just a quick question: did anybody consider using the qumi with an anamorphic prism/lens system? Just because, as far as I can see it, this little projector has rather convenient amount of pixels (1.15M) to work with, which is 75% of the pixels actually used on BD discs when storing 2.4:1 aspect ratio movies. But if we use just the qumi to project these movies, then 50% additional pixels will remain black above- and below the image, in the black bars.
This is already a problem for a 16:9 AR projector (only 33% additional pixels "sleeping" there) but the qumi is 16:10 AR. This means losing that 50% additional light and resolution. As the qumi is not a projector with plenty of light (and pixels) to work with, utilizing these resources would mean a lot in my opinion. But the above also means that we need anamorphic optics with 1.5x extraction/compression ratio and these are not the standard products today. So, we have two options:
1. use "tunable" prism systems and set them to 1.5x extraction/contraction, or
2. custom build anamorphic prism systems for the qumi
Now I have some prisms at home and made some tests. The first experience i had was that the qumi works nicely with the achromatic prism system that I have build for my Samsung a600 projectors. But life is not always as nice as we hope: as the qumi has an 1.55:1 throw ratio to start, widening the image with another 50% results in about 1.03:1 throw ratio, and this has really unhappy results. The nonlinearities start to dominate at the edges of the image, the image is significantly distorted there (the extraction ratio is much more than 1.5 at the edges) these areas of the image also get darker and the astigmatism makes the image unfocused in these areas. So, I guess we can consider this direction unrealistic. Only projectors with longer throw seem to be suitable for building horizontal expansion systems.
The other possible way is to shrink the image vertically, this would leave the 1.55:1 throw ratio intact and the vertical compression would have to handle much (1.6x) smaller angles, so, I tried using simple prisms of optical glass ... and it worked nicely. There was some vertical false color (first order lateral color) introduced at the top- and bottom parts of the image (these were singlet prisms of BK7 optical glass) but it was really enjoyable, and naturally, the increased brightness of the image was obvious. If there are people who use similar singlet prism systems for horizontal expansion (it is a fact that these are still offered by some manufacturers) then I guess this MUST be a realistic compromise. Soon I will make a temporary housing for the prisms and try this in a home theater environment to see how acceptable the false color will be. Also, I want to see the surface brightness of the image in a darkened room on my 210cm (about 83") wide 1 gain screen. I really hope the qumi will be suitable for viewing movies with the kids in 2D, so, we can use the larger projectors exclusively for (polarized) 3D (where the qumi would run out of light). Taking into account the power consumption (and bulb costs) of my UHP lamp projectors, if the qumi works well in 2D than it was already worth to buy it.