On Jun 6, 2004, at 5:14 PM, Michael Harvey wrote:
1. On the Spectrolino / Spectroscan you have the option to set the
instrument to read up to 5 points per colour patch and ask it to average the
readings on the fly.
And as I understand it, this is how the EyeOne spectro works by default. It averages numerous rapid readings as you move the spectro over the patch. Any of you color geeks want to confirm? I can see the value in that.
2. You can read multiple test charts from the same printer and then average
the results in Profile Makers - Profile Editor. Clearly this is not
something that I have found benefit in.
I can see where one could pose that one carefully and accurately read target would produce a better result than several less carefully produced/read targets that are averaged together. But if multiple targets are produced/read just as carefully as the single above; and then the results averaged; how could that possibly have anything but a positive effect on the accuracy of the resulting profile? Since no output system is perfectly repeatable such a method would seem to build a profile that better describes the mid point of the output device's typical performance. Isn't that the goal of an accurate profile? Are you just suggesting that the difference is not enough to be worth the effort? I would think the value of such a method is dependent on how repeatable a particular device is over a range of typical printing conditions (temp, humidity, dry down time and so forth).
On Jun 7, 2004, at 6:22 AM, matthew ward wrote:
If you measure it lots of times you can start making statements about how accurate your measurement is.
You've stated quite clearly and succinctly my understanding of averaging as well.
Thanks
Bob Smith
=============================================================== GO TO http://www.prodig.org for ~ GUIDELINES ~ un/SUBSCRIBING ~ ITEMS for SALE
