#SetPatch Crash/A3000 3.1
50 messages in this thread
Hi, everyone!
I'm looking for as much help as I can on a weird problem I've been
experiencing since I installed 3.1 on my "vanilla" A3000/25 last week. You
may have seen this on AmigaUser, but I thought I might get a few more
responses here…
Upon booting, the system crashes with a "Software Failure Error:8000 0004
Task:07C0DEB0" flag. The only way I can get the machine to boot
successfully is to comment-out the "C:SetPatch QUIET" line in the
Startup-Sequence. All SetPatches I have after V37.10 (V37.38, V40.14,
V40.16) crash the system, while V1.38 (WB1.3) and V37.10 (the old 2.04
SetPatch) work fine.
Prior to the upgrade, I had been using 2.04 KS on disk, with bootstrap
ROMs. After the upgrade, I have the 3.1 ROMs and WB from Expert Services.
My A3000 is generic, with the only change made (that I am aware of — I
bought it used almost 3 years ago) is the addition of 1 Meg Chip Ram
(total 2M) and 4 Meg Fast Ram (in 1×4 ZIPs) that was installed before I
got the machine. 3.1 without setpatch runs fine.
Here is the System Configuration info:
Processor: CPU 68030/68882fpu/68030mmu InstCache:ON DataCache:ON
WriteAllocate:ON BurstModes:OFF FreezeModes:OFF CopyBack:OFF
Versions: Kickstart V40.68, Exec V40.10, Disk V40.42
Ramsey: Normal, Code $0d, PageDetect:OFF BurstMode:ON WrapMode:OFF
Gary: Normal, Code $00 Timeout:ON Type:BERR
Battmem Status: Timeout:2 Sec LUNs:Only 0 SyncXfer:OFF HostID:Unit6
Amnesias: Originally "LOST", but I used SCRAM to "revive" them before
installation.
Any ideas as to why the SetPatches crash? Maybe a hardware jumper? I can't
figure it out… I'm not sure what the newer SetPatches are trying to do.
Hope anyone can help… I seem to be running fine without the SetPatch so
far, but I'd like to be safe. Thanks!
— Dave
Dave,
Your opening remarks suggested you have a softkick machine. If so, I
think you need a softkick version of 3.1. It seems to me that I remember they
mentioned in the packing stuff that came with mine and listed available
versions, that a softkick version was available. You might want to check on
that.
As for SetPatch, you should use the version that comes with 3.1, which
is 40.16. Be sure you don't have more than one setpatch in the c: directory
and that it's the right version. Of course, the only other thing to suggest is
to be sure you installed the ROMs in the correct sockets. I think ROM 0 is in
the bottom socket and ROM 1 is in the top, but I don't know if that's correct
since I can't see them in my machine right now to know for sure.
Steven:
Yes, I think I have a softkick machine, as it was originally configured
that way. Kickstart was actually a file in the Devs: directory.
I did not get packing material with my 3.1 upgrade listing versions. I was
only aware of an A3000 version, as opposed to an A2000 or A4000 version. I
will check with Expert Services on that.
Right now I am running 3.1 fine (without any SetPatches), so I tend to
think the ROMS are installed properly. At least, the system thinks they
are!
My machine has 2 sets of ROM sockets. One was where the old ROMs were, the
other is right beside it and are just pads (no sockets). Both sets are
labled ROM0 and ROM1, so I just placed the new ROMS where the old ones
were.
— Dave
Dave,
I think you may want to talk with Expert Services. I do seem to
remember that when I ordered mine they asked me if I had a hard or softkick
3000 or 3000T. This may be unimportant, but it's worth checking out. I have a
3000T and use SetPatch 40.16 without any problems, so it would seem that if you
are having a problem, it may very well be the ROMs you have are not the right
set for your machine, and that could easily explain the problem you're having.
