For LCD tablets, which include Wacom TabletPC and PL/Cintiq tablet, calibration is needed to resolve parallax issue. As the word parallax itself stands, the issue lies in the point/direction you look at the LCD. So, after rotating the display, a new calibration is needed since you now changed the way you look at the LCD.
 
The difference of TopXY/BottomXY among systems can be caused by looking at the LCD from a different point/direction while calibrating the tablet. Unless you calibrate all the tablets from exactly the same point and direction, which is very hard if possible, you'll see different TopXY and BottomXY on different systems no matter whether they are identical or not.
 
Hope this helps.
 
Ping
 
On 10/28/05, Jeremy Moles <jeremy@emperorlinux.com> wrote:
I was wondering if someone could please clarify something for me
quickly:

We currently have a number of __identical__ (inasmuch as we can tell,
though they obviously aren't) IBM X41 Tablets that we have been working
with for quite some time. For the most part, everything is working
okay--or can be made to work w/ some coaxing. :) However, I'm looking
for clarification on one small issue that I can't find any pertinent
documentation on; which is saying a lot, since my google-fu is usually
pretty strong.

My question is:

What exactly are TopX, TopY, BottomX, and BottomY--and how do specify
them?

>From what I can tell The Top* and Bottom* pairs are 2D coords pertaining
to physical positions on the tablet, correct? Relative to the
orientation,right? That seems to be the case 100% of the time in
"normal" mode. However, when we rotate the display CW, technically, if
this were true, we would only have to switch around BottomX and BottomY
and everything should still work? As it stands, I have to go in and
tweak these values by hand (while rotated) to get the pointer to
behave.Even stranger is that for all of these systems with the exact
same part numbers (and sequential serial numbers!!!) these values are
always different.

*boggle*