[comp.sys.atari.st] Phantom Typist

AMEIJ@vax.oxford.ac.UK (Jan Ameij) (02/08/90)

As requested:

a) What happened: Tempus II started delaying by tens of seconds between
keypress and effect. This was only in one file: another could be edited fine,
but on reloading the first the same happened. Hard reboot solved the problem.

b) Hardware: Mega 2, nothing else

c) Software: foldr200.prg, filefix.acc, S.I.Ramdisk @ 1Mb, Tempus II

End of Report.

RiddCJ@computer-science.birmingham.ac.UK (Chris Ridd) (02/08/90)

  Yep - this one strikes me too!  I've never ever seen this problem mentioned
in the ST mags or anything - is it common?

  I sometimes get some kind of infinitely repeated keys, but normally only
during bootup.  I do boot however, with a small clock setter (SUPCLKRD - the
Supra one) which sets the IKBD's version of the time/date (the new version
does, at least).

  I'm pretty careful about touching the mouse - I don't think it matters if
the mouse is moved, or not - it seems to happen 'randomly'.  Also, sometimes
the whole machine crashes after doing all the \AUTO\*.PRGs, with 3 bombs.
Again, 'randomly'.  Maybe it has something to do with the IKBD's time! :-)

  System: 1040STFM, Mono, TOS 1.0, AUTO contents in order: SUPCLKRD, FOLDR180,
DCACHE, LGSELECT, G+PLUS.  I've also had it with a bare bones AUTO: SUPCLKRD.
Other progs to look at/change the IKBD during bootup seem to die too, so Supra
isn't the culprit!


Personally I suspect a bug in the IKBD's ROM.  Any 6301 hackers out there?

    Chris

/*
 * Snail mail address:
 * Chris Ridd,                  "Wave after wave, each mightier than the last
 * School of Computer Science, 'Til last, a ninth one, gathering half the deep
 * Birmingham University,        And full of voices, slowly rose and plunged
 * UK                             Roaring, and all the wave was in a flame"
 *
 */

Z4648252@SFAUSTIN.BITNET (Z4648252) (02/14/90)

Scott Flinn writes:

"Has anybody else noticed any correlation between rapid sequences of
keystrokes (involving multiple backspaces) and the appearance of the Phantom?"

    I think that Scott has sort of hit the nail on the head.  But heck if
I can get the creature to show up at my beckoning.  I wonder if it takes
a certain combination of keys that occur upon fatfingering.  That would
explain why it is so difficult to get the critter to show up.
    Furthermore, I still feel that it is on the keyboard side and not the
motherboard side.  Seems the fact that it has happened also with Spectre
would be a significant clue.

Larry Rymal:  |East Texas Atari 68NNNers| <Z4648252@SFAUSTIN.BITNET>

R.D.Chafer@sysc.salford.ac.UK (02/16/90)

I have experienced the Phantom Typist a couple of times lately, only since
I bought a hard disk though (a Third Coast Technologies 45MB SCSI which I
would reccomend to anyone). The software I use is all my own and all the
keyboard interactions are via evnt_multi. No fancy keyboard tricks! Could
the use of fast DMA be affecting anything?
 
 
=========================================================================
From:  Robert Chafer
 
       FTN77 Group
       Computing Centre             Telephone: +44 61 745 5678
       University of Salford,       Fax:       +44 61 745 5666
       Salford M5 4WT
       United Kingdom
 
E-mail:
    JANET:   chafer @ uk.ac.salford.sysc
    ARPANET: chafer%uk.ac.salford.sysc @ nss.cs.ucl.ac.uk
    BITNET:  chafer%uk.ac.salford.sysc @ uk.ac
        or   chafer%uk.ac.salford.sysc%ukacrl.bitnet @ cunyvm.cuny.edu

Bob_BobR_Retelle@cup.portal.com (02/21/90)

I assume, based on KBad's comments that Alan Page is a poor programmer too,
or possibly I didn't sprinkle enough dried chicken blood around my ST, as
I've had the "Phantom Typist" hit me while using FLASH as a text editor...
 
