justify their huge fees and it just does not hold water. It's like saying,
>when I die, existence comes to an end. Of course it does - for you! For the
>rest, it carries on, erm, existing. So, colors exist, even without an
>observer.

A couple of comments:

- I didn't want to start a philosophical discussion in the same vein as "if a tree 
falls in the forest". But being a scientist at heart (which is not necessarily a 
departure from philosophy) I tend to take the existential view. So, I view it as 
someone else posted. Light is full of many wavelengths, objects selectively absorb and 
reflect wavelengths, people selectively sense wavelengths and colour is how it is 
assembled in the brain.

- the science behind this backs it up. When you take a reading of a colour patch on 
paper with a spectrophotometer (like the Eye-One Photo for instance), the instrument 
determines the spectral reflectance of the patch. In software (in-instrument or on 
your computer), the patch's spectral curve is combined with an illuminant curve 
(typically D50) and human eye response curves. The result is Lab. A number assigned to 
a human colour response.

So even with philosophical arguments aside, we need an object, an illuminant and an 
observer.

>
>BTW, where can I buy your Toast Guide�? I just cannot get my wife to
>reproduce the toast the way I like it, and consistently.

that may be a calibration problem <g> (either on the part of your toaster or your wife)

Regards,

Steve


________________________________________________________________________
o  Steve Upton              CHROMiX        www.chromix.com
o   (hueman)                               866.CHROMiX
o    [EMAIL PROTECTED]                     206.985.6837
o  ColorGear   ColorThink   ColorValet   ColorSmarts   ProfileCentral
________________________________________________________________________
-- 


===============================================================
GO TO http://www.prodig.org for ~ GUIDELINES ~ un/SUBSCRIBING ~ ITEMS for SALE

Reply via email to