When you say you didn't get packing material, did you not get a box
with a black and red cover on it with boxes to be checked for machine
configuration? Did you buy your upgrade from Expert Services or somewhere
else? Expert put a notice on the services that there were people distributing
a 3.1 upgrade, but that it was not an authorized upgrade. I do hope you didn't
get yours from another supplier.
Steven:
I got my upgrade from Expert Services. By "packing material" I thought you
meant a reciept or list. On the side of the slip cover of the box, there
is a red dot on "AS330 A3000, 3000T. That would appear to be the right
version.
I will try to get in touch with Expert Services and see what they have to
say… but the longer I run without a problem SANS SetPatch, the more I
tend to think I'll just leave it alone! <g>
— Dave
Dave,
I can confirm that 3.1 will work on previously soft kicked 3000, it does on
mine, Setpatch and all.
However, some "non standard" batmem settings that worked under 2.x will give
3.1 real grief. Having SCSI set to synchronous xfer will normally cause a
complete boot failure, for example.
The mother board on my 3000 is identical to yours, complete with empty ROM
socket positions. The ROM 0 & ROM 1 legends are reversed on my board, the
socket numbers are as per the intallation docs tho`. I can`t see your machine
booting at all if you did have ROMs reversed. It almost sounds like setpatch is
trying to patch where your ROMs aint, if that is at all posible.
The only problem I have with 3.1 is, that when moving larger icons around the
WB screen, the icon will flicker or disapear until the pointer stops moving.
This only occurs while running the ROM image from RAM (i.e cpu fastrom).
Robert.
Robert, the phrase "previously soft kicked" is the key here. The new
ROMs, mine at least, and I was not aware of there being more than one
choice that was 3000 specific (maybe for the 3000T). My original ROMs
used the soft kickstart. My 3.1 ROMs do not. I deleted the Kickstart
file in the DEVS: directory, which was taking up half-meg of real estate
for nothing.
DJ
On AP from Queens, the thrill and excitement of the Big Apple
Where a little paranoia is always a healthy attitude …
Robert:
Could you check your system for me via a utility and tell me what your
CPU, Gary, Ramsey, and BattMem settings are? Maybe one of my settings is
off. Also, could you tell me general info about your system (amount of
memory, page or Static column ram, type of Ram chips, etc. — do you have
any newer revisions of chips (ie Buster). Thanks.
BTW, my Buster is the original, appears to be rev. 7.
— Dave
Dave,
Your mailbox should now contain an IFF screen dump of various Scrammer windows.
View with your newly acquired Multiview 🙂
My 3000 sounds to be the same as yours, along with a busted Buster, dodgy DMAC
and a ropey Ramsey;) No chance of replacing these over here, C= UK were/are
just box shifters. Spares and knowledgable techs are very hard to find here.
Robert.
Robert: I will send you an IFF screen dump of my system. Our systems seem
to be virtually identical, except that you have SCRAM. I have sinced
changed the following:
1) Set SCSI Timeout back to 128ms (it works, but didn't help the SetPatch)
2) Killed my Amnesia bits (didn't help the SetPatch) I revived them again.
I issued a command from the shell "CPU BURST" and my machine behaved VERY
similarly to the crash from SetPatch! The Power light blinked about 7
times, and the Software Failure banner came up, with an error 8000 0006
instead of 8000 0004. We may be on to something. I think I'll check my RAM
chips and see what they are.. they may be SCRAM/Page mixed…
Thanks again!
BTW, you have a PAL Amiga… where are you from? I have a good friend in
South Carolina named Bob (Robert) Luckett… I'm sure you're not him! <g>
— Dave
Dave,
I think that RAM or possibly Ramsey is your problem. One of Setpatch`s
functions is to turn on Instruction Burst. The error codes are non recoverable
CPU traps for illegal instruction and CHK instruction. Starting to look
promising eh?
Have you tried Scrammer`s memory testing facilities? It should be able to
reveal the RAM type and test each bank, I don`t know how thourgh the test is
tho`. Any ideas what these amnesia bits are for? Don`t think they have any
relevance here, just curious.
