Datacolor User to User Support Group.

On Nov 8, 2008, at 11:15 AM, John Arnold wrote:

> David,
>
> I have a few more questions that have sprung up while contemplating  
> the Spyder 3 Print,
> which by the way is starting to shape up as the winner in my mind.
>
> 1} You made a comment in a dpreview forum recently. You said  
> "Datacolor spectro reads
> much smaller patches, so less ink and paper is used to provide a  
> much greater number of
> patch measurements. It also measures quickly: overall profiling  
> time, even with spot
> measurements, is comparable (or even faster) than other solutions  
> with swiped
> measurements." Can you please elaborate on the "or even faster"  
> portion of your
> comment. The spot measurement feature and time involved measuring  
> has been one of
> the concerns that I have had.
>
>
The 3.5 software configures the spectro so that it physically
takes measurements more quickly than any of the earlier versions.
There's a "measurement time" associated with every patch; it's the
time that the spectro itself takes for the measurement, and when
you use the software, you'll "hear" this as a small gap between the
two click sounds that you get as feedback when you take a measurement.

In the 3.5 software, the time between the clicks is smaller, because  
the measurement
itself is several times faster.

In practice: this means that the spectro not only "feels" more  
responsive as you
move from patch to patch and measure (either by pressing on the nose,  
or pressing the
Return key)... because you're spending less time audibly waiting for  
the 2nd "click"
so that you can move to the next patch... but that overall, your  
measuring time will
go down overall, because you can move more quickly (still taking care  
not to go TOO
fast).

For example: with 3.0 and earlier software, it would generally take me  
close to 5
minutes to measure the 225 patch target (which admittedly is faster  
than most people
will go, but I measure many of these). The 729 patch target would take  
closer to 15
minutes.

With the 3.5 software, I can measure the 225 patch target in a little  
more than 3
minutes, and the 729 patch target in under 10. To me, that's a big  
difference, and
the speed for the 729 patch target means I'm more likely to simply  
print and measure
that, without worrying about whether or not the extra patches will  
benefit the
profile.

(In fact, the novelty of the faster measurements still hasn't worn off  
for me,
so I actually enjoy measuring the 729 patch target now... the time it  
takes
me to measure 225 patches is so small that it's almost an afterthought).
> 2) One of the participants in the same dpreview forum said that the  
> Spyder is supported in
> most third party applications. Is that true? I wasn't quite sure  
> what the person was
> referring to.
>
>
The Spyder3 (for display calibration) is supported by ColorEyes  
Display from
Integrated Color; this is the best known 3rd party display profiling  
application
and I'm sure that's what they were referring to. If you have an Eizo  
display, it's
also now directly supported by their ColorNavigator 5.1 software.

So it means that if you buy any product of ours that contains a  
Spyder3, there are
certain other software packages that can use it, beyond ours.

For printer profiling, Spyder3Print can export Lab values from patch  
measurements into
a tab delimited text format, so technically it's possible to use it to  
measure color
targets for use with someone else's printer profiling software,  
although I haven't
spoken with anyone who's tried using it that way. A more common "3rd  
party" use is
with QTR (a 3rd party utility for profiling and printing with  
specialty B&W inksets);
Spyder3Print has a "Measure" tool that lets you measure any color, or  
sequence of
colors, that you can touch the nose of the spectro to, so you can, for  
instance, print
QTR's step wedges and export the results to text files that contain  
either Lab or
density measurement values. (In fact, this feature also means that you  
can use the
spectro and Spyder3Print as a densitometer).
> 3) I noticed in the Northlight-images review that "saturation" is  
> the default rendering
> intent in the software. Why is that when in most published material  
> these days I seem to
> see "relative" as the suggested rendering intent for photography? I  
> presume that's not a
> big deal either way since you can change it anyway?
>
>
The way that we build the profile, as far as the Saturation rendering  
intent goes,
is different than most other 3rd party and OEM profiles. They put data  
in their Saturation
intent that's intended to boost colors for business graphics and line  
art, but which isn't
generally well suited for printing photographic images.

In our profiles, we use a different philosophy for the Saturation  
intent. For photographic
images (where the colors are all, or mostly, inside the printer  
gamut), the data in our
Saturation intent makes beautiful prints, so it's perfectly well  
suited for general
photographic printing. For out-of-gamut colors, the Saturation intent  
pushes the print
towards the most saturated colors that the printer driver/paper  
setting/inkset can
produce. This means that you get the best possible saturation in your  
print for out-of-gamut
colors, with little or no mixing of other color inks (yellow, for  
instance, will print with
pure yellow ink, and blue will print with the maximum blue ink mixture  
that the driver
produces based on your paper type setting). For most people, this  
yields the best results,
and that's why we recommend it as the "default" rendering intent with  
our profiles.

If you like, you can choose either the Perceptual or Relative  
Colorimetric rendering intents
as well, which provide some color mixing for the out of gamut colors  
to make the printed
hues more exact, at some expense of saturation.
> 4) I have read in Bruce Fraser's "Real World Color Management" that  
> they generally
> recommend reflective spectrophotometers over reflective  
> colorimeters, generally finding
> the former more accurate. Can you comment on that and try to put it  
> in perspective as it
> relates to the Spyder 3 Print as I know Spyder 3 uses a colorimeter?
>
>
We've got published stats on the accuracy of the Datacolor 1005  
spectrocolorimeter with
respect to other devices:

http://spyder.datacolor.com/product-compare1005.php

The numbers speak for themselves. Also, if you read what Uli Zappe has  
had
to say here:

http://lists.apple.com/archives/colorsync-users/2008/Jul/msg00001.html

And also, the results of the big "Printer Profiling Shootout" here:

http://www.luminous-landscape.com/essays/fancy-graphics2.shtml

when you look at the scores and chart in the conclusion at the bottom,  
you'll see
that we scored a 45, compared to the JK3 score of 41. JK3 is the  
EyeOne Pro
with their reflective spectophotometer, vs. our reflective  
spectrocolorimeter.

These two independent 3rd party reviews reinforce the "goodness" of  
the measurements
coming from the Datacolor spectro. If we're doing as well (Uli Zappe)  
or even BETTER
(Gerard Kingma) than the much more expensive XRite/Gretag systems that  
are using
spectrophotometers, we can't be more satisfied than that.
>

David Miller
Senior Software Developer, Digital Color Solutions
Datacolor