Mailing Lists: Apple Mailing Lists
Image of Mac OS face in stamp
RE: Colorsync-users Digest, Vol 2, Issue 3
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Colorsync-users Digest, Vol 2, Issue 3

in Reply to Vol 2, Issue 3:

   3. Colorimeter vs. Spectro

In answer to previous postings, this posting is intended to provide a
clear understanding of the differences between filter based
spectrometers, diffraction grating based spectrometers and colorimeters.

When obtaining color measurements from a monitor, a filter based
spectrometer with 20nm (like Pulse) or even 10nm resolution, can't
capture enough precise spectral details of the light emitted by
monitors. However, diffraction grating based spectros (like Eye-One)
feature a 3.5nm physical sampling interval, providing more than 100
channels compared to 15 channels from a 20nm filter device with no over
sampling. The 100 channels overlap due to 10nm optical bandwidth of the
spectral module. Therefore, grating based devices don't miss any
critical information and are able to resolve the wavelength of a laser
line (the sharpest edge possible) ultimately resulting in a resolution
far more precise than 1 nm! This is why colorimetric results calculated
with Eye-One's spectras measured on a monitor agree very precisely with
those calculated from spectrometers featuring an optical resolution of

A filter based spectrometer will not capture all the critical data and,
therefore, does not support measurements on monitors, which is why a
user of such a device must choose to use a colorimeter or a diffraction
grating based spectro for monitor measurement.

GMB spectral devices are stable over time and feature an excellent
inter-instrument agreement which is independent of the measured
spectrum. This is due to the fact that the filter characteristics
required to obtain the colorimetric results are set in software and do
not depend on optical characteristics of filters (like colorimeters)
which change from lot to lot and also change over time.

The lot-to-lot differences of these filters can be compensated to a
certain extent for a given spectral distribution during calibration at
the manufacturing site. However, this is at the cost of introducing
errors when the spectral distribution of the monitor is different from
the one assumed during calibration.

Advantages of diffraction grating based spectros:

1) 10nm/3.5nm grating based spectrometers like the Eye-One perform
better than colorimeters when monitors with different spectral
characteristics must be matched.

2) The consistency of the measurement results is better when the same
measurement device can be used for reflectance and emission measurement
(like with the Eye-One or Spectrolino) because both colors are measured
with the same resolution of more than 100 channels, plus there are no
inter-instrument agreement errors brought into play.

3) The inter-instrument agreement of 10nm/3.5nm grating based
spectrometers (like the Eye-One) don't depend on lot-to-lot changes of
optical filter characteristics.

Colorimeter clarifications:

Marc Levine  wrote>>Incidentally, it is my impression that Colorimeter
filters are engineered to
match the human eye response and not the spectral character of the
which they are measuring

The opposite is true. Manufacturing colorimeters using filters to match
the precise response of the human eye is not advisable. First, it is
simply too expensive to build filters which feature (in combination with
the applied photodiodes) the complex shape of the colorimetric weighting
functions or a linear combination thereof.

Secondly, due to several technical reasons that would go beyond the
scope of this message, manufacturers choose different filter curves in
their colorimeter design. Nevertheless colorimeters are still able to
provide good results although the sensitivity curves do not match the
human eye response as precisely as a grating based spectrometer.

A grating based spectrometer measures monitors more precisely than a
colorimeter because it features over 100 channels capturing all the
detail information of a monitors spectral characteristic and the filter
curves are implemented by software and not by physical filters like in a
colorimeter. Therefore, if you have an i1 spectrometer you don't need a
colorimeter to measure monitors. However, if you wan't to measure
monitors only, a colorimeter does a good job at an affordable price. If
you plan to buy a more expensive spectrometer for measurement of
reflective targets and you chose a filter based product that can not
measure monitors like pulse, you may reconsider your decision, because
an i1 delivers higher resolution results on paper and monitors.

Conclusion: filters of colorimeters are designed to deliver best
performance on the devices they are designed for - monitors. Filter
based spectros provide good performance for reflective measurement but
are not conducive for emissive measurement. Diffraction grating based
spectros feature precise color matching functions implemented
numerically and independent of the measured device and provide the
superior choice for measuring both emissive and reflective.

Andreas Brant
GretagMacbeth Corporate Support
e-mail: email@hidden

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

Visit the Apple Store online or at retail locations.

Contact Apple | Terms of Use | Privacy Policy

Copyright © 2011 Apple Inc. All rights reserved.