I`m in the UK, so I`m fairly sure I`m not the same Bob Luckett<g>. Not an
amazingly common name tho`, so he might just be a long lost relative!
Rob.
Rob:
I agree that memory is the most likely culprit. I noticed while I
installed the 3.1 Roms that the Zip chips are not all the same (some are
TI, some NEC, etc) so maybe this could cause the problem.
Another possibility is that SetPatch might expect to see SCRAM-type memory
if you have 1×4 ZIPs, whereas to the best of my knowledge, I have PAGE
mode RAMs. If you attempt to enable CPU BURST with Page-mode RAMs, the
machine will crash, as the Page RAMs cannot handle the burst mode. I have
tried Scrammer's memory-tests, and, true to its documentation, any attempt
to test Bank-1 (other than a write-test) results in a machine crash. In
fact, I had to cold-boot after I tried it, because a warm-boot resulted in
my seeing "0 Fast Mem" in the titlebar, and I subsequently paniked. <g>
I don't know the signifigance of the amnesia bits, other than the LOST
flag indicates that power was interupted to the battmem chip at some
point, and its settings may have been lost.
I will take my machine apart during my upcoming vacation (between
installing plank paneling in the kitchen and going on leaf-peeping trips
<g>), and I will check all the chips (especially the RAMs and their type
to see of that helps things out. Thanks for your help!
BTW, concerning Bob Luckett, he is a good friend who I worked with in
local television here for a few years. He has gone out of broadcasting,
and is now an Occupational Therapist, which I think is rather fitting,
since he managed to survive TV. I live in Connecticut, and Bob now lives
in S. Carolina. Maybe you are related! <g>
I once was traveling through Virginia (I have relatives there), and on my
way through northern Virginia I passed through a small town named
Lucketts. There I saw Lucketts Elementary School, Lucketts Fire Dept,
Lucketts Lions Club, etc. I took several pictures of the town for Bob —
he got a kick out of it! If I had a scanner I'd send a copy or two off to
you via Email.. you would probably find it amusing as well!
Later!
— Dave
Dave:
This is getting interesting! Have you tried running SetPatch 40.16 with
the NoCache switch? If it is a burst-mode problem, you could use
"SetPatch NoCache" followed by "CPU Cache NoBurst" in your
Startup-Sequence.
Wayne
Wayne:
*********** BINGO!!! ****************
Check this out:
3.System3.1:> version c:setpatch
setpatch 40.16
3.System3.1:> setpatch nocache
Amiga Operating System
Copyright 1985-1993 Commodore-Amiga, Inc.
All Rights Reserved.
SetPatch (V40.16) has been installed. Patch list:
Enabled Advanced Graphics Modes
graphics/WriteChunkyPixel() patched
Memory Pools
3.System3.1:> cpu cache
System: 68030 68882 (INST: Cache NoBurst) (DATA: Cache NoBurst)
3.System3.1:>
Well, we've got the culprit nailed down. SetPatch is trying to enable
BURST mode on the CPU, and the memory chips can't handle it! This means
that I really have to inspect that RAM. I will do it this weekend.
Thanks for mentioning the SETPATCH NoCache switch…. I didn't think of it
(actually, that must be a new switch, as I was not aware of it).
I will let you know what I find "under the hood"
Thanks a LOT <you win what's behind door #3…>!
— Dave
Dave, as far as I know there is no soft-kick 3.1. I have a 3000 with 3.1
and was not given an option. The fact that you are booting at all tells
me the ROMs are installed properly (the ROM-0 chip goes in the ROM-1
socket and viceversa).
It's easy to fine out if you're soft-kickstarting: do you get a "Loading
Kickstart X.XX" while cold-booting? If you cold-boot and hold both mouse
buttons down, do you get a choice of booting OS revisions from floppy or
harddrive? If the answers are yes, then you have a soft kickstart system.
