Datacolor User to User Support Group.

******* Measuring Targets and Checking Measurements ********

- Make sure you attach the spectro to a powered USB port. The safest  
USB ports (you'll always
know they have power) are those built directly into the computer.  
Sometimes there's enough power
in the USB port on a keyboard for it to work; other times, not. If you  
plug the spectro into a
keyboard USB port, it may not "show up" on your system, or even if it  
does, there may not be
enough power for it to work correctly. If you plug it into an  
unpowered hub: it won't work at
all. If you plug it into a powered hub which doesn't supply enough  
power, or which has other
devices connected that draw a lot of power, it may not work at all, or  
it may not work
consistently. If a powered hub isn't getting any power (it's DC power  
cable is loose, or it's
power adapter isn't plugged in, etc) then you make think there isn't a  
problem (some or all
other USB devices on the hub may work fine, not all USB devices  
require much power) but the
spectro won't work because it definitely needs a certain amount of  
power to work.

- The basic technique for building profiles is as follows: Run  
Spyder3Print/PrintFIX PRO; step
forward one screen, then click on Skip To Profiling Process (bypassing  
the Print Quality Check
and Media Setting Check screens, which are not required). Select the  
225 patch target (recommended
for most profiling). Click the Print button, to print it, with all  
color management disabled in
the driver (In OSX, use the Print Settings pane to choose your paper  
type, output quality/speed
settings; use the Color Management pane to turn color management OFF;  
in Windows, select the
printer, go into the printer preferences/properties, and use the  
printer driver's UI to do the
same thing). This will give you a target print for measuring.

- The target prints should always be dark and saturated. Darker is  
better.

- The measured colors will always be darker than the "pure" colors in  
the Target window
display. Don't let this alarm you, it's expected. The profile that you  
build will compensate
for this.

- Don't perform the white calibration of the spectro on your paper -  
it will throw the
measurements off in subtle ways. -Always- perform the calibration of  
the spectro (prior to
taking a series of measurements) in the base; on the round white  
calibration tile.

- Put one or two blank white sheets of the paper you're profiling  
beneath your target, when
measuring.

- When measuring, it's easiest to switch from Split display mode to  
Measured, so you can see full
color patches as you measure them. Easier to see errors this way. The  
measurements, on your
calibrated screen, should "look" like the target patches that you're  
measuring.

- Don't measure "blindly"; pay close attention to what you're  
measuring. If your
measurement for the black patch (always the first patch in every  
target) looks lighter than
the other patches near it, you'll know instantly that something is  
wrong. If your measurement
for the white patch ("paper white", which has no ink) comes out cyan,  
you'll know something is
wrong...! Paper white won't show up as a pure, perfect white when you  
measure it... expect an
L value of the low to upper 90's, and a and b values close to, but not  
exactly equal, to 0. It
should display as a light gray, or something with a very SLIGHT color  
cast that's very close
to neutral. If it looks like a "color", it's wrong. If you see  
scattered patches in the darker
measurements that are noticeably lighter or less saturated than their  
neighbors: something
is wrong. If you see patches that have noticeably changed color with  
respect to their neighbors:
something is wrong. You can observe these "by eye" at the end of each  
row as you're measuring
(see other hints below) and also after you've finished the entire  
target, by toggling the display
between Pure and Measured and observing how the patches for the entire  
target change as a
group. Your eye is a very good judge of color and this is the best way  
to check for problems.
The overall effect you should see, when toggling the display of the  
complete measurement set
between Pure and Measured modes, is a darkening of the colors.  
Neighboring patches should visually
maintain their relationship, there shouldn't be any "wild" flips of  
colors, or any "wild"
changes in lightness or saturation, when you look at groups of color  
patches or the target
measurements as a whole.

- Measure across a row; after the bell, look up and visually check  
your measurements. Look for
totally incorrect patches; or any that stand out incorrectly from  
their neighbors, particularly
in the shadows. (Lighter/less saturated than they should be). If you  
see any like this, use the
left arrow, go back and remeasure.

- When you're done with the entire target, do one more visual check  
this way, before you close
it.


David Miller
Senior Software Developer, Digital Color Solutions
Colorvision