MathGroup Archive 2000

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Re: Bug: Mathematica 4.0 vs. Win2k on ThinkPad

  • To: mathgroup at smc.vnet.net
  • Subject: [mg25959] Re: [mg25944] Re: [mg25931] Bug: Mathematica 4.0 vs. Win2k on ThinkPad
  • From: John Fultz <jfultz at wolfram.com>
  • Date: Tue, 14 Nov 2000 03:46:53 -0500 (EST)
  • Sender: owner-wri-mathgroup at wolfram.com

That has not been my impression at all.  Since Windows 2000 came on the 
scene, Wolfram has only confirmed one bug specific to Windows 2000 (a small 
repaint problem in the help browser which only occurs in 8-bit color 
mode).  Other than that, I've not even heard rumors of problems specific to 
Windows 2000.  Additionally, I've closely monitored this other forums for 
potential compatibility issues specific to Win2K (and Win ME, too), and 
none have arisen other than the one I mentioned above.  Not to mention that 
my daily usage of Mathematica under Win2K for the past year hasn't turned 
up anything.

Now that I take a closer look at Mr. Jeppson's original email, I see that 
there is one interesting thing about the cursors that are missing...they 
are cursors which contain no color, but only invert the screen (e.g. black 
is inverted to white, white is inverted to black).  The cell and text 
insertion cursor invert only, but the cell bracket cursor is a painted 
cursor (e.g. black with a white background).  This is not so unusual...in 
fact, the default system text cursors (which you'd see in MS Word and 
WordPad) do the exact same thing, although it is possible that the system 
treats them somehow differently since they are the system defaults.

I wonder if there might be some strange video driver thing or, as you 
suggest, a pointer device driver thing.  I did a bit of checking on 
deja.com and ibm.com, and it does seem to be the case that earlier versions 
of the pointer driver that shipped with the T20 are flaky (nothing quite 
like this, but several other problems which seem to make it worth upgrading 
nonetheless).

Oh, and I suppose something else to check would be to make sure your 
default window background color isn't set to gray.  Just the right shade of 
gray (i.e. exactly halfway from black to white) would have the affect of 
making the cursor appear invisible, since the inverse of gray is 
gray.  This wouldn't be specific to Mathematica, of course, if you're using 
the default system cursors.

John Fultz
jfultz at wolfram.com
User Interface Group
Wolfram Research, Inc.


At 01:40 AM 11/10/2000, Mark Harder wrote:
>My impression is that there have been some problems with Mathematica that
>are specific to Win2k systems; i.e. they didn't occur on WinNT systems.  One
>thing that hasn't been tried in your case is running your Mathematica 4.0.2
>on a think-pad with WinNT.  Can you get a friend, dealer, or IBM itself to
>try this with you?
>
>Another thought: What is the pointing device on the ThinkPad? Could your
>problem have to do with a non-standard pointing device (eg. that little red
>joystick thing on the keypad)? Maybe there are alternative drivers out there
>that will interface better with Mathematica and/or Win2K.  I ask this
>because I think there are lots of steps the operating system must carry out
>to decode the screen position of the pointer based on pointer motion, then
>decode intra-window position based on the desktop display, then pass this
>info to Mathematica, which has to be able to use that info to detect the 
>"pointer
>entering/leaving" events that depend on pointer position, etc., prior to
>changing the screen display appropriately, so its easy to imagine the bug
>lying in the pointer driver info somewhere.
>
>-mark harder
>
>-----Original Message-----
>From: John Jeppson <jepp at kern.com>
To: mathgroup at smc.vnet.net
>To: mathgroup at smc.vnet.net
>Subject: [mg25959] [mg25944] [mg25931] Bug: Mathematica 4.0 vs. Win2k on ThinkPad
>
>
> >Can anyone confirm (or refute) this bug I'm getting.
> >
> >Mathematica 4.0.0  (or 4.0.2)
> >Windows 2000
> >IBM Thinkpad T20   (machine type 2647)
> >
> >Fail Case:
> >
> >1.)   Open a new blank notebook
> >
> >2.)   Type:    2 + 2
> >
> >3.)   Run it.   -->  Output line says:  4
> >
> >4.)  Now move the cursor over the window. While over the input or
> >output text, the cursor should be a vertical bar with little  v's at
> >top and bottom.  Over the area between the input and output lines the
> >cursor should flip over on its side and become a horizontal bar
> >indicating the location where you can insert a new cell.
> >
> >5.)  On my machine this does NOT happen. The cursor won't change by
> >moving around within the text areas of the notebook window (although
> >it does change shape over the window frame and the little brackets
> >that lie along the right side of the notebook cells).
> >
> >Wolfram tech support has been unable to duplicate this because they
> >apparently don't have an IBM ThinkPad running w2k.  My machine itself
> >has been back to IBM and declared okay, but still has this bug.
> >
> >It does not seem to  be my CDROM disk since I can reinstall 4.0 on my
> >Dell desktop (Windows 98) and it works correctly, and Wolfram tech
> >support sent me another disk (4.0.2) which acts just the same.
> >
> >It can't very well be a program conflict because my machine was
> >completely reinitialized by IBM to the original factory state, and
> >Mathematica was the first and only program installed.



  • Prev by Date: Q: Symbolize: How to create a table of indexed symbols?
  • Next by Date: DumpSave
  • Previous by thread: Re: Bug: Mathematica 4.0 vs. Win2k on ThinkPad
  • Next by thread: Re: Bug: Mathematica 4.0 vs. Win2k on ThinkPad