Also, how many disks are in your 3.1 set? Is one a Kickstart disk? If
so, you would have seven disks instead of six.
Another thing you may want to do is save the original startup-sequence
then add echo statements throughout it to find out exactly where in the
startup it's hanging with the proper 3.1 setpatch on.
DJ
On AP from Queens, the thrill and excitement of the Big Apple
Where a little paranoia is always a healthy attitude …
DJ:
I'll try the startup-sequence echo idea… thanks.
My system USED to (under WB2.04) come up with "Loading System 2.x"
message, but now that I have installed 3.1, it is just a black screen
until the CLI window opens with a copyright message like "Commodore Amiga
Operating System and Libraries C1991-1993 CBM", etc. I have six disks in
the 3.1 set.
— Dave
In other words, no soft kickstart. You can save 1/2 meg in your system
partition by getting rid of the kickstart file in DEVS: Save it to a
floppy if you're not sure <g>.
DJ
On AP from Queens, the thrill and excitement of the Big Apple
Where a little paranoia is always a healthy attitude …
I'm starting to wonder if you can you use the ROM's on a soft-kicked A3000?
Wayne:
>>>I'm starting to wonder if you can use the ROM's on a soft-kicked A3000?
How can we find out? Last time I checked, West Chester rings "The number
you have reached is not in service…" :^(
I think I'll have to tear the machine apart again and check ALL jumpers.
Maybe the SetPatch thinks I have SCRAM or something. If it enables burst
mode on the CPU, I would think that would crash the machine. Perhaps a
jumper needs to be adjusted telling the machine I have KS in ROM. Who
knows? At this point, I'm reaching.
But, like I said, since the machine is humming along even as we speak with
3.1 and NO SetPatches, I may just leave it alone.
Either that, or try to find a newer A3000 motherboard. <good luck>
— Dave
The ROMs that came with MY 3.1 for the 3000 are NOT soft kickstart ROMs.
They replace the original 3000's soft kickstart ROMs, so now you no longer
have a soft kickstart environment. Read my message to Dave to find out
if you have a soft kickstart.
DJ
On AP from Queens, the thrill and excitement of the Big Apple
Where a little paranoia is always a healthy attitude …
My 3000 was originally soft kicked. IT was a temporary measure by C= till
2.0 stabilized, which allowed them to ship product, without having to make
new roms at each revision. When 2.04 came out, the Roms that came with
it, replaced the soft kick roms, and the select at boot time capability
between 2.0 and 1.3 was lost.
After that, unless you left the 2.0, and 1.3 images on your hard drive,
there was no difference, between the earlier 3000's and the later 3000's
as far as rom handling was concerned. Personally I would have like the
soft kick to have remained, that was a simple disk update would have been
available.
I'm looking at your message and a couple of responses to it regarding soft
kicked 3000's at the same time. Once you've put in the roms, you're
machine is no longer soft kicked. Soft kicked 3000's actually do have roms
that need to be yanked out prior to insertion of new roms. The new roms do
not fit in the sockets of the soft kicked roms but in adjacent sockets.
Some older soft kicked 3000 mother boards require an add on tower to
install new roms.
I'm not sure that any of this has pertinence to your situation, but I'm
pretty sure there would be problems if two sets of roms were present or
the new roms were installed in the wrong sockets.
-Richard
Richard:
Thanks for your help…
We might be on to something. Here is what I found when I replaced the
ROMS:
The ROMs originally in the machine… Labeled "390629-02 ROM0v5 C1990
Commodore-Amiga 9042D" Were sitting in sockets near the mouse port.
Adjacent to them were two "pads" labeled ROM0 and ROM1, with no sockets.
The original ROMs were in sockets labeled U180 and U181, also ROM0 and
ROM1 while the pads are U182 and U183. I pulled the old ROMs and installed
the new ones per instructions in the sockets.
