• Please review our updated Terms and Rules here

Computpro RAM16 and IMSAI

8

8008guy

Veteran Member
Joined
Jun 25, 2014
Messages
584
Location
Salt Lake City, Utah
Hi,

Can anyone tell me if I can get a Compupro RAM16 working in an IMSAI? From the few links I can find it doesn't appear to be promising. I don't mind making modifications to the memory card. I just don't want to make mods to the IMSAI.

What 64k cards work well on the Altair/IMSAI buss?

len
 
I see it is a 16 bit SRAM of 64K bytes. The manual says it supports 8 bit read/write.
You might check the manual for things like jumpering unused addresses and the sXTRQ*.
Dwight
 
I've had both the RAM16 and the RAM17 working with my IMSAI and various other machines. One thing you'll run in to, is that the RAM16 is an IEEE-696 board and grounds pins 20 and 70. This will cause the one-shots in the front panel to be disabled, basically making the front panel act totally dead. I usually cut the traces to pins 20 and 70 on the front panel itself, since the "disable the one-shots" feature isn't used by any known IMSAI boards, and it does cause a fair bit of headache. Herb Johnson has a bit of information on this problem on his site:

http://www.retrotechnology.com/herbs_stuff/s100_pin20.html

If you don't want to modify the front panel, you can put tape over the pins -- a piece of cut down Kapton tape folded over pins 20 and 70 works nicely (they line up front-to-back). You can also cut them on the RAM board, as they're used as redundant grounds, but the IMSAI backplane, if it's an original, doesn't know anything about that and doesn't provide ground on those pins. Some IEEE-696 boards have jumpers to isolate pins 20 and 70, but I don't remember the RAM16 being one of those.

Personally, 64K static RAM boards using the 6116 are some of my favorite for using in *any* S-100 system. The CompuPro RAM16 and RAM17 are good cards (most CompuPro products are), as is the 64K card sold variously as Coex, Tanner, or Digital Research:

http://www.s100computers.com/Hardware Folder/Digital Research Computers/64K Static RAM/64K RAM.htm

Feel free to PM or email me if you keep having issues with your RAM16 -- I can get it set up for you on my IMSAI if you like, test it out, and send it back.
 
That makes sense. I did get a bit closer by pulling the bank select ic and grounding the bsel line. That let me see the board since the above a15 lines were floating. I could have flipped all the switches off as well on the bank select.

Anyway, doing that I could read, but not write.

I have no issue with cutting the traces on the ram board for lines 20 and 70. i will try that in the morning.

Thanks

Len
 
I have Ram-17s running on my Fulcrum I-8080 Clone. under CPM-80 and CPM-86 using the Compupro Dual Processor Board, and the IMSAI setup Compatibility Switches that the Ram-17 comes with.

I have not had Ram-16s running in the Fulcrum I-8080 Clone. The Ram-16 does not have IMSAI Compatibility Switches like the Ram-17 has.
 
From what I recall, the RAM-17 has the ability to work with the IMSAI Front Panel, if selected, as Microcomputer Solutions said. The RAM-17 also has switches to poke a hole in the memory map so one can deselect the RAM that may conflict with some other PROM in the system. The RAM-16 is meant to be a complete bank of 64K RAM and does not have either of those abilities. It has only switches to select which bank of 64K RAM it will be.

smp
 
glitch said:
I've had both the RAM16 and the RAM17 working with my IMSAI and various other machines. One thing you'll run in to, is that the RAM16 is an IEEE-696 board and grounds pins 20 and 70. This will cause the one-shots in the front panel to be disabled, basically making the front panel act totally dead. I usually cut the traces to pins 20 and 70 on the front panel itself, since the "disable the one-shots" feature isn't used by any known IMSAI boards, and it does cause a fair bit of headache. Herb Johnson has a bit of information on this problem on his site:

http://www.retrotechnology.com/herbs_stuff/s100_pin20.html

If you don't want to modify the front panel, you can put tape over the pins -- a piece of cut down Kapton tape folded over pins 20 and 70 works nicely (they line up front-to-back). You can also cut them on the RAM board, as they're used as redundant grounds, but the IMSAI backplane, if it's an original, doesn't know anything about that and doesn't provide ground on those pins. Some IEEE-696 boards have jumpers to isolate pins 20 and 70, but I don't remember the RAM16 being one of those.

Personally, 64K static RAM boards using the 6116 are some of my favorite for using in *any* S-100 system. The CompuPro RAM16 and RAM17 are good cards (most CompuPro products are), as is the 64K card sold variously as Coex, Tanner, or Digital Research:

http://www.s100computers.com/Hardware Folder/Digital Research Computers/64K Static RAM/64K RAM.htm

Feel free to PM or email me if you keep having issues with your RAM16 -- I can get it set up for you on my IMSAI if you like, test it out, and send it back.
Click to expand...

No Joy... :(

I cut the ground traces to pins 20 and 70. There was also a via I removed to they were not shorted to each other. I can see the memory board and step through the memory, but write still does not work.

Can you see if you are doing anything else special?

The PLD is a 16v8 equivalent. If I need to I can redesign the logic in that to only support 8 bit mode.

len
 
The front panel probably isn't asserting *sWO (pin 97) which indicates a write operation when low. The RAM16 uses this to control one of its buffers. You can confirm this by trying to write from memory using the CPU itself -- the front panel just asserts MWRITE, but the CPU board will assert *sWO as well. Using a ROM monitor or something will accomplish that. You can program a 2716 EPROM or 2816 EEPROM with test code and plug it into the RAM16, as well. To correct this, you can either modify the RAM16 to use one of the other status signals that the front panel *does* provide, or add a bit of circuitry on a prototype card to assert it only when the machine is halted.

You shouldn't need to reprogram the PAL, as long as the request line for 16-bit mode isn't being asserted, the card will run in 8-bit mode. On my systems, this hasn't required any modifications, but I do typically have a terminator in my IMSAI so the line isn't just floating.
 
Same with my IMSAI. I put Kapton tape over pins 20/70 and removed the data comparator 25LS2521 and set the switches accordingly to fix the board at block 0. I also use the CompuPro boards in my Gazelle (leaving in the 25LS2521 and setting the memory address accordingly). I've never had to modify anything else.
 
glitch said:
Yeah, Tanner/Digital Research/Coex are all the same card, and they do work well in older S-100 systems without modification. They support 24-bit addressing, but it's easily disabled with a DIP switch. I don't recall if they ground pins 20 and 70.
Click to expand...

Mine was grounded, I just cut the traces. It can be repaied if necessary, allthough I only ever plan on using it on older systems.

I got my rom board configured and an eprom burned. I'm just getting my serial card set up and will test the memory card via a monitor in the next day or two. As long as I can write to it I am not to plussed about front panel access. Thank you everyone for your assistance!

Len
 
You must log in or register to reply here.
Back
Top