#A1200 system lockup
30 messages in this thread
Has anybody come across a problem with A1200s which has the symptom of the
system locking up randomly (as far as I can tell) after anywhere from 1/2
to several hours use. Screen locks up, drives stop clicking, sound
becomes a monotone at the last note played and all keyboard input is
ignored except the ^AA sequence for reboot.
It seems to happen when there is one of the mem/clock/maths co-processor
boards in the belly regardless of brand but different brands seem to have
an effect on the rate at which this happens.
Lu, 2500/30 APpy downunder
My 1200 has never done this…
I have a microbotics '030 accellerator, SCSI+ controller, 8M fast ram, 3
floppy drives (one HiD), and a bigfoot power supply. I only reboot once a
week when I'm not testing programs.
You might be having an overheating problem, not a strong enough power
supply, or some other hardware problem. Could be a software problem
too…
Bob Comer — Hakuna Matata!
Cheyenne, Wyoming U.S.A.
Software problem is unlikely as it happens regardless of program or
whether I boot off the harddrive or a floppy. The overheating or power
supply are intersting possibilities … I may have been taking the power
supply for granted since it's the C= one that came with it.
Your `bigfoot' power supply, is that C= or 3rd party?
Lu, 2500/30 APpy downunder
>I may have been taking the power supply for granted since it's the C= one
>that came with it.
The one that came with my 1200 was pretty weak… Power supply problems
can do some wierd things.
>Your `bigfoot' power supply, is that C= or 3rd party?
3rd party (Micro R&D) 200W. 🙂
Bob Comer — Hakuna Matata!
Cheyenne, Wyoming U.S.A.
> 3rd party (Micro R&D) 200W
Thanks, I'll look into it …. I take it that you're happy with it.
Lu, 2500/30 APpy downunder
Yep, I am. (I have two, one for my 1200, and one for my 500.)
Never had any trouble with either one.
Bob Comer — Hakuna Matata!
Cheyenne, Wyoming U.S.A.
Thanks, I'll keep an eye out for a local distributor.
Lu, 2500/30 APpy downunder
LB >> Screen locks up, drives stop clicking, sound becomes a monotone at
the last note played and all keyboard input is ignored except the ^AA
sequence for reboot.<<
I haven't had these problems, and I've been using a DKB A1202 board with
16Mhz FPU, 5-megs RAM, and clock.
I have seen some funky sprite patterns, but I believe that was due to
compatiblity problems between certain games & AmigaDOS 3.0.
Have you checked the power supply? It may be a loose power connector, or
overheated powersupply.
Ron. – ("I'm not stupid, I'm not expendable, and I'm not going!"
– quote made by any low level RPG player.)
Would a faulty power supply give lock up symptoms? I've already changed
the power supply under warranty because it carked it but I assumed that
this sort of thing either lived or died and the computer along with it.
Lu, 2500/30 APpy downunder
LB >>Would a faulty power supply give lock up symptoms?<<
Its possible. If the powersupply is not providing enough power, it can
affect the computer's chips (weak signals, and overheating chips).
Since you've already changed the powersupply, check other factors. You
might have a bad RAM chip, one or both CIA chips, flakey keyboard, or it
could be bad software (one of the files normally loaded at bootup may be
corrupted).
Ron. – ("I'm not stupid, I'm not expendable, and I'm not going!"
– quote made by any low level RPG player.)
I thought the CIA chips were only in question for serial/parallel IOs ?
I came across an A1200 diagnostic program which said my `blitter' was
faulty but I seem to recall somebody saying this could happen with
diagnostics because there were different Agnus versions for the A1200.
Lu, 2500/30 APpy downunder
LB >>I thought the CIA chips were only in question for
serial/parallel IOs ?<<
Basicially, yes. But sooner or later, they get the blame for anything
that goes wrong. 🙂
—
LB>> an A1200 diagnostic program which said my `blitter' was faulty but<<
An diagnotic program that isn't designed for the AGA chip set will most
likely fail the graphics hardware of the A1200/4000.
Ron. – ("I'm not stupid, I'm not expendable, and I'm not going!"
– quote made by any low level RPG player.)
LB>> an A1200 diagnostic program which said my `blitter' was faulty but<<
> An diagnotic program that isn't designed for the AGA chip set will most
> likely fail the graphics hardware of the A1200/4000
The program was apparently specifically for A1200 but I don't have any
information as to its source. It was something I downloaded from a BBS,
no docs, nothing else.
Lu, 2500/30 APpy downunder
Lu,
not quite complete. Other things are tied to the 8520's as well, such
as your floppy control signals, your mouse, your keyboard, etc.
Basically, if it is I/O, it goes through the 8520's, with the exception of
video.
"You do trust me, don't you? Of course you do."
— "To Play the King" (BBC)
OK, but wouldn't the 8520's be either dead or alive? Even if one of tem
decided to collapse, wouldn't the floppy or keyboard or whatever just stop
responding and the rest of the system carry on as usual 🙂
Lu, 2500/30 APpy downunder
Lu,
with electronics, things are never simple. Given that any 8520
consists of two seperate circuits sharing some signals, such as the clock,
you have the possibility of having only one-half or the totality of a
problem.
Actually, it is even worse than that. Since some signals on a
particular half of an 8520 are split between devices, a single input or
output can change the output only with respect to that device. It is only
when you swap the 8520's that you have a chance of catching the particular
input or output that was affected.