There is a jumper near them, but the schematic says it is an "A2000 ROM
compatibility jumper" so I left it alone.
The odd thing is that about a year ago (with softkick WB 2.04) I
downloaded SetPatch v37.38 from CIS, and tried to replace my v37.10, and
the same thing happened (the system would not boot). I stored the old
setpatch away, only to dust it off and try it again now… SAME result as
v40.16.
I think there might be something else going on here besides ROM chips /
softkick. There might be a memory select problem, or a CPU thing.
What would you think about running the system with NO setpatch at all? It
seems to work great without it! <g>
— Dave
Dave,
Here are my TOP TEN things that you can do about your set-patch problem.
10. Buy a Tandy Computer
9. Give M. Ali a call and ask for his advice.
8. Solicit help from Jay Miner via a Oijii board.
7. Upgrade your computer to work with your OS upgrade.
6. If your not in a hurry, ask someone at Commodore UK for help.
5. If your really not in a hurry ask someone at Utilities Unlimited.
4. Ask Expert Services for your money back.
3. Ask Commodore Ltd. for your money back.
2. Ask a hypnotist to erase the term "set-patch" from your mind.
1. If it fixed, don't break it!
<VBG> Marty
"When the eagles are silent the parrots begin to jabber"
-Sir Winston Churchill
Dave, your 3000 sounds like most 3000s. There is an empty space where it
looks like two more ROM sockets might go, but are not installed. I think
those might have been for network booting, I'm not sure, but they are
perfectly normal as they are (empty). The 3.1 ROMs go in the old ROM's
sockets. They are obviously in the right place, or you wouldn't boot,
with or without setpatch :-).
SetPatches are meant for specific OS releases. NEVER run a setpatch from
a different OS release, or your system will shrivel up and turn into a
stinky puddle. OK, I exagerate but you get the drift <g>. It is best to
find out why the patch is making the system crash, because to me that
indicates something somewhere ain't kosher. You might want to comment out
the QUIET from the setpatch command to see what it's doing, and put an ASK
command right after it to stop until you can read the messages. You will
also find out if the SetPatch is crashing or if it's something else
further down the startup sequence.
DJ
On AP from Queens, the thrill and excitement of the Big Apple
Where a little paranoia is always a healthy attitude …
DJ: Thanks for the tips… I will try checking out the SS with 'debug'
lines… <g> and let you know what happens.
— Dave
DJ:
I just tried your suggestion about echo statements in the SS. I got
NOTHING before the System failure!
I then had the system "kick out" of the startup-sequence by putting in an
invalid echo line at the beginning. I then ra"SetPatch", and the system
turned to an oozing goo, pausing only to blank the screen. No output at
all from SetPatch.
This is very confusing. I ran AIBB to check the performance of the system,
and it came though with flying colors WITHOUT setpatch. In fact, in some
areas (writepixel for instance) it is TWICE the speed of 2.04! And this is
with the FASTROM option off.
I will talk with Expert Services next and see what they have to say.
Thanks for your help!
— Dave
Keep us posted! This is a weird one. Did you try running setpatch
without the quiet? Also try running "version c:setpatch" after a
successful boot without it. Mine says 40.16, and it's 13,484 bytes long.
The date is July 6, 1994.
DJ
On AP from Queens, the thrill and excitement of the Big Apple
Where a little paranoia is always a healthy attitude …
DJ:
I tried running SetPatch without the quiet, but got no readout.. the
screen just stays (or goes to) black before the Software Failure banner
comes on.
The Info for my SetPatch is the same as yours (13,484 bytes, v40.16,
7/6/94)
I'll keep you posted.
— Dave
Dave,
in your case, I would suspect one of two things. First, and foremost,
the Setpatch is corrupt, but I believe you've reloaded it a few times. If
so, you have a hardware conflict somewhere in your system, and a bloody
subtle one as well. You might want to take apart your system and make
absolutely sure that every socketed chip is firmly in its socket. Usually
that will take care of the problem. If it does not, then I believe you
have a chip failure. I'd then try to find someone that could troubleshoot
it.
