Quote:
Originally Posted by
Fat Dave 
Chuck Antsey's post above on the iris
I disagree with his findings. You can check this in two different ways:
1) You can set the first variable in the service menu to "1", which tells the projector to set the iris to corresponding entered value for Iris Open Reg or Iris Close Reg. That allows you to see the changes in real time to setting different values. It changes all the way down. This doesn't accurately replicate what the projector will do in a real world situation, because it will open up to the actual value of the variable you're changing, not the calculated intermediate value that the scene would call for, but it does show the finite adjustments of the iris.
2) You can measure the difference with a meter on a paused scene. It will show a difference as you make these changes, similar to what I showed in my HW30 testing.
I reevaluated whether the dynamic iris uses in-between positions and I did find that it does. The preference for the algorithm is to be fully open or fully closed but it is willing to play around in between as necessary. However, whether the iris is fully open or some fraction of closed isn't based upon the average brightness of the scene but whether high IRE pixels are present, in any quantity. I had up the test pattern from the AVSHD disc where 0 to 24 are shown flashing. This image is all 0 when flashed off, which I paused, yet the iris remained fully open because the the numbers were fully white even though the occupied maybe 2% of the image. The DI appears to prefer keeping whites brighter in any quantity than blacks darker. My subjective opinion of this is I think this is preferable because when I forced the projector to use the closed iris position, the turning of the numbers darker (appear very gray and not subjectively white) was far more noticeable and less desirable than the advantage of darkening the blacks.
However I still stand by my initial assessment that the majority of the Open/Closed Reg value is about gamma remapping but obviously willing to reevaluate with additional data. I looked at the iris directly and went from my factory setting of 313 to 0. The iris does close a little more but not much. My problem with your conclusion that it is 100% iris changing is the method used to evaluate, a paused scene, could be changed by either changing the iris position OR by gamma remapping OR both. We know that when DI in active there is gamma remapping going on as the iris changes position.
There is an easy way to test this and I would do it myself except I don't have a meter that can do it.
Go into the Iris control menu to set the Closed Reg (currently at factory) and make sure Iris is in the closed position
Project 0 IRE, 20 IRE, 40 IRE, and 60 IRE patterns. Measure the light of each.
or
Project a multi-IRE pattern that you can measure the different IRE around the screen.
Now set Closed Reg to 0
Project 0 IRE, 20 IRE, 40 IRE, and 60 IRE patterns. Measure the light of each again.
Now if the Closed Reg only controls the iris opening and nothing else, all of those IRE patterns should increase in lux by the same multiplicative amount when comparing Closed Reg = 0 and factory setting (~300). I expect however that the 20, 40, and 60 IRE patterns will increase their lux by a higher percentage than the 0 IRE pattern, which is only affected by the iris opening at least if the gamma doesn't allow showing blacker than black. If the percentages are not the same then Closed Reg definitely controls gamma remapping in addition to possibly other things.