Seems odd that someone who's been so involved with the ST has never heard of
this "feature" that's existed as long as the ST has been available...
 
BobR

mwjester@wsucsa.uucp (02/23/90)

In article <27180@cup.portal.com>, Bob_BobR_Retelle@cup.portal.com writes:
> I assume, based on KBad's comments that Alan Page is a poor programmer too,
> or possibly I didn't sprinkle enough dried chicken blood around my ST, as
> I've had the "Phantom Typist" hit me while using FLASH as a text editor...
>  
> Seems odd that someone who's been so involved with the ST has never heard of
> this "feature" that's existed as long as the ST has been available...
>  
> BobR

I used to use Flash occasionally as an editor while online ( before I switched
to Uniterm :^) - thanks Simon!), but I never ran across the phantom typist
while doing so - in fact I've never had it happen to me.  I have the 85 ROMs
and normally have some fairly minor accessories running - the control panel
and so forth.  What kind of things were you doing when it hit?

Max J.

Bob_BobR_Retelle@cup.portal.com (02/26/90)

Max J asked:
>I used to use Flash occasionally as an editor while online ( before I switch
>to Uniterm :^) - thanks Simon!), but I never ran across the phantom typist
>while doing so - in fact I've never had it happen to me.  I have the 85 ROMs
>and normally have some fairly minor accessories running - the control panel
>and so forth.  What kind of things were you doing when it hit?
 
I've used FLASH as my "editor of choice" for all my ST-Log articles, my
Dungeon Master hintbook, and most of my other text editing.. it doesn't have
all the "bells and whistles" of the "real" wordprocessors, but then again,
my applications have never required any of those features..
 
Essentially the same symptoms as reported by others, using other programs,
occurred...  I would be typing in new text at the bottom of the text buffer..
A series of quick backspaces would trigger the "phantom typist", resulting
in the last 10 or so characters being repeated over and over, until the syste
crashed...
 
It didn't happen often, but when it did, it was terribly annoying...
 
I wonder if Ken knows where I can buy a new supply of dried chicken blood..?
 
BobR

petey@fox.acslab.umbc.edu (Jack P. Axaopoulos; (C)) (03/03/90)

Moue crashes:

Your st will definitley crash if you hold down any of the mouse buttons
or click a button at the wrong time during booting.

just a note...

dlh@raider.MFEE.TN.US (Dana L. Holt) (03/05/90)

In article <2891@umbc3.UMBC.EDU>, petey@fox.acslab.umbc.edu (Jack P. Axaopoulos; (C)) writes:
> 

> 
> Your st will definitley crash if you hold down any of the mouse buttons
> or click a button at the wrong time during booting.

	Well, I would like to add that holding down BOTH mouse buttons
during boot will make the keyboard processor (6301) enter external ROM mode.
It looks for ROM chips which the ST doesn't provide so of course it crashes.
So, if anyone says their ST crashes at boot when they hold both mouse buttons
down, say: "It's supposed to do that!".. just another note...
					dlh@raider.MFEE.TN.US

Any questions? (You can buy a piggyback RAM upgrade to about 8K for the 6301)

bwhite@oucsace.cs.OHIOU.EDU (Bill White) (03/06/90)

	After days and days of using FW+ in an attempt to get the
Phantom to appear, I finally did it.  Unfortunately, I didn't have my
logic probe with me right there, but I did manage to get it after the
machine had bombed (four bombs; don't remember whether that's bus error
or what).
	Anyhow, counting from the top end of the connector, the values
I got were:
			L
			H
			H
			P
			H
			(gap in connector)
			P
			H

"L" and "H" are Low and High.  P indicates a pulse.  Let me explain:
at first, there was no data on P (I think it was high).  Then I pressed
the spacebar, and I got a "make" pulse and a "break" pluse.  After
doing this twice more, the pulses all of a sudden didn't stop -- they just
kept on coming.
	I have no idea whether this is an accurate representation of what
was actually going on, because I think I might have had the probe set in
CMOS rather than TTL.  I'm connecting a buffer and a set of LED's for
further verification so I don't have to type with my ST's case off.
	Anyone out there have a logic analyzer they'd be willing to send
me for further testing???


