All 5020/6020 Owners, Please Help! Please answer this Poll in an attempt to get a response from... - Page 3

There's two parts to focus uniformity. First, just use the focus pattern built in to the PJ. It's best if you focus an area of center. Don't expect perfect uniformity, that's not going to happen and you don't need it. But do expect that you should be able to see decent focus on all parts of the screens. The corners are going to be worst. Then, put up the patterns you mentioned. That'll show fow well each of the colors is focused. All three colors should be reasonably close. You'll see some flare on each color. But it shouldn't reach the point where you're seeing what looks like two or even three pixels instead of one. Now, if you're using a lot of lens shift, you'll see worse flare, again especially in the corners. That's expected; that's life. What you don't want to see in the corners or edges is a color shift with a white field. For instance, if the red pixels are flaring badly, you'll see a pink tint in that area when it should be white. How much is too much? If it is bad enough that you can see it in a snowy scene or a cloudy scene. That'll happen if you can easily see it with a white test pattern. If you really have to look hard to see it on a white test pattern, odds are you won't be able to detect it while viewing normal material.

Remember, focus is extremely sensitive. 1/8" focus ring movement can easily take a focused picture and take it out of focus.

Don't worry if every pixel doesn't look perfect. They aren't going to. A little flare, a little bit out of focus, a little mis-convergence, is not going to hurt your picture. A lot is. When in doubt, post up pictures. Look at actual scenes. Move the menu around the screen. Does it look worse in the "bad" parts of the screen than in the "good" parts? View scenes from the closest seating position.

Here's the deal. Everything is a tradeoff. If you're going to be sitting very close to the screen--less than 1.1 x screen width--there are better projectors. if you're at 1.2 or beyond, will you be able to see the errors your PJ has? If not, the torch modes on the Epson are very nice to have. The 3D is great with RF glasses.

The problem seems to be--as indicated by this poll--that Epson's quality control is not where it used to be. People have always gotten a few bad units with every brand. That's what warranties are for. But I've been through a number of bad Epson PJs trying to get a good one this year. Hopefully I'll have more to report on this soon...

I discovered something about the problem with inverting some 5020/6020s. Some people have reported problems, which may not have been problems in some cases.

When you adjust the pixel convergence, you tell the red pixels to go up. (For instance.) When you invert the screen, up becomes down. However, the pixel alignment is still telling the red pixels to go up. Except that Up is now down, so the alignment adjustment is now telling the pixels to go exactly the wrong direction. So instead of aligning the pixels, the adjustment you had is un-aligning the pixels. Am I making sense? I tested this by not moving the projector, but changed the projection setting back to "Front" from "Front/Ceiling". Sure enough the pixels went out of alignment.

So, if you test a 50020/6020 on a bench before you ceiling mount it, and if you enable "LCD Alignment" while testing on that bench, and then you ceiling mount it, it's going to look like something moved inside because the convergence will now be horrible. The solution is to "reset" the LCD Alignment and start over.

Hopefully this information will help some people.