You did remove all boards from the system when you conducted your
experiments? If not, try that first. If so, you are in deep kim-chi.
Brian:
The likelihood of a corrupt setpatch file is practically nil. I used
different copies of SetPatch (from the install disk, from the WB disk,
etc) with the same results.
I have no boards on my system. I will be taking the machine apart and
double checking everything, and making sure all chips are seated. I also
noticed while I installed the 3.1 ROMS that my memory chips are not all
the same (some are TI, and some NEC for example) so that may be the
problem.
— Dave
DJ:
The Command "CPU BURST" from a shell produces almost exactly the same
crash as the SetPatch, the only difference being the error reported (8000
0006 vs. 8000 0004 for SetPatch & the task #) — the power light blinking
and behavior is the same!!!
I think we may be on to something…. time to check the RAM (and make the
donuts) <g>
— Dave
I've had RAM problems with my 3000, none of them permanent, but they did
require me to "fiddle around" with ALL the chips. I hate those zips! I
was getting "out of memory" when booting, or booting and getting 0 fast
ram available (!), and also just plain crashing. So don't give up. The
last time took about three hours of jiggling, put everything back
together, crash, taking the backplane board & disk chassis off again, etc.
… you know the drill! I left no chip untouched (careful with that
static!) and eventually was rewarded.
DJ
On AP from Queens, the thrill and excitement of the Big Apple
Where a little paranoia is always a healthy attitude …
DJ:
I'll be checking those Rams as soon as I get a chance (probably this
weekend or early next week). As I mentioned to a few other people, I
remember looking at the Rams and seeing different manufacturers (TI, NEC,
etc) for some chips. I'll have to check the numbers on them and their
seating. I may have mixed page & burst chips!
If SetPatch attempts to set CPU Burst ON, that could be the problem… do
you have SCRAM? I have Page Ram, to the best of my knowledge… and CPU
Burst will crash the machine. Maybe all SCRAM what I need… oh, boy! $25
a chip! :^(
Later!
— Dave
Dave, I did get SCRAMs when I upgraded my memory several years ago, but I
find it hard to believe they would by default use a setting that would
crash half the CPU population or more! Wait a bit before you shell out
for new chips.
Different manufacturers should not make a difference, unless the quality
of one is so bad that the chips are defective 🙂
DJ
On AP from Queens, the thrill and excitement of the Big Apple
Where a little paranoia is always a healthy attitude …
DJ:
Thanks to Wayne, the culprit is nailed down. Look at this:
3.System3.1:> version c:setpatch
setpatch 40.16
3.System3.1:> setpatch nocache
Amiga Operating System
Copyright 1985-1993 Commodore-Amiga, Inc.
All Rights Reserved.
SetPatch (V40.16) has been installed. Patch list:
Enabled Advanced Graphics Modes
graphics/WriteChunkyPixel() patched
Memory Pools
3.System3.1:> cpu cache
System: 68030 68882 (INST: Cache NoBurst) (DATA: Cache NoBurst)
3.System3.1:>
I ran SetPatch NoCache and it works! This means that SetPatch IS trying to
enable burst mode, causing the machine to crash. It is the only answer.
What is your Patch list on your machine?
— Dave
Then the morons (presumably at VillageTronic) who mastered the disk
changed all the dates from the dates they were really created on to the
date they mastered.
Oh, well. At least this should only impact the historians … Anyway,
I'm just grateful to them to be able to have 3.1, so I can be at the same
OS level as my Windows buddies at work 🙂
DJ
On AP from Queens, the thrill and excitement of the Big Apple
Where a little paranoia is always a healthy attitude …
Dave,
Excuse the delay in responding, but you inspired me. I went ahead and
installed thge 3.1 ROMS I've had sitting here for the last month. Now I
know why I was delaying it. These 3000's are sure great but their a pita
to work on. Not only did I have to put up with the hassle of disassembling
my machine, I also had to sweat out whether or not I was going to break
pins I had to straighten out on the ROM chips. Expert sent me some badly
mangled chips. Looks like they were quality checked by a kindergartener or
something.
