Re: LaCie 526 - problems in color rendering
Re: LaCie 526 - problems in color rendering
- Subject: Re: LaCie 526 - problems in color rendering
- From: Roger Breton <email@hidden>
- Date: Sat, 09 Feb 2008 13:24:38 -0500
Hello Raimar, Rolf, Richard, Steffen, Ken, Graeme und Derek,
If anyone is interested in obtaining a copy of the RIT article everyone
keeps refering to in this thread please contact me offlist. I've managed to
obtain a copy from UGRA -- I've never found it on the RT web site anywhere.
I'd be interested in discussing its methodological merits with members of
the list. If anyone happens to have its exact bibliographical reference
please post it for everyone's benefit. I wasn't able to obtain this
information from my UGRA contact :(
Let me say that there are many aspects of the methodology in this experiment
I am uncomfortable with.
First, an ACD23" was used to setup the test stimulus. I'm not sure that's
the best monitor for testing the effect on the independent variable since we
know of its proverbial non-uniformity? Second, the experimenters used GTI
lightbooth illumination. I'd be curious to see what results would have been
obtained under JUST-Normlicht illumination. Third, the article does not
document how they achieve a "gray background with a uniform noise
chromaticity of D50 chromaticity". An key variable in the experiment. For
all we know, it could have been higher or lower. Fourth, why wasn't the
experiment first conducted in low light, such as in darkness? To establish
some kind of perceptual ground? Or prove that even in total darkness it
wasn't possible for the test observers to scale the controlled stimulus to
match the gray background?
To me, that experiment has a number of methodological weaknesses which makes
me question its validity.
Last but not least, why strictly focus on the scaling of one gray stimulus
at the expense of all colors? Until I'm able to obtain further information
about this article and replicate its findings for myself or with a group of
test subjects I will exercise skepticism with regards to its conclusions.
UGRA would do well to post the full reference of the article on its web site
or at least include a link to the RIT site -- if they are going to continue
refering to the said article.
Best regards / Roger Breton
> The UGRA exactly uses the RIT empiric data for their recommendation. To be
> precise, they recommend 5.600-6.000K for PT 1 - the arithemtic average is
> 5.800K.
>
> Regards
>
> Raimar
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden