The XP21N came close to meeting a high quality projector for my viewing conditions, but the screen door was still a bit too pronounced for my setup.
Screen size 8' x 4.5'
Screen to audience 14'
Screen to projector 19'
HTPC with ATI Radeon All in Wonder
I tried the Cygnus IMX depixelation lens (about $900) to reduce the pixelation. The design of the IMX is quite clever for those who aren't familiar with it. It does not blur individual pixels or the image. Instead, it uses a pair of lenses to create a ghost of the image that can be shifted diagonally (or otherwise) by a user adjustable amount. The effect is equivalent to having 2 projectors half as bright overlaying their images with an offset that is a fraction of a pixel.
My test procedure was to (1) display an image where screen door was more prominent, (2) view it without the IMX from a distance where the effect was noticeable, then (3) view it again with the IMX at a distance where the picture was equivalent to without the lens.
The Good: Where properly calibrated, the IMX lens does reduce the screen door by 30% or better. In other words, the screen door effect for an image at 18' without the lens will look approximately the same at a distance of 12'.
This is a dramatic improvement in two ways. Obviously, it means I can sit where my normal viewing position is without a screen door effect. Second, it the primary artifact becomes pixelation and scaling due to resolution rather than screen door.
The Bad: Adjusting the IMX lens is tedius, even with a binocular. Not a surprise since I've heard it many times before.
The Ugly: I can't get the IMX to ghost uniformly across the screen. In fact, some areas of the screen are always shifted by more than a pixel regardless of any of the other areas whereas I would need consistent subpixel shift on all areas. This means there's a lot of bluriness for non-uniform areas.
I'm wondering whether this might be due to the shorter throw distance of the projector and the mounting of the XP21N. Perhaps the shorter throw distance and mounting away from the projector lens requires more uniformity across a greater region of the IMX optics than for other projectors.
The way that I test for uniformity is to place some 1 pixel width text on the middle of all 4 sides of the screen (left, right, top, bottom), then try to get the same offset for the ghosted image for all 4 positions. I haven't even tried the more rigorous test of placing the text in the all 4 corners yet.
Unfortunately, I'm fighting time and only have a couple more days to decide. There are two questions I have:
1. If you have an IMX, do you see uniform offsets across the entire screen?
2. I have a 4" dim green blob below the center of the screen. It's quite noticeable during credits, dark scenes, and dark shadows in that area. Does anyone having an XP21N see this?
Regards,
Piney
[This message has been edited by piney (edited 04-23-2001).]
Screen size 8' x 4.5'
Screen to audience 14'
Screen to projector 19'
HTPC with ATI Radeon All in Wonder
I tried the Cygnus IMX depixelation lens (about $900) to reduce the pixelation. The design of the IMX is quite clever for those who aren't familiar with it. It does not blur individual pixels or the image. Instead, it uses a pair of lenses to create a ghost of the image that can be shifted diagonally (or otherwise) by a user adjustable amount. The effect is equivalent to having 2 projectors half as bright overlaying their images with an offset that is a fraction of a pixel.
My test procedure was to (1) display an image where screen door was more prominent, (2) view it without the IMX from a distance where the effect was noticeable, then (3) view it again with the IMX at a distance where the picture was equivalent to without the lens.
The Good: Where properly calibrated, the IMX lens does reduce the screen door by 30% or better. In other words, the screen door effect for an image at 18' without the lens will look approximately the same at a distance of 12'.
This is a dramatic improvement in two ways. Obviously, it means I can sit where my normal viewing position is without a screen door effect. Second, it the primary artifact becomes pixelation and scaling due to resolution rather than screen door.
The Bad: Adjusting the IMX lens is tedius, even with a binocular. Not a surprise since I've heard it many times before.
The Ugly: I can't get the IMX to ghost uniformly across the screen. In fact, some areas of the screen are always shifted by more than a pixel regardless of any of the other areas whereas I would need consistent subpixel shift on all areas. This means there's a lot of bluriness for non-uniform areas.
I'm wondering whether this might be due to the shorter throw distance of the projector and the mounting of the XP21N. Perhaps the shorter throw distance and mounting away from the projector lens requires more uniformity across a greater region of the IMX optics than for other projectors.
The way that I test for uniformity is to place some 1 pixel width text on the middle of all 4 sides of the screen (left, right, top, bottom), then try to get the same offset for the ghosted image for all 4 positions. I haven't even tried the more rigorous test of placing the text in the all 4 corners yet.
Unfortunately, I'm fighting time and only have a couple more days to decide. There are two questions I have:
1. If you have an IMX, do you see uniform offsets across the entire screen?
2. I have a 4" dim green blob below the center of the screen. It's quite noticeable during credits, dark scenes, and dark shadows in that area. Does anyone having an XP21N see this?
Regards,
Piney
[This message has been edited by piney (edited 04-23-2001).]
