Re: [NTLK] Jaggies? 7-2008

From: Doug Parker <ispinn_at_gmail.com>
Date: Sat Jul 05 2008 - 11:49:32 EDT

> Another condition that I have seen is what I think of as true
> jaggies , the line being drawn seems to jump dramatically
> around the contact point making HW impossible.

Check out

     http://www.ispinn.com/jaggies

for a video snippet of jaggies-in-action.

The currently accepted theory is that the jaggies are caused by
the sum total of multiple pressure points on the screen coming
from a stylus *plus* points from the underside of the bezel.
(COL...the venerable Newton was some 15 years ahead of the
pack on the *multi-touch* concept!)

Some things to check are:

1) Try rotating your screen 180 degrees and see if the misalignment
goes away. Rotate the screen back when you're through--the 180
twist is a temporary testing step only. Oh--try other rotations as well.
This is my most valuable suggestion!

2) Try handwriting without touching the edge of the Netwon's bezel.
Try a heavy versus light touch. Look for repeatable differences.

3) Try writing in ink, in different sizes, and examining your scratchings
to look for obvious jaggies appearing on your scrawls. Compare with
different screen rotations.

4) Do vertical lines as in the link above, using the sketch setting.

Save your data. Why? BECAUSE I SAID SO! ;-) Seriously, just do it.

<things-that-make-you-go-hmm>
Isn't it curious to anyone that the jaggies only seem to appear in the
horizontal direction? I mean that when you watch the link above,
while making sweeping *vertical* lines, jaggies appear as slight
left-to-right misalignment bumps. Try making sweeping *horizontal*
lines. You won't get any slight up-and-down jaggies. I as yet have no
theory on this observation.
</things-that-make-you-go-hmm>

(1) *may* eliminate or move the pressure points around the bezel
edge so that jaggies are eliminated or at best reduced. (2) may or
may not help, but I included it as another testing step to get more
data. More data gives more info with which to decide how to remedy
the problem. (3) Again, is just for more data. Same for (4).

I have a 2100 that has no jaggies since I fixed it with the bezel trim
procedure, but I do notice that the alignment gets ever so slightly
off as my scrawling gets further into the bottom-right corner of the
screen. (See ** below.)

If your sweeping, vertical lines have jaggies, as in the link above,
do try the bezel-shaving procedure. If your alignment is slightly off,
well, I don't have a solid answer for that one right now. **It *may*
have something to do with the screen's old age and normal usage--a
pristine screen has a uniform surface, and a pressure point is
converted into an accurate x,y coordinate.

Older screens have surface material rubbed away from normal use
and *may* cause misregistration--the non-uniform surface begins to
relate touch points to x,y voltage levels slightly off from where they
once were. A test would be to get a slightly-off Newt and replace its
screen with a virgin screen (digitizer) and compare.

Compare it with what? *That's* why I told you to save your data! :-D

Doug

====================================================================
The NewtonTalk Mailing List - http://www.newtontalk.net/
The Official Newton FAQ - http://www.splorp.com/newton/faq/
The Newton Glossary - http://www.splorp.com/newton/glossary/
WikiWikiNewt - http://tools.unna.org/wikiwikinewt/
====================================================================
Received on Sat Jul 5 11:57:23 2008

This archive was generated by hypermail 2.1.8 : Mon Jul 07 2008 - 10:30:00 EDT