User-agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.4) Gecko/20030624 Netscape/7.1
See comments below:
Marc Levine wrote:
In the case where a spike falls between
the narrow bands of a spectro - the instrument will fail to capture accurate
visual data which may potentially compromise the accuracy of any subsequent
profile.
I believe that most spectrophotometers have spectral channels that
slightly overlap. Therefore, they would not miss narrow spikes. The
spectral resolution in most spectrophotometers is divided into 24 or 32
channels. This give you much better spectral resolution than any wide
band filtered instrument.
In addition, densitometers and colorimeters have filters whose spectral
characteristics change with time. The sensors themselves change
spectral and gain characteristics with temperature. None of the current
calibration procedures supplied with these types of instruments can
fully correct for these types of changes. If you get into reflective
type instruments then the spectrum of the illuminate can also change
with time. It is possible to calibrate a spectrophotometer with a
standard white reflector and compensate for all of these problems. This
procedure allows you to normalize the readings of all of the spectral
channels to compensate for changes in either the sensor or illuminate.
You can also emulate any of the densitometer or colorimeter filter types
with weighting factors for each of the spectral channels. Since these
weighting factors are stored in the firmware of the instruments they do
not change or fade.
When I worked full time in this area and exchanged measurements with
labs at Kodak, Dupont, Imation, and FUJI we used nothing but
spectrophotometers and used standard tiles to calibrate and exchange
data. We never used densitometers or colorimeters because of the short
comings I described.
I am speaking of instruments like the Gretag Spectrolino and the X-Rite 938.