Basically, from an electronics technician's viewpoint, the 8520's are a
troubleshooting nightmare. If I hadn't spent several years tracking
similiar problems in the remoting gear for the Precision Approach Radar, I
wouldn't even want to address this one. Too many signals shared too many
ways. The best we can do is to swap the 8520's and see if the problem
moves. If it does, then test to see where the problem moved to. Usually
that will tell you, within a pin or two, where the problem is. Not
exactly useful, since you will have to replace the 8520 anyway, but it
does tell us that the problem was within the 8520 in the first place.
Ummm, one more item. I've seen quite a few cases of half-dead circuits
on IC's. Given a certain set of inputs, the output would be screwed up,
but if you gave the chip any other set of inputs, it worked fine. Usually
this is a case of true IC failure. Basically, the chip is easily confused
as to what it is supposed to do. This only happens about 10% of the time.
Most failures are on the input or output side, at the chip lead level,
thankfully.
With electronics, dead or alive is not enough. You have to ask,
exactly WHAT is dead or alive. Life was much simpler in the DTL and TTL
days <sigh>.
"You do trust me, don't you? Of course you do."
— "To Play the King" (BBC)
Lu,
and to add to Ron's list, connectors that are not quite connected,
chips that have worked their way loose, and dirty circuit board edge
connectors. Quite a few things can go wrong, sad to say.
As for the powersupply, if you are electronically inclined, you might
want to check it under load to see what the voltages are. They are one of
the first things I check in any electronic system, after switchology
(switch positions).
BTW, all my random lockups here have been due to chips becoming
unseated or edge connector problems.
"You do trust me, don't you? Of course you do."
— "To Play the King" (BBC)
You wouldn't happen to have the pin outs for the power supply would you?
The A1200 came without a manual.
Lu, 2500/30 APpy downunder
Lu,
perhaps. I've only got A500, A1000 and A2000 schematics. If the
pinout is similiar to the A500, it uses a square connector, then it would
be the following:
Pin arrangment:
4 5
3
1 2
Pin 1: +5 VDC
Pin 2: Shield, tied to ground inside the A500
Pin 3: -12 VDC
Pin 4: +12 VDC
Pin 5: Ground
You could use a multimeter to check.
"You do trust me, don't you? Of course you do."
— "To Play the King" (BBC)
I can verify that the 1200 uses the same pinouts as the 500…
Bob Comer — Hakuna Matata!
Cheyenne, Wyoming U.S.A.
Bob,
I suspected as much. C= wouldn't have any reason to switch to a
different power supply as the requirements, voltage-wise, would be the
same. I know I wouldn't, economically speaking.
"You do trust me, don't you? Of course you do."
— "To Play the King" (BBC)
Thanks for that – I take it that arrangement is looking at the plug from
the power supply in the orientation it actually plugs in.
Lu, 2500/30 APpy downunder
Lu,
I always try to present such information in the exact manner in which
you would see it, if you were trying to plug the connector into your eye.
Simpler that way, neh?
"You do trust me, don't you? Of course you do."
— "To Play the King" (BBC)
Are you using CPU FASTROM? This is the only problem I have had with my
Microbotics M1230XA expansion board, and the only program that I noticed
it causing problems for is Lemmings II. When I tried to run this with
FASTROM, it gave a scrambled display. If I remember correctly I could not
exit from this most of the time. By fiddling with the things in
User-Startup I found that commenting out the line "CPU FASTROM" fixed the
problem. Maybe you're also running some program that doesn't like having
this software remapped.
Clay Spence
Thanks, I'll look into CPU FASTROM but I think it may be unlikely as the
problem occurs regardless what I boot from … what does CPU FASTROM do?
Lu, 2500/30 APpy downunder
> … what does CPU FASTROM do?
Well, I thought it used the MMU to remap the libraries in the Kickstart
ROMs into fast RAM, which should speed up system calls somewhat. I can't
find the reference that said this, though. The AmigaDOS manual says it
"copies data from ROM into 32-bit RAM, making access to this data
significantly faster", which sounds somewhat different.
Clay
Clay,
The CPU FASTROM copies the KickStart image from the ROM(s) into a 32-bit
RAM area, then uses (requires) the MMU to do 'remapping' so the systen
accesses the KS RAM image instead of the KS ROM when making calls & jumps.
This increases the speed of OS access/use *IF* the ROM was only 16-bit
(used in A1000/500/2000) while the RAM is 32-bit, or if the 32-bit RAM is
running significantly faster than the 32-bit ROMs (used in the
A3000/1200/4000).
Ron. – ("I'm not stupid, I'm not expendable, and I'm not going!"
– quote made by any low level RPG player.)
Lu,
CPU FastROM copies the contents of the Kickstart ROM to your fast ram
with the intended purpose of speeding up accesses to ROM library
functions, of which the Amiga has quite a few. Some programs, and some
hardware, balks at this manipulation of the hardware memory space since
they ignore the MMU built-in to many machines. Instead, they usually go
straight to a hardware address. Not nice, not compatible, and not
authorized by C=, such as they are these days.
"You do trust me, don't you? Of course you do."
— "To Play the King" (BBC)
> CPU FastROM copies the contents of the Kickstart ROM to your fast ram
> with the intended purpose of speeding up accesses to ROM library
Time to delve into the startup sequences by the looks of it, thanks.
Lu, 2500/30 APpy downunder