Originally Posted by Fat Dave
By what measurement methods have you established that there is a change in gamma on this projector with dynamic iris engaged?
By studying how these devices work:
It isn't just a question of a little flap that opens or closes according to picture level. There is other processing going on with the signal as well. Otherwise when the scene is dark anything that is brighter than black will look dim, so the processing attempts to maintain the same brightness for those elements of the picture.
Imagine this process in action:
You have a dark black sky, perhaps a camp fire scene, so the ground area may be very dark, so almost black, but you have the flames of the fire and the reflection of the fire on the faces of the people around it. If the iris just closes down because it's a dark scene, then the flames and the faces will get darker too, making them look dull. The near black ground will blend to black too. The electronics increases the contrast (within limits of course) to try to make the flames/faces look as bright as before the iris closes down. Try measuring the gamma of a display (with the iris fixed) and at different contrast settings and guess what? The gamma will measure differently.
Now this won't work for a star scene because the stars might be at peak white and the sky at black, so closing the iris will just lower the black, and also lower the brightness of the stars (the contrast can't go any higher as they are already at peak white). A scene like this is already using the maximum native contrast of the display, so no amount of iris and contrast changing can increase the contrast of this scene. The better iris systems will be able to detect this and won't try to 'fix' what it can't so you won't get pumping of brightness for no gain in contrast.
FWIW I pulled apart my first projector a Panasonic AE1000 and made lots of measurements of the effect the iris had (I only had an i1LT back then but it was still pretty good for measuring gamma if not so accurate for greyscale or colour gamut). I was troubled by the poor dark scene performance and wondered if there was any way to improve it by tweaking and better understanding how it worked. I made many measurements with the dynamic iris enganged, but physically unplugged to see how the image was manipulated. Eventually I moved away from a sequence of dynamic iris AExxxx models as I realised that my preference is for better native contrast that DILA/LCOS can provide, but it was an interesting voyage of discovery.
While the above shouldn't
cause a colour shift, in a dark scene the contrast might change such that a 60% part of the image gets boosted to 80% (and the iris closes down to bring the brightness optically back to 60%). It's not unusual for there to be a difference in greyscale between 60% and 80% points especially if the display only has a two point greyscale correction.
I should add that I haven't done these measurements on the 50ES, but the principle is likely to be similar (it's more the algorithms used to decide the iris and contrast boost that differentiates between different projectors as well as the native contrast available). For scenes that don't contain peak whites, then the iris can close down some, the contrast can go up to make the non black areas brighter (up to just under the point of clipping) meaning that the black level goes down, but hopefully the other parts of the image are maintained at the same brightness they should be.