Re: [NTLK] Pen Alignment followup

From: Johannes Wolf (jwolf_at_xe.estec.esa.nl)
Date: Tue Mar 05 2002 - 10:21:39 EST


Regarding the requested info on how to disassemble your Newt:
go to our Mr. PCBmans website and have a look at his backlight project:
http://www.members.tripod.com/mrpcbman/backlight.htm
there you will find a fairly good description with photodocumentation on how
to disassemble your Newt.

Johannes

> -----Original Message-----
> From: newtontalk-bounce_at_newtontalk.net
> [mailto:newtontalk-bounce_at_newtontalk.net]On Behalf Of Sean Luke
> Sent: Tuesday, March 05, 2002 3:46 PM
> To: newtontalk_at_newtontalk.net
> Subject: [NTLK] Pen Alignment followup
>
>
>
> Yesterday I posted stuff indicating that my new Newton was off-kilter of
> the pen by several millimeters in the middle of the screen over on one
> side, but was fine in the corners. Here's a follow-up.
>
> (1)
>
> Deep in the C++ includes in NewtonDev is a C++ header file called
> "VirtualMemory.h". In there it has, among other things, a structure
> called SGlobalsThatLiveAcrossReboot, which contains, as you might guess,
> all the stuff that's guaranteed to stay in memory even through a
> reboot. Part of that is:
>
> // tablet calibration info
> long fTabletValid;
> // Table data valid
> long fTabletXScale; //
> Table x scale
> long fTabletXOffset; //
> Table x offset
> long fTabletYScale; //
> Table y scale
> long fTabletYOffset; //
> Table y offset
>
> What this means is that the Newton uses just two variables (scale and
> offset) per X or Y dimension. That's pretty bad IMHO, but that's what
> it is. Basically when you click on the left "X" during pen alignment,
> it stores the (x,y) coordinate where you clicked on (that's the
> offsets). Then when you click on the right "X", it computes the scale
> from the left "X" to the right "X".
>
> So if things are looking good around the locations where the "X"'s are,
> but not good in the middle, there's not a dang thing you can do about it
> in software. DANG!
>
> BTW, there is a NewtonScript function called CalibrateTablet() which
> calls the pen calibration program (written, it seems, in C++ not
> NewtonScript), and returns. There's no other way to calibrate. The
> calibration program is the screen with the Newt scribbling on a Newton
> that says "Your Newton needs to be calibrated for the way you write..."
> etc.
>
> (2)
>
> I took my new 2100 apart last night but could not figure out how to get
> at the screen. It seems to be buried deep underneath a pile of other
> stuff, and I couldn't remove that stuff. Has anyone else had experience
> getting to the screen layer? How do you do it?
>
> (3)
>
> I've discovered another problem with the 2100: the latch is basically
> broken -- it doesn't slide shut -- in fact it tries to spring to the
> "open" position. I'd also like to get to the screen layer to examine
> the latch and see what I can do about it. If nothing else, I will see
> if I can swap out the motherboard with my trusty MP2000.
>
> Sean
>
>
> --
> Have you checked the FAQ lately?
http://www.guns-media.com/mirrors/newton/faq/
This is the NewtonTalk mailing list - http://www.newtontalk.net
To unsubscribe/manage: visit the above link. Questions:
victor_at_newtontalk.net

-- 
Have you checked the FAQ lately? http://www.guns-media.com/mirrors/newton/faq/
This is the NewtonTalk mailing list - http://www.newtontalk.net
To unsubscribe/manage: visit the above link. Questions: victor_at_newtontalk.net



This archive was generated by hypermail 2.1.2 : Tue Apr 02 2002 - 14:02:10 EST