[comp.sys.atari.st] Phantom

micro@imada.dk (Claus Pedersen) (05/08/90)

We hear about the Phantom all the time, at first I didn`t believe in
`it`... (for a year or so) and then it stroke me in Tempus.

I were able to have `it` for 3 hours (moving the mouse really speeds
things up).

Ok I were able to quit Tempus, returned to the desktop and every thing 
still were acted very slow (clicking and moving the mouse simultaneously is
kind of tricky...).

But now to the interesting parts, disconnecting the keyboard on my
MEGA (and thereby Reseting the keyboard cpu) did not affect the phantom,
notice that pressing the reset botton on an MEGA does not reset the keyboard
cpu...

More interesting things, starting new GEM programs did not affect the
phantom either...

BUT NOTICE - starting a non GEM program momentally stopped the phantom.
I started BUG - the debugger from the turbo C package, and everything
was as nothing had happened. I looked after suspicious code on the system
vectores (viruses) but I were not able to find anything.
When returning to the desktop the phantom were there again...

Then I killed the phantom, it`s easy - just press CTRL-ALT-DELETE and off
he goes... (or the little botton at the back called RESET).

Where does this leave us ?
  - Not hardware (surely not the keyboard).
  - Not in the TOS (Bug uses Gemdos, Bios, Xbios).
  - Could be in GEM somewhere (I had a drawing program running too -
    so it could not be in the VDI)
  + this leaves us with the AES... 
    - what about the dispatcher (evnt_xx calls) ??

I guess this problem is for the people at Atari to fix...


- Klaus    (micro@imada.dk).

boyd@fsucs.cs.fsu.edu (Mickey Boyd) (05/08/90)

In article <751@imada.dk>, micro@imada.dk (Claus Pedersen) writes:
>We hear about the Phantom all the time, at first I didn`t believe in
>`it`... (for a year or so) and then it stroke me in Tempus.
>
> [.....]
>
>- Klaus    (micro@imada.dk).

  Does this ever happen in Tempus II?  Is this bug fixable?

-- 

    -----------------------------------------------------------------------
    ---------------------------------+-------------------------------------
              Mickey Boyd            |    "Nobody can be exactly like me.
          FSU Computer Science       |      Even I have trouble doing it."
        Technical Support Group      |
      mail:  boyd@fsucs.cs.fsu.edu   |              - Tallulah Bankhead
    ---------------------------------+-------------------------------------
    -----------------------------------------------------------------------

bright@ccu.umanitoba.ca (Bob Bright) (05/09/90)

In article <9005081510.AA06370@fsucs.cs.fsu.edu> boyd@nu.cs.fsu.edu (Mickey Boyd) writes:
>>We hear about the Phantom all the time, at first I didn`t believe in
>>`it`... (for a year or so) and then it stroke me in Tempus.

>  Does this ever happen in Tempus II?  Is this bug fixable?

Yes, as least one user has reported being struck by the phantom in
Tempus II.

Fixable?  Well, first we gotta find it before somebody can figure out
a fix.

BBB
-- 
Bob Bright <bright@ccu.umanitoba.ca>
Dept. of Philosophy
University of Manitoba
Winnipeg, Man  R3T 2N2  (204) 474-9680

kron@aiki.Eng.Sun.COM (Kenneth &) (05/10/90)

Just a note I haven't been following the phantom dicussion closely but I
ran into the problem the other day while running Word Perfect 4.1.
Just thought I'd add a data point.  And quite by accident I also discovered
(as has been mentioned) that moving the mouse helps speed things up0>

kk
--

Kenneth Kron
-----
The opinions I gave were mine, the opinions you form are yours.

grahamt@syma.sussex.ac.uk (Graham Thomas) (05/29/90)

Bob Bright's recent posting on the two phantoms was extremely useful. 
First Word Plus users may be interested to know that I've copied it to
Simon Jones of GST Software Products.  They're just about to return to
work on improving/enhancing FWP for the ST.

Just to recap on FWP's relation to the bugs - version 2.02 (UK) used to
be hit by both the phantom repeater (3 characters repeating across the
current text line followed by bombs) and the phantom slowdown (keyboard
input and mouse clicks taking an age to be acted upon - Bob's message
gave some useful specifics on this).

Version 3.14 (UK), the current version, no longer seems to be affected
in the same way by either.  (Has anyone with v3.14 or equivalents in
other countries seen them?)  Instead, at unpredictable times, though
always during keyboard input rather than mouse operations, FWP
'freezes' - i.e. stops accepting either keyboard or mouse input.  If you
want to carry on writing, you have to reset, and you lose all your work
since the last save operation.  Things like screensavers and the
alt/help screen dump routine continue to operate (the latter is
interesting given Bob's surmise that the phantoms involve keyboard input
not getting through to GEM) but there's no way to continue word
processing or get back to the desktop or wherever you came from.  (I
don't know if program switcher like Revolver, twist, Juggler or K-Switch
would let you get out - does anyone have experience of this?)

I think this is a displacement of one of the old bugs.  It may or may
not have something to do with changes in the way FWP3 deals with
keyboard input.  FWP3 introduced keyboard alternatives for most mouse
operations, and the program now takes over the keyboard more
comprehensively than the old version used to.  (A side effect of this is
that Simon Poole's KEYEDIT program no longer works with FWP3.  I used to
use it with FWP2 to assign foreign language characters to the numeric
keypad.)

If anyone has any more comments on bugs in FWP3, I'd be willing to relay
them to the authors (assuming the message I just sent gets through OK). 
Or you could try mailing direct: the address I used is
sproducts%cix.uucp@ukc.ac.uk - your route may be a little different. 
Please don't mail general questions about First Word Plus or other
GST/Timeworks products - Simon Jones is a programmer rather than a sales
or marketing person.  I think he'd quite like to get bug reports, but
please be as specific as possible.

Graham

-- 
Graham Thomas, SPRU, Mantell Building, U of Sussex, Brighton, BN1 9RF, UK
 JANET: grahamt@uk.ac.sussex.syma  EARN/BITNET: grahamt@syma.sussex.ac.uk
 ARPA:  grahamt%syma.sussex.ac.uk@nsfnet-relay.ac.uk
 UUCP:  grahamt@syma.uucp    Phone: +44 273 686758    Fax: +44 273 685865