Guess what. No problems here with 3.1. I had to set a 2 sec timeout on my
boot drive, which worked perfectly before, to get it recognized at bootup
but that's the only thing I ran into. My system now boots a little slower
than before. The step from 1.x to 2.x is sure much greater than from 2.x
to 3.x. I'm not altogether convinced it was worth the money and hassle.
Your machine was definitely hardkicked. I'm not sur where your problem is
coming from. Sound like it is some kind of memory thing. What happens when
you run setpatch after WorkBench has loaded? If it works without SetPatch,
it works and that's all that's important. You're not going to damage
anything.
-Richard
Richard:
Yes, I noticed that the difference between 2.x and 3.1 is almost subtle,
with the exception of the graphics being drawn much more quickly.
I will be checking the machine out thoroughly when I get some time (during
my vacation — between installing plank paneling in the kitchen and leaf
peeping) <g> and I'll let eveyone know what I find.
At this point I would suspect you are right.. it is a memory thing. I
noticed while I had the cover off the last time that the memory chips are
different (ie some ar TI and some are NEC, etc) so that might be the
problem.
— Dave
Dave,
Having ram chips from different companies shouldn't matter too much. I've
got a real mixture. I also have a one page mode chip and the rest scram.
If you have a mixture of page and scram that could be a problem. The page
mode should go in the first socket. This makes the machine treat all the
ram as page mode which works O.K. with scram. However, if the machine
treats the ram like scram when page mode is present, there could be some
problems.
-Richard
Richard:
Wayne advised me to run SetPatch with NoCache. This is what I got:
3.System3.1:> version c:setpatch
setpatch 40.16
3.System3.1:> setpatch nocache
Amiga Operating System
Copyright 1985-1993 Commodore-Amiga, Inc.
All Rights Reserved.
SetPatch (V40.16) has been installed. Patch list:
Enabled Advanced Graphics Modes
graphics/WriteChunkyPixel() patched
Memory Pools
3.System3.1:> cpu cache
System: 68030 68882 (INST: Cache NoBurst) (DATA: Cache NoBurst)
3.System3.1:>
So, SetPatch is apparently attempting to enable burst mode on my machine.
The only unanswered question is WHY…. which I will find out this weekend
when I check the Ram chips… I'll let you know what happens.
Thanks!
— Dave
Dave,
As you may have noticed on the box, there are different versions of 3.1
for the A2000, A3000 and A4000 computers. Aside from a handwritten check
in the block on the box, I'm not sure how to tell which version you have.
Is it possible that you have the ROM for the wrong machine? Perhaps
someone knows by part numbers….
wmc – via Autopilot!
Wayne: The box says it is the A3000 package, and the ROMs have stickers
that say they are the A3000 Roms. There might be a possibility that they
put the wrong stickers on the Roms… maybe we could tell the difference
via KickStart versions? I noticed via forum traffic that each machine
model has a different version number for kickstart. Mine is v40.68.
— Dave
Dave,
I have the single ROM A2000 version so hopefully someone else will chime
in with a way to discriminate between the A3000 and A4000 versions.
wmc – via Autopilot!
Dave,
I to have upgraded a plain A3000-25 recently to OS3.1. I have 2meg.chip
and 4meg.fast static column ram. First of all, were the 1×4 zip chips you
installed all static column ram or all page mode or a mix of both? If page
mode is mixed with static mode chips or if page mode were installed the
system can't use the static mode features. I forget what those are but it
was worth putting in static mode chips for me. Next, did you set jumper
J852 on the mother board to indicate to the system what type of ram you
were installing? The shorting bar should connect pins 1 and 2 for 1Mx4
DRams whether static or page mode.
