KDS-R50XBR1 Post OB Replacement Issue UPDATE Overscan "fixed", further geometry issue

EDIT
Thanks to KewlK the overscan problem mentioned in the OP has been evened out on the 1080i input, and I have to go in and adjust the amount in the other resolutions, however a further geometry issue has been realized, outlined in this post below.

Main Issue
I know these TVs have always been a "fun" topic here, but lets keep the typical haters out, I'm looking for people with experience with this issue.

On Tuesday of this week (4/21/09) I had my OB replaced for the first time on my June 2006 built KDS-R50XBR1 due to the green blob issue.

I was here for the replacement and everything looked fine, and immediately post replacement the tech reset my settings and we connected the HD cable box and PS3 to take a quick look and the difference was obvious immediately (just switched to default pro to check). He packed up and left and I didn't get a chance to really calibrate after that.

On Wednesday I got a chance to run the HD-DVD version of DVE from the 360 add-on. When I got to the geometry images I immediately noticed how off the screen was.

I tried changing resolutions from the 360 to 720 and 1080 and the problem was persistent. However since my 360 (component input) was replaced in between my last calibration and the OB replacement it was possibly a defect of the source, so just to eliminate that (and since my BD DVE won't arrive until later today), I downloaded the HDNet patterns and streamed them through my PS3 (HDMI input) and saw the same results.

I then decided to do a less scientific method and hit up Xylon's WALL-E comparison thread. I used the first image of WALL-E with the hubcap since it had too reference points on the picture, and I could easily see just how much of the screen was missing. It's OK on the left (about the maximum "ok" overscan, but still more than previously, and I'm not happy with it), however the right is missing approximately 1.65" to overscan. I have attached an image that displays what I'm seeing using Xylon's original.

The top is pretty much fine, but the bottom is missing about 3/8" to 1/2" inch as well.

Now this seems really odd to me on a non CRT display so the only thing I could think is that either the block isn't seated properly (which I don't even know if it has the room to NOT be seated correctly), or the projector in the block is defective.

I called the repair service back and talked to the tech who basically said he never had a replacement customer complain of overscan problems, but also admitted he didn't know of any who were using geometry patterns to calibrate either. He said I should call Sony back since they covered it.

I did that, which was the typical CS exercise in frustration with a lackey typing stuff in and (painfully) reading it back to me like a robot, he then recommended I call the repair service, and I told him I did, and they said to call him, and then he just repeated himself (the fun part was when I mentioned calibration, at which point I imagine he typed in the word "calibration" and he got a screen that he read about how sony doesn't cover the work of calibration services... I laughed at him pretty hard).

So I just called the manager of the repair place and he listened to the problems and said they'd discuss how to proceed (stating that he didn't feel he needed to come out with the level of checking I did). But I figured someone here might be familiar with why this is happening so I can account for any possible BS I might get fed.

Reference Image of what I'm seeing


Secondary Issue
I honestly can't remember if this was a problem or not with my first block, but when calibrating the color settings using DVE and the filters, Blue and Red were simple to balance, and green is off by quite a bit no matter what settings I change. I read a few threads on this and it seems to just be the way it is, but any comments on that too?

much thanks for your help KewlK. I was very willing to make mods in the service menu myself, but I was having a lot of trouble finding the right adjustment for the positioning (a lot of early threads referenced HPOS and VPOS which I knew was wrong, and what's funny is the tech called me and told to change those very settings, and I just said OK and blew him off). About the time you posted I did find the HST_POS and VST_POS reference in the XBR1 owners thread, but I do appreciate the clear guidance.

Moving on, I do output the PS3 (and thereby BD) and the 360 (and thereby HD-DVD) at 1080i, so regrettably I won't be able to adjust it any further. I've really never seen any obvious deinterlace effects on this set and it seems a shame for movie watching to bump down to 720p just because of the early limitation with these units. I will make those changes as you directed for SD (the Wii) and 720p (for the cable box when in that mode) to benefit those however.

Moving on, I made this change after I got my BD DVE HD Basics yesterday. When using the newer overscan test pattern that was not in the HD-DVD (the one with percentages), I was able to get 1080i to an even 2.5% all the way around in the center of each side. I emphasize that because I do apparently have some kind of geometry issue still.

As I said in the middle of each side I now have a very even amount of overscan the entire way around, but the 2.5% L markers in each corner are NOT in the same positions. The bottom left sits right about where it should, the top left has its left edge in the proper position, but its top edge almost touching the top of the screen. The bottom right's bottom edge is just barely visible on the screen with the right edge off, and the top right is basically completely off the screen (I say basically because a hair of the edge of the top is visible when looking up at the top edge of the screen).

So there's some obvious distortion of the test pattern, and although I'd say it's very close still, it isn't perfectly square.

Now is this an issue of the block, it's seating, something else? Or is this also something I just didn't take clear notice of with my first block that I just have to live with?

I figured I'd have to deal with the green issue, and as you said it's not terrible, just frustrating.

One final note, is there a clear reference that says that the tech is responsible to adjust the alignment of the screen after an OB replacement? I've seen references to this but no evidence of its validity. However I'm honestly glad he didn't, because he apparently didn't even know how to do it (instructing me to do the wrong thing). He never put up any test patterns to check the color either, he was content in showing the non-used tuner input snow, and I did some other checks then using test patterns and the old memory stick input check with color turned down.

Thanks again,
aC

If I knew I could have just adjusted the blocks positioning then I wouldn't have asked the question, now would I?

In regards to that, I'm not sure I understand what you're getting at in regards to the pixel mapping. Doesn't the HST_POS and VST_POS effectively just shift the location on the 3 panels that the image is rendered? How would that alter anything in regards to the mapping?

Also, I will note that when I saw the geometry issue after adjusting the HST_POS I did go back to its original setting and the problem was still there, so that is unrelated to the shift.

Thanks again KewlK for all the info. Any comment on what like.no.other. said though? I haven't heard anything like that from any of the comments on adjusting positioning on this set before until his now.

Wow... what a reminder of the dichotomy of AVS. Great, informative, helpful post by KewlK. Then the rest of the nonsense driven by like.no.other.