• Please review our updated Terms and Rules here

Ast six oak premium missing IC? 5150 with 16-64 board

offensive_Jerk

Veteran Member
Joined
Jul 13, 2009
Messages
1,226
Location
Wisconsin
I've got a "new" IBM 5150 with the 16-64 board and 10/27/82 bios.

It came with this SixPakPremium card, but no matter what I try, I get 201 errors. 1090/1095, 201. I pulled the card out with no switch changes and I got the 1090 201 error.

I set the IBM to 64k and took the card out, and no errors.

I set the AST card to switch 2-7 and 2-8 to Off, Off, (640k)but still getting errors. Starting memory on the AST set to 64k.


The last thing Ive tried is leaving only bank 0 on the card populated and setting the AST as "up to 256" conventional memory and get below error. I swapped the banks with different memory, etc.
0400 201.

I am setting the 5150 dip switches according to the correct 1982 bios on minuszerodegrees chart.

When I look at the picture of the card on the manual there is an IC missing (u58) compared to my card. And I'm not talking about the gameport missing ICs, u66 and 72
 

Attachments

  • Screenshot_20231008-164808.png
    Screenshot_20231008-164808.png
    3.2 MB · Views: 13
AST SixPakPremium: 4 banks of 256 KB = 1024 KB.
So, for 640 KB in the machine, only 576 KB of that is to be enabled.
64 on motherboard + 576 on SixPakPremium = 640 KB

My take on switch settings:

Motherboard SW1: Switches 3 and 4 set to OFF
Motherboard SW2: Switches 1 through 5 set to ON-OFF-ON-ON-OFF, i.e. 10/27/82 BIOS, 640K total in computer
SixPakPremium SW1: Switches 1 through 4 set to ON-OFF-OFF-OFF, i.e. start card's RAM at address 64K
SixPakPremium SW2: Switches 7 and 8 set to OFF ("Up to 640 KB")

Re the SixPakPremium SW2 settings above. From my reading of the AST SixPakPremium manual, the SixPakPremium will work out from its SW1 and SW2 that it needs to provide 576 KB.

Are the aforementioned switch settings what you have set?
 
Just to reiterate:

IBM 5150 16-64 with 1982 bios, POSTs fine without the card and switches set to 64k.

With the AST fully populated with 1MB (all 4 banks)

I am assuming what needs to happen is I set the 'starting memory' on the board to 64k, then set the AST conventional memory to 'Up to 640 KB' sw 2-7 and 2-8 to OFF OFF.

Then, I am guessing I need to set the IBM SW2 to 10110. (640 k conventional memory)

What I believe should happen is the AST would use banks (256 each) 0-3 to backfill conventional memory to 576k (minus the on-board 64k). Since the three banks are used (768k) it would have 704k (minus the on board 64k) and have 192k available as extended memory.
 
So, RAM chips re-seated and motherboard/SixPakPremium switch settings verified.

Errors of 1090 201 and 1095 201.

Per the 'Address Block Indicator' section of [here], the '10' portion is an error at address 64 KB, which corresponds to the first RAM bank on the SixPakPremium.

90 and 95 is the bit error pattern, per [here].
90 hex = 10010000
95 hex = 10010101

I pulled the card out with no switch changes and I got the 1090 201 error.
Note that the 90 is not unexpected for non-existent RAM in an IBM 5150. Per the 'Multiple bit failure' section of [here], I see 98.

So a question becomes, is the RAM functionality on the card completely faulty, or partially faulty.

Are you in a position to see what Ruud's Diagnostic ROM makes of the situation ?
 
AST SixPakPremium: 4 banks of 256 KB = 1024 KB.
So, for 640 KB in the machine, only 576 KB of that is to be enabled.
64 on motherboard + 576 on SixPakPremium = 640 KB

My take on switch settings:

Motherboard SW1: Switches 3 and 4 set to OFF
Motherboard SW2: Switches 1 through 5 set to ON-OFF-ON-ON-OFF, i.e. 10/27/82 BIOS, 640K total in computer
SixPakPremium SW1: Switches 1 through 4 set to ON-OFF-OFF-OFF, i.e. start card's RAM at address 64K
SixPakPremium SW2: Switches 7 and 8 set to OFF ("Up to 640 KB")

Re the SixPakPremium SW2 settings above. From my reading of the AST SixPakPremium manual, the SixPakPremium will work out from its SW1 and SW2 that it needs to provide 576 KB.

Are the aforementioned switch settings what you have set?
Yes, I just confirmed.

You are confusing extended memory with expanded memory.
The SixPakPremium optionally provides expanded memory.
Ugh, Yes I am. I always mix those up.
Are you in a position to see what Ruud's Diagnostic ROM makes of the situation ?

Sadly, I don't have a way to burn ROMs unless I can do it through an XT-IDE somehow.

I am tempted to throw this in a different 5150 64-256k board. The manual really wants the default to be 256k starting address. I wonder if there is some sort of compatibility issue with a 16-64 board using the '82 bios?
I will say that the error seems to have slightly changes since swapping chips around.

At first I was not getting the 'parity error' message, and the system would seem to 'hang' usually with the drive light on. It tried to boot once, then looked like it reset (this was during initial testing.)

Currently, with the switches set as above, I get '1055 201' then it halts at PARITY CHECK 2
Removing the card displays '1090 201'

Also, I read the manual before looking over the switches, and I came to the same conclusion as the previous owner, except the base address was set for something other than default. I changed it back to default once I noticed problems. I don't *think* it would cause any conflict, because the only other two cards in the 5150 are the floppy controller and a small sized MDA clone.

I also tried a different slot.
 
Also, I read the manual before looking over the switches, and I came to the same conclusion as the previous owner
I will add that I purchased two IBM 5150s from this person, and noticed that the newer 5150 was incorrectly configured to not use 640k of memory. Maybe the person before him set it that way.
 
SW2-2 is not 'fully home' and therefore could be registering as OFF.

View attachment 1265658

I noticed that too, when troubleshooting. Flipping it back and forth gives a tactile 'click' and seems to work, it's just there's some tolerance in the plastic where the button droops a bit from gravity. It jiggles to a certain extent, but I feel like the metal contacts are still making connection. I did not confirm that though.

We have seen a lot of people confuse OFF and ON, which is why I created the web page at [here].
I'll confess I've been to that site before, when overthinking the switches. Thanks for that page, actually, it brought be back from insanity a time or two. Between overthinking the switches, and possible hardware issues, you start to question everything. And sometimes the manuals have it illustrated in black and white, and you start to wonder if the 'black space' means, that's where the switch goes for 'on', or does that signify the absence of the switch.
 
Sadly, I don't have a way to burn ROMs unless I can do it through an XT-IDE somehow.
I suggest that this capability be added to your to-do list.
Some EPROM/EEPROM options require an adapter. See [here].

I wonder if there is some sort of compatibility issue with a 16-64 board using the '82 bios?
Seems very unlikely to me.

I am tempted to throw this in a different 5150 64-256k board.
Worth a try.
 
Did some heavy editing in this post, I was too quick to post.

(edit) Not same error in a different 5150 64-256k machine. I changed the switches on the card to start at 256, sw1-4 Off, Off, On, Off:

"4055 201"

I did not adjust the settings on this 5150 since 4 banks are still filled equaling 256k.
 
Back
Top