Wonky TI99/4A
Simon D. Williams
I have a silver TI99/4A that I picked up at a thrift store several years
ago, and I've just got round to playing with it ;-)
Seems to be something screwy with it... it boots to the splash screen with
a smattering of grabage characters. Pressing [1] twice gets me the BASIC
prompt, but keyboard entry and character display is mucked up. Characters
seem to be missing a row of pixels on one side or the other. Also, the
characters echoed onscreen are wrong, though consistently so, with the
exception of capital letters.
Is there a self-test mode or some other way of diagnosing the problem? At
teh moment I have no way of loading programs.
] SIMON D WILLIAMS
winston19842005
Pine.NEB.4.64.09...@retro-net.org, "Simon D. Williams"
<ad...@retro-net.org> wrote:
No self-test mode, just speculation.
Possibly bad VDP RAM, VDP chip not refreshing the RAM properly or GROM that
contains the VDP character definitions either bad or weak.
If you can type a small program in:
10 FOR I=32 TO 159
20 PRINT CHR$(I);
30 CALL CHAR(I,"FFFFFFFFFFFFFFFF")
40 NEXT I
50 GOTO 50
That is 16 "F"'s in line 30. And see if all the characters turn to a solid
block and hold it. You can change all the "F"'s to "0"'s to clear them and
check that too.
If they hold, then refresh is ok (probably is ok anyway, as the characters
should "decay" over time with a refresh issue).
If they change, you may have a refresh problem. If they are not solid blocks
or clear then you probably have a bad VDP RAM, especially if some are
solid/clear and others are mixed.
If it works, then break the program (FCTN-4). Note the last characters
printed - they should still be solid blocks. The others should change back
to normal definitions.
Press enter a couple of times to see if the characters change. If so, you
probably have a weak GROM. I have a TI computer that starts out ok, but as
you press enter the characters start to change. Apparently, every time you
press enter the computer reloads the standard character set from GROM.
In any event, you can always try the "drop test", as an Apple user you
should be familiar with that, drop the computer an inch. Also, reseat the
chips - and refresh the heat-sink grease on the VDP chip - big chip behind
the keyboard amidst all the small RAM chips.
There was a diagnostic module to check VDP RAM, but they were not all that
numerous, probably mostly used by technicians. I don't have one, I have the
program on disk.
Simon D. Williams
> No self-test mode, just speculation.
Next best thing ;-)
> If you can type a small program in:
No such luck... it prints * INCORRECT STATEMENT when I enter a line.
Dropping it didn't help... I'll take a screwdriver to it when I'm feeling
more adventurous.
] SIMON D WILLIAMS
winston19842005
Try a simple line, like:
10 REM
Simon D. Williams
> Try a simple line, like:
> 10 REM
It will let me enter that, but anything more complicated gives me
* INCORRECT STATEMENT as sson as I hit [ENTER]
I also get the same error if I try to RUN or LIST
FWIW, I took it apart and reseated the socketed chips, blew the dust out
and played hide-and-seek with the loose screws ;-)
] SIMON D WILLIAMS
winston19842005
Weird. Gives me the impression that what you are typing is not what is being
seen by the computer. With the video issues you are having, can you see that
you are typing "RUN" or "LIST"?
Even if there is no program, a LIST should give "* CAN'T DO THAT", and never
an "* INCORRECT STATEMENT" (TI's version of sytax error until XB where they
actually use "* SYNTAX ERROR")
Simon D. Williams
> Weird. Gives me the impression that what you are typing is not what is being
> seen by the computer. With the video issues you are having, can you see that
> you are typing "RUN" or "LIST"?
>
> Even if there is no program, a LIST should give "* CAN'T DO THAT", and never
> an "* INCORRECT STATEMENT" (TI's version of sytax error until XB where they
> actually use "* SYNTAX ERROR")
I have to hold down [SHIFT] or use [ALPHA LOCK] to get the display to
match what I'm typing. Unshifted QWERTYUIOP shows as 17%2495)/O which I
just realised is what you'd get if you subtracted 64 from the ASCII value
of teh input... well, that gives me something to play with...
] SIMON D WILLIAMS
winston19842005
Hmm - Input or Output problem?
Another thing to add. ASCII values being what they are, it MAY be helpful
for you to know that an "A" on the screen in BASIC, while a CALL GCHAR will
return a 65 for the value, the actual value of the character on the screen
is 65+96=161 - i.e. to save memory, the TI actually uses character 161 to
display an "A".
That may help to determine if the problem lies in the inputting of the
character or the outputting to the display. I'm starting to think the former
rather than the latter. Sure would help if you had access to routines to
PEEK and POKE into VDP RAM - do you have anything else with your TI?
Simon D. Williams
- All regular numeric and symbol keys work, shift functions correctly
- Uppercase letters work correctly
- [FCTN] characters work except:
[~] = >
[|] = <
[{] = ;
[}] = =
[`] = space
Which are maddeningly consistent with the ASCII - 64 thing.
Keyboard encoder?
] SIMON D WILLIAMS
Simon D. Williams
> character or the outputting to the display. I'm starting to think the former
> rather than the latter. Sure would help if you had access to routines to
> PEEK and POKE into VDP RAM - do you have anything else with your TI?
Unfortunately it's just a bare-bones box with the power supply and RF
converter.
] SIMON D WILLIAMS
MarkWills
Sorry to hear of your TI woes. The '64' problem leads me to think
there could be a problem either on the keyboard (broken track?) or the
keyboard connector. Failing that, the 74LS156 on the mother would be
my next suspect. These are easy to get and very cheap. To remove, just
snip the pins off the chip with snippers, and throw the legless chip
away. Now you can unsolder the pins one at a time, using a solder
sucker to open the plate-through holes. Solder in your new '156, et
voila!
More info here:
http://www.nouspikel.com/ti99/titechpages.htm
Follow the keyboard link.
Mark