For the OS3.1 roms. I assume that, like me, your machine had a ROM tower
for the softkick ROMs. If so, did you put the new ROM's in the tower and
re-install the tower back in the computer. It's very easy to get the tower
reversed. I installed the following: new ROM 3000-1 in tower rom-socket
U180 and new ROM 3000-0 in tower rom-socket U181. I put the tower with the
new ROMs back in the same place from which it came. Make sure that the
tower pins get into the sockets alright. My tower sits slightly askew when
plugged in correctly. I then installed the system software with the
install disk.
I had to set BattMem to 2sec SCSI timeout but I DID NOT revive the
Amnesias like you did. I have the SyncXfer set to ON only because both of
my hard drives support SyncXfer and it's faster than ASyncXfer.
Other than a few differences our system's are the same, therefore, some
other hardware problem directly affecting ver40.16 of SetPatch is at work
here. Good Luck.
/ / Flying The A3000 on AutoPilot 1.74
\/\/ from Durham,Ontario.
Pete: Thanks for the reply.
I did not install the Ram chips, so I do not know what type each
individual chip is. I bought the computer used about 3 years ago, and it
already had the 4M fast ram installed. The system recognizes the RAM, so
I'm pretty certain the jumper is set correctly, although when I get a
chance I will look at it.
For the ROMs, I don't know what you mean by a "Rom Tower". My machine has
sockets for the Roms, but they are not removable (at least, not without
de-soldering them). I followed instructions and but Rom-0 in U181 and
Rom-1 in U180. They seem to be working fine.
Insteresting that your Amnesisas were also lost! Perhaps this is a
difference looking into. I may try to "kill" them.
Seems my drive will work with Sync or ASync transfers. It didn't seem to
matter which was selected.
I guess I'll have to look at the board carefully and see if I can find any
odd jumpers, Ram chips, etc. I'll let you know what happens.
Thanks again!
— Dave
Dear Dave,
I can't claim to have as much computer acumen as you and the rest of
the computer experts out there, but I have had nothing but trouble with my
A3000T. In my atempt to take advantage of the newest CMB BridgeBoard, I
purchased a second hard drive. To make a long story short, after exhausting all
my efforts to solve the problem of crashing, and crashing and crashing, I
resorted to, not one but two computer repair companies which are authorized by
CBM. The first company, after working with my A3000T for 4 months, gave up in
disgust, and to my surprise, this company went out of business. The second
company discovered that by replacing the motherboard, the problem should be
solved. Well, the problem is solved, until you run the machine for aprx 45
minutes. Then, a crashing we will go!
And yes Dave, I shall try, yet another ploy to this heady problem. I'm going to
try upgrading from os 2.1 to os 3.1. Oh, but the first 3.1 ROM chips I received
were defective. So, I wait for the post to deliver, yet another hope, sprung
ever eternal, to solve the: "The A3000T that could", or "Have some of us been
taken for a ride on the good ship Lollipop?", or "Was CMB ever my neighbor, Mr.
Rogers?"
Bob
Bob:
My goodness! You've been through quite a bit with your A3000T.
Unfortunately, I would not count myself among the "computer experts" here;
I'm mainly a "knowledgeable user"!
Back to your problem, I have never installed or worked with a BridgeBoard.
However, based on what you described (crashing machine after aprx 45
minutes) I would tend to think you may have a ventilation problem, and the
machine is overheating. If the BridgeBoard is a 486 model, then I would be
even more suspect of the overheating problem (the 486 is a notoriously hot
chip) and if your A3000T is an 040 model or if you have an accellerator,
that makes the overheating even more likely. It could also be a conflict
with the OS (but I find that unlikely), as I have found 3 of my games on
my HD will not run under 3.1!
That's the only thing I can think of off the top of my head; maybe others
here might chime in with some more help for you…
Good Luck!
— Dave