-- 
|   Bill White			Internet: bwhite@oucsace.cs.ohiou.edu	|
|	BILDO'S TWELFTH LAW:						|
|		The purpose of love is to distract people from		|
|		killing each other.					|

dlh@raider.MFEE.TN.US (Dana L. Holt) (03/07/90)

In article <2891@umbc3.UMBC.EDU>, petey@fox.acslab.umbc.edu (Jack P. Axaopoulos; (C)) writes:
> 
> Moue crashes:
> 
> Your st will definitley crash if you hold down any of the mouse buttons
> or click a button at the wrong time during booting.
> 
> just a note...

	This is just another note. When you hold down both of the ST's
mouse buttons during boot, it tells the keyboard processor (6301) to
enter "external ROM mode". The 6301 looks for ROM to execute, but
the ST doesn't provide it with any chips, so it crashes!
	dlh@raider.MFEE.TN.US

pegram@uvm-gen.UUCP (pegram r) (05/03/90)

From article <90.115.09:59:04@ira.uka.de>, by grunwald@Tokyo.ira.uka.de (Grunwald Betr. Tichy):
> May be your mouse cable is broken. This makes the mouse response awful, until
> the mouse breaks down and there is no more response. Because the cable moves
> the error is not stable. Try what happens, if you stretch the cable. But do
> it careful. I had that problem for several months until I had to recognise,
> that the mouse keys did not work and the pointer moves only in one direction.

> Knut Grunwald

Yes, that is a common problem, but it's not the phantom typist.
Wesley Kaplan mentions that it happens with rushed keyboard input,
especially with lots of deletions.  The same thing happened to me in
the exact same situation with Tempus II under Neodesk 2.03 and
including Turbo ST 1.6.  An awful lot of the original system was
replaced with that software running.  As was stated here on the net
before, the problem is relatively high up in the Tos - Gem software
hierarchy and possibly includes a problem with the IKBD processor
firmware.  I have the crash dump (it happened shortly after I
verified that I had finally seen the elusive typist) that Neodesk
provided and will post the thing when I remember to bring it in to
work.  Hope you ST wizards can make something of it.

Bob Pegram (Internet: pegram@griffin.uvm.edu)

P.S. 
   1985 520ST (no modulator), Tos 1.0, 2.5meg RAM, + hard disk. 
				RBPIII 8-)

pegram@uvm-gen.UUCP (pegram r) (05/08/90)

I coulda been a contenda, but....  everybody else has posted
interesting info on the Phantom first.  My 2 cents is this dump from
after my machine crashed when I experienced the phantom typist.  I was
using Tempus II, under Neodesk 2.03 with Turbo ST enabled.  I was
typing quickly and goofed (read pressed 2 keys at once) when I
deleted.  The last two characters typed started to repeat, and when
the characters reached Tempus' maximum line length, the machine
crashed.  Moving the mouse made the characters speed up.  Oh yeah, I
have Tos 1.0.

A Bus error has occurred (2 bombs)

	d0: 00200069	a0: 00000000
	d1: 000006DC	a1: 00000005
	d2: 00002300	a2: 0011A712
	d3: 0000004C	a3: 0011A732
	d4: 00000000	a4: 0011A746
	d5: 00000000	a5: 0011A74E
	d6: 00000005	a6: 0000799E
	d7: 00200000	a7: 00007972

	ssp: 00007972
	usp: 0012756A
	instr.: 1083
	sr: 2300 = {t:0 s:1 interrupt:011 x:0 n:0 z:0 v:0 c:0}
	pc: 00FE65EA

Good luck, those of you who can figure out what's going on here.

Bob Pegram (Internet: pegram@griffin.uvm.edu)