Digital BW, The Print

Paul: <SNIP ... While we're on the subject, note the extent to which 
"reciprocity failure" of the accelerated tests gets more serious as the 
illumination level
increases.  ... In short, if one tries to project "years of display" without 
considering reciprocity failure, the exaggeration of the predicted life could 
be huge.>

Yes, indeed. Reciprocity failure is the Achilles Heel of accelerated testing, 
IMO, significantly *underestimating* fading. And what's interesting is that 
Henry Wilhelm has known this all along, at least since the mid-'70s. In fact, 
he devotes a whole table to it in his 1993 book. Yet, the latest published 
testing results from Wilhelm all assume a reciprocity failure of ZERO. Not 
quite sure why. Supposedly, his new test methodology takes this into account 
somehow. We'll see.

Harald Johnson
author, "Mastering Digital Printing"
DP&I.com ( http://www.dpandi.com )