Re: ICColor vs DTP41
Re: ICColor vs DTP41
- Subject: Re: ICColor vs DTP41
- From: email@hidden
- Date: Tue, 18 May 2004 06:35:31 -0700 (PDT)
Roger,
Here a scale based on some calculations I did on tens of thousands (not one by
one if you ask ;-) )of randomly selected colors uniformly distributed within the
AdobeRGB or Apple RGB spaces.
On average, for small color differences (less than one), dE*ab is about twice as
large as CIE94. So, a 0.1 dE*94 is about equal to 0.2 dE*ab, which easily
explains why a manufacturer would use the former standard as a specification
(plus the fact that it is generally accepted that, indeed, it represents the
perceived difference more accurately...)
Note: the above result if for the standard CIE94 variety, not the Textile
version, which gives smaller numbers and an even bigger ratio.
Hope this helps.
Danny
On Mon, 17 May 2004 23:20:18 -0400, Roger Breton wrote:
>
>
I9ve read somewhere (Greg Barnett) that the ICColor, apparently, has some
>
inherent bias in reading colors? In trying to substantiate that basis, I9ve
>
been reading on the ICColor and the Spectrolino specs. For one thing, I
>
found it odd that the ICColor accuracy is expressed in terms of dE*94, which
>
is a weighted color difference metric, whereas the Spectrolino accuracy is
>
expressed in terms of dE*ab, making it impossible (to my knowledge) to
>
directly compare the two units: can 0.3 dE*94 (ICColor) be likened to 0.3
>
dE*ab (Lino)? From my limited color knowledge, this suggests that the Lino
>
is more 'accurate' than the ICColor.
>
>
At any rate, I find the repeatability of the ICColor pretty low in
>
comparison to the Spectrolino less than 0.2 dE*94 vs 0.03 eE*ab for the
>
Spectrolino? FWIW, the i1 has a better repeatability than the ICColor :
>
0.1E* 94 vs 0.2 dE*94. I know these differences are tiny but for the amount
>
of money we have to pay for these instruments, I wonder if the ICColor is
>
all that worthed?
>
>
Inter-instrument agreements
>
>
Average 0.3 E* 94 (ICColor) vs 0.4 E* 94 (EyeOne) vs 0.3 dE*ab (Lino). The
>
ICColor seems to be 'better' than the EyeOone but it's not possible to
>
compare it with the Lino.
>
>
DTP41 specs are quoted in terms of dEcmc with a value 0.3. Inter-instrument
>
agreement is quoted at 0.2 dE (does not say dE*ab or dEcmc or what, just
>
dE?). So there we have it, no standard of comparison across manufacturers or
>
even within the same manufacturer.
>
>
Bottom line, is the ICColor any good?
>
>
Roger Breton | Laval, Canada | email@hidden
>
http://pages.infinit.net/graxx
>
_______________________________________________
>
colorsync-users mailing list | email@hidden
>
Help/Unsubscribe/Archives:
>
http://www.lists.apple.com/mailman/listinfo/colorsync-users
>
Do not post admin requests to the list. They will be ignored.
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.