Re: Monaco Profiler and DTP41 measurement issues
Re: Monaco Profiler and DTP41 measurement issues
- Subject: Re: Monaco Profiler and DTP41 measurement issues
- From: Graeme Gill <email@hidden>
- Date: Tue, 11 Mar 2003 11:23:29 +1100
Dikta wrote:
>
I have been using Monaco Profiler and the Xrite DTP-41(UV/T) with great
>
success, but recently I am getting more and more problems doing the
>
measurements. The DTP-41 is acting as the pathes are skewed while
>
reading, but they are going very straight into it. I am mainly
>
measuring RA-paper or films (duraclear or trans).
>
>
In some cases I have rubbed the paper lightly with a cotton cloth, to
>
see if it could be fingerprints, and sometimes this helpes, but last
>
time I inspected the patch carefully and could not see any marks, but
>
still the program or the DTP41 refused to read the strip, I gave up
>
after trying about 30 times...
This isn't a problem unique to Monaco. In my experience the DTP41 is
rather more fussy over the spacers between the patches than the
manufacturer would like to believe. Putting down black or
white patch spacers is not sufficient in itself, unless the
layout of the patches is very carefully controlled so as to
avoid certain colors being adjacent. Randomizing the patch
layout may raise your chances of running into this sort of problem.
Another "gotcha" is cutting the strip too short at the front.
I'm told that the DTP41 is looking for a certain density
change in one of the R/G/B density channels. To achieve this,
colored patches are needed between different patch values,
and even then it's not possible to achieve acceptable
density differences between any arbitrary combination of
patch values.
I've managed to get very reliable results for a DTP41 by:
Using an approximate (profile driven) density model
for estimating patch spacer contrast.
Generating colored spacers to maximize density
change between patches and spacers.
Stochastically optimizing patch placement to make the
minimum patch/spacer contrast as large as possible.
As far as I know, no commercial package goes to this
trouble. For sufficiently small test charts, it is even
possible to do without patch spacers if you're prepared
to arrange the patches carefully enough.
Graeme Gill.
_______________________________________________
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.