Tweet
I received one of these a bit before Christmas and have used it for about 6 movies now and feel like I can comment on its value to LCD (and to a lesser extent , DLP) owners.
The way it functions is fairly basic in principle. There is a two optical element system which splits the beam (and image) and offsets one of these by an adjustable amount. So after installing the unit, your first step is to align the pixels of the original ad 'ghost' image (ghost being a misleading term as the two images are similar in brightness, although the ghost is a bit dimmer). Once aligned, and with your projector sharply focused, you then adjust the ghost position until it is offset from the original enought that the ghost pixels are shifted up from the originals by about the amount of the width of the gridlines (screen door). You then rotate the assembly so the the ghost pixels are offset diagonally from the originals and are thus lighting both the horizontal and vertical gridlines.
The instructions then detail further steps to insure you have a uniform effect across the entire screen, but this was not necessary for me. If you PJ/screen setup is not geometrically ideal, you may find you need to do these.
THe net result is that you now have very close to 100% fill factor with no perceptible softening of the image. Accordingly, the improvement is the greatest with LCD. I have a Studio Experience 20HD, which is a clone of the Sanyo PLV-70, a 1366x768 2200 lumen 900:1 CR LCD PJ (more like 1600 lumens at 600-650:1 after calibration). I have a 119" diagonal 16:9 2.8 gain screen that I sit less than 11' feet from. At this distance, I was always aware of the SDE from my PJ, but valued the immersiveness of having my nose in the screen over moving back far enough to obviate it.
Now it is not a problem. This device does exactly what it is advertised to do. Some caveats - my older LCD has fairly noticeable vertical banding, moreso than my friend's HS20 or likely any of you with Z2's or newer Panasonics, Yamaha's etc. But this device does nothing to mitigate that. I had not really expected it to, but had a small hope it would be less perceptible if the image had been smoothed. Nope.
Interestingly, the lack of the visible grid has made some source artifacts MORE visible. On pans, I can now more easily see pixelization, as it abruptly disappears when the pan stops or slows, whereas before it was similar in texture to the grid from the PJ. Also, changes in or lack of focus in the source media are starkly more visible. In the 2nd movie I watched, I had begun to wonder if I had bumped the thing somehow and softened focus, or if it was indeed softening the image and I was just noticing it now, when it shifted to a different scene in the movie (which I had seen before on my system, BTW) and the focus was then as sharp as ever. I saw this same phenomenon in other discs as well.
This all sort of underscores the idea that digital projectors can give a false sense of a sharp image because of the high frequency sharp dark grid of the pixel array. Once it is removed, you can then more easily see the actual differences in sharpness of the source.
Anyway, if you have an LCD PJ, and have any issue with SDE, this is a really really good solution for you. And if you are contemplating a PJ purchase, and think LCD is a good solution but are dismayed by SDE, this is a real and true fix.
I spent the better part of an hour adjusting the IMX, but could probably slightly improve on the results with a bit more time. I would say the net achievable result is to put the smoothness of the image at a bit better than a good DLP. There is no softening of the image from a viewing standpoint, although, realistically, as with all optical devices, it is not perfect and you are seeing a little loss of pixel to pixel contrast. This does not mean your image is losing intra-frame contrast, unless the image you are looking at is alternating columns of black and white pixels, the bleed is only into immediately adjacent pixels and is faint (not the ghost pixels, but much fainter diffuse light). In mine, the ghost pixels themselves are positioned slightly beyond the grid in the veritcal axis. This is the part I feel I could improve on with a little more time. The process could be performed much more easily with someone to help and I did it alone, requiring about 40 trips from Pj to screen and back.
I don't have any screenshots at present, but will try to take some full screen, medium level (to show smoothing) and closeup (to show pixels) when I return home.
BB
The way it functions is fairly basic in principle. There is a two optical element system which splits the beam (and image) and offsets one of these by an adjustable amount. So after installing the unit, your first step is to align the pixels of the original ad 'ghost' image (ghost being a misleading term as the two images are similar in brightness, although the ghost is a bit dimmer). Once aligned, and with your projector sharply focused, you then adjust the ghost position until it is offset from the original enought that the ghost pixels are shifted up from the originals by about the amount of the width of the gridlines (screen door). You then rotate the assembly so the the ghost pixels are offset diagonally from the originals and are thus lighting both the horizontal and vertical gridlines.
The instructions then detail further steps to insure you have a uniform effect across the entire screen, but this was not necessary for me. If you PJ/screen setup is not geometrically ideal, you may find you need to do these.
THe net result is that you now have very close to 100% fill factor with no perceptible softening of the image. Accordingly, the improvement is the greatest with LCD. I have a Studio Experience 20HD, which is a clone of the Sanyo PLV-70, a 1366x768 2200 lumen 900:1 CR LCD PJ (more like 1600 lumens at 600-650:1 after calibration). I have a 119" diagonal 16:9 2.8 gain screen that I sit less than 11' feet from. At this distance, I was always aware of the SDE from my PJ, but valued the immersiveness of having my nose in the screen over moving back far enough to obviate it.
Now it is not a problem. This device does exactly what it is advertised to do. Some caveats - my older LCD has fairly noticeable vertical banding, moreso than my friend's HS20 or likely any of you with Z2's or newer Panasonics, Yamaha's etc. But this device does nothing to mitigate that. I had not really expected it to, but had a small hope it would be less perceptible if the image had been smoothed. Nope.
Interestingly, the lack of the visible grid has made some source artifacts MORE visible. On pans, I can now more easily see pixelization, as it abruptly disappears when the pan stops or slows, whereas before it was similar in texture to the grid from the PJ. Also, changes in or lack of focus in the source media are starkly more visible. In the 2nd movie I watched, I had begun to wonder if I had bumped the thing somehow and softened focus, or if it was indeed softening the image and I was just noticing it now, when it shifted to a different scene in the movie (which I had seen before on my system, BTW) and the focus was then as sharp as ever. I saw this same phenomenon in other discs as well.
This all sort of underscores the idea that digital projectors can give a false sense of a sharp image because of the high frequency sharp dark grid of the pixel array. Once it is removed, you can then more easily see the actual differences in sharpness of the source.
Anyway, if you have an LCD PJ, and have any issue with SDE, this is a really really good solution for you. And if you are contemplating a PJ purchase, and think LCD is a good solution but are dismayed by SDE, this is a real and true fix.
I spent the better part of an hour adjusting the IMX, but could probably slightly improve on the results with a bit more time. I would say the net achievable result is to put the smoothness of the image at a bit better than a good DLP. There is no softening of the image from a viewing standpoint, although, realistically, as with all optical devices, it is not perfect and you are seeing a little loss of pixel to pixel contrast. This does not mean your image is losing intra-frame contrast, unless the image you are looking at is alternating columns of black and white pixels, the bleed is only into immediately adjacent pixels and is faint (not the ghost pixels, but much fainter diffuse light). In mine, the ghost pixels themselves are positioned slightly beyond the grid in the veritcal axis. This is the part I feel I could improve on with a little more time. The process could be performed much more easily with someone to help and I did it alone, requiring about 40 trips from Pj to screen and back.
I don't have any screenshots at present, but will try to take some full screen, medium level (to show smoothing) and closeup (to show pixels) when I return home.
BB