Originally Posted by Eric_S
I have successfully corrected the overscan on my XS955!
The image is much brighter and sharper now as an added bonus to seeing the whole image. I suppose it's because there is more energy from the scanning electron guns inside being focused in a smaller area than before.
Probably you're seeing more image (wait until you see those score bars during ball games, or info crawls on news channels, etc., and realize just exactly how much information there really was to be seen... including OUTSIDE what was the old image perimeter you were looking at before your tweaks).
Also, everything inside the constant screen size is now somewhat smaller (since you have more image in that screen) although there's more to look at, which has an automatic "sharpening" effect to our eyes and brains.
I did this by adjusting the VPOS, VSIZ, HPOS, HSIZ setting in group 2170D. I adjusted it as small as could, but it still cut off part of the screen.
In my own case with my XBR960, it wasn't just size and positioning in the horizontal and vertical directions which required adjusting... it was also centering.
So I had to play with VPOS, VSIZ and VCEN (2170D-1), along with HCNT, HPOS, HSIZ (2170D-2), and then VDHP, VHDS, VDVE and VDVS (MID3).
I also worked on tweaking pin cushion and corners, to get as much rectangularity and linearity on vertical and horizontal edges as I could. I cannot tolerate "bowing" and "curvature" at the corners of my screen.
From what I can tell the difference in the two sets of screen adjustments is that the first one is the size of the actual raster, and the other is a sort of digital image resizing to fit the image onto the raster.
I always describe it like with Photoshop... "background", and "image". If the background's not large enough you lose image (like cropping). If the background's too large you have extra black space with no image on it. If the image's too small you're not taking advantage of all the screen real estate. If the image's too large then the smaller background will result in loss (crop) on the edges of the image.
Fairly delicate process of adjusting three tweaks all together, but eventually you do zero in on where just +/- one click on anything obviously causes you to say "no... that's worse". Now you know you're done.
I have lined them up closely, but they do not line up pixel to pixel. Will this cause any negativity?
This is an analog set, and is therefore not pixel-perfect.
I'd say just get as close as you can, so that you're really happy... and be done with it.