Curved DLP Screen.  

Do a rough calculation here.... The difference comes out to about 0.625 inches of depth, with relation to the projector's position. I would say that there isn't that much that can be done... It would be VERY difficult for the home brew dude to get that accurate with such a curve at that size, and it wouldn't do that much for you. (This is the curve associated with he projector's position ONLY...)

It has much more to do with the distance you are sitting from that screen size than anything else. If you wished to keep the brightness up, you would have to move the projector forward... and get a lensing system that allowed you to accurately place such an image size... that much closer to you. Such a lensing arrangement for that big a screen, at such a close distance would probably employ corrective elements that would wreak havoc with your plans for a curved screen...focus at the outer edges would be screwed. This is where good old CRT projectors score big. Almost every one of them can handle such situations right out of the box. (The CRT projector would be closer to the screen)

It all depends on what is available for your projector as far as lenses goes, and if you want to go through the tremendous bother and work required here...

The angle characteristics of how far off axis the outer edges are from you right now, compared to the center angle, is what your problem is... the screen is simply too big, and you are too close to it.

I have a feeling that the lenses are available to do what you want, but the construction characteristics of the screen would be a bit tricky... the screen would be good for only the one set-up.

------------------
---Place Signature Here---

Actually, if you wanted to reduce hotspotting by moving the projector, you would move it *farther* away. That would reduce the angle of the light reflected at the edges of the screen and direct it closer to the viewing area and less into the walls on the side.

As for the depth of the curve, you should probably experiment with your projector when it is installed (before you make your screen...). Focus it on a screen sample or a wall, then take a piece of paper and hold it at different distances from the wall. See what range of distances produces a sharp image (look for sharp inter-pixel gap) on the paper and that is the maximum depth of the curve you should use. Also, measure at the center, edges, and corners as the results may not be the same (try and stay within the smallest value). Obviously, sharpness is a matter of preference and you can push it a bit, I believe when someone measured an ISCO + D-ILA the depth of field was about 6" (which is quite a bit more than the 1" you want). I don't know what the half-gain angle of the m2500 is, but any kind of curvature will improve your image even though some hotspotting may still be visible.

I assume you are going to construct a horizontally curved screen? In this case it would definitely be achievable and a curve with a depth close to the focal depth you measure will provide an improved image with a wide-oval shaped hotspot rather than a smaller circular hotspot, definitely an improvement. Even if your curve is not perfect it will still provide a better image (as long as you can't notice the imperfection).

Regards,

Kam Fung

DOH! you are right Kam... i have compound curve on the brain as of recent. i keep forgetting that there is a perfectly good single curve screen that is easy to implement!

------------------
---Place Signature Here---

You can do a test with a piece of bendable plastic, and a white bedsheet hung from it.

It is best to do this test with a few hours of viewing lots of different material under your belt. 'Simple the screen may be to make, but live with it long you will....'(Yoda screen lore quote of the day)

------------------
---Place Signature Here---