You bring up a very good point.
I have to point out that ANSI lumen specs on paper are just that. Paper. What counts is real world calibrated numbers. For example in direct comparisons with a competitor that has 1500 ANSI lumens on paper, after the units were both configured to best settings with test signals, not only did we have better contrast (which didn't surprise me), but we were brighter as well (which surprised me greatly).
Alot has to do with the effeciency of the light engine itself. Let me explain.
You start with three color filters. Each has a certain perctntage of light blockage. Blue is the darkest, followed by red, then green which is the most transparent.
Therefore, if the algorithms that control the mirror timing are adjusted for the maximum amount of blue output, red and green would still have a way to go. No matter what color temp you are shooting for, blue will be maxed. The key is to have that happen at EXACTLY 100 IRE, otherwise there will be a color shift as you get brighter. This is the problem with many of the projectors. In addition, white crush also occurs, but everyone has that problem; the only way to beat that is to make sure the overall contrast level is set so as to achieve maximum dynamic range. If you look at our PJ, as the color temp gets higher (bluer) the overall brightness decreases. This is because to get the picture bluer and still maintain flat tracking of the gray scale (which is another area where our competitors suffer), you just decrease red and green. On our PJ, the LOW setting is designed for 5400k to match B/W film. It is also the brightest setting with flat tracking. But so far, a couple of the tests of the Sim2 have shown grey scale tracking issues as well as color shifting at the top end, if ANY peak brightness is to be achieved.
PS. If Kevin didn't have the latest version of the PJ, whose fault is that? Typically if a product of ours is being reviewed (especially by a rag that has as high of a viewership as HT), I'm in touch almost constantly with the reviewer, to try to head off potential issues, and to make sure that if something like un "update" occurs (especially if it actually makes a visible difference) that the reviewer knows about it and gets it.