KepTronics SRE-1M

[Roger Hanscom]

Anyone familiar with this S-100 board?  I was trying to build it from a blank PCB that I acquired a long time ago.  It is a 1 MB static memory.  I was hoping that it would be useful for one of my Cromemco boxes, but so far it is non-functonal.  It has a phantom option.  I wired the chip select for the on-board RDOS ROM on the 16FDC to pin 69 of the bus.  I didn't expect that would work, and was not surprised that it didn't.  I believe that the Cromemco FDC boards require some R/W memory to function, so if the 16FDC is running in the ROM, phantom would assure that the SRAM wouldn't be available?  Is that right?  So would a board like this ever be useful in a Cromemco box?  Is it possible that a 256KZ could be used at low memory addresses, and the 1M memory be addressed above that?  Does Cromix expect contiguous memory addresses?  Just wondering.

Roger

[curt mayer]

https://bitsavers.computerhistory.org/magazines/S-100_Journal/v01n05_1987_Winter.pdf

has a complete writeup of the board, with design information, bus timing, schematics, the works.

On Wed, Feb 12, 2025 at 11:17 AM Roger Hanscom <norwe...@gmail.com> wrote:

Anyone familiar with this S-100 board?  I was trying to build it from a blank PCB that I acquired a long time ago.  It is a 1 MB static memory.  I was hoping that it would be useful for one of my Cromemco boxes, but so far it is non-functonal.  It has a phantom option.  I wired the chip select for the on-board RDOS ROM on the 16FDC to pin 69 of the bus.  I didn't expect that would work, and was not surprised that it didn't.  I believe that the Cromemco FDC boards require some R/W memory to function, so if the 16FDC is running in the ROM, phantom would assure that the SRAM wouldn't be available?  Is that right?  So would a board like this ever be useful in a Cromemco box?  Is it possible that a 256KZ could be used at low memory addresses, and the 1M memory be addressed above that?  Does Cromix expect contiguous memory addresses?  Just wondering.

Roger

--
You received this message because you are subscribed to the Google Groups "Cromemco" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cromemco+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cromemco/CAJF4YRtuJP4HSkKKkJmepjeyqD7kje%2BF5zNBZDa9m6h_4-TtvA%40mail.gmail.com.

--

--curt mayer

+1 (707) 540-3441

I refine, they debug, he/she patches, they kludge.

[Roger Hanscom]

Thanks for the response, Curt.  I already have the full documentation booklet that came with the (blank) PCB.

Roger

[Peter Higgins]

This board should work in your Cromemco box when used at the bottom of the S100 memory space. By using the RDOS ROM chip select signal to simulate PHANTOM signal, the RAM board is only disabled when the CPU is accessing addresses within the range of the ROM itself. That same mod also works with the S100Computers 4M RAM board.

Routing the RDOS ROM chip select signal to S100 bus pin 69 is the correct modification to make on the 16FDC board. However since you did not mention it I must ask - on the RAM board, did you (a) run a jumper from S100 bus pin 69 to the pin of the IC (shown as a 7421 in the schematic)  the the S100 PHANTOM signal is currently connected to and (b) also on the RAM board, cut the trace from pin 67 (which is the actual S100 PHANTOM signal line) going to that same IC pin? This is necessary since S100 bus signal 67 is the actual PHANTOM line but cannot be used for this mod, since Cromemco uses S100 line 67 for another purpose hence the need to use S100 bus signal line 69 which is normally unassigned.

Incidentally... thanks for mentioning this board. I have a couple of these sitting around unused that I acquired along with a few other boards a while back. I thought that someday I'd have to try finding documentation for it, and of course I'd long forgotten the designer had written it up in the Winter 1987 issue of S100 Journal.

[Peter Higgins]

One more thing... the full 1M of RAM on this board only works with 68K Cromix, which operates in a single contiguous memory space. It will not work with Z80 Cromix, which only works with Cromemco's RAM boards that support their memory banking scheme.

[Roger Hanscom]

On Thu, Feb 13, 2025 at 1:02 PM Peter Higgins <higgin...@gmail.com> wrote:

<snip>

>> I thought that someday I'd have to try finding documentation for it, and of course I'd

>> long forgotten the designer had written it up in the Winter 1987 issue of S100 Journal.

I have the complete documentation for the board, including a very legible schematic.  I could scan it, or photocopy it, and send it to you if you'd like.  Just let me know.  It has a print date of Dec. 1986, so maybe there are other revisions of the board/document??  My board has (c) 1986 in the silkscreen.

Thanks for the info. about phantom.  I'll have to take a look at that again.  I must say that there is no 7421 on my board.  I didn't realize that the normal phantom pin (67) was used by Cromemco for something else.  The SRE-1M has phantom going to S-100 pin 67.  Maybe that's why I can't get anything to work right?  Back to the drawing board!  <grin>

Also, I knew that a board like this wouldn't make much sense (let alone work) for Z80 Cromix.  My box will never have more than one user (me), so a lone 256KZ should be more than sufficient for my Z80 Cromix.  As far as contiguous memory is concerned for 68k Cromix, the sys. admin. manual I have states that memory would not have to be contiguous, but I'm having a hard time believing that.  It might be an error?

Roger

[Peter Higgins]

My reference to the S100 PHANTOM signal providing input to a 7421 was based on the logic diagram provided in the "S100 Journal" article - which likely was changed in the production versions of the board. Nevertheless - based on your board's design:

- you need to trace where the S100 PHANTOM signal (normally pin 67) goes to on your RAM board, and break that connection

- instead, feed the signal from S100 pin 69 to wherever the pin 67 connection used to go

That *should* work.

The discussion regarding the problem with using the "real" S100 PHANTOM line is found here:

https://groups.google.com/g/cromemco/c/T_eui2YJEQ0/m/FkrSFaYtBwAJ

Apparently the DPU, XPU and XXU boards pull pin 67 low all the time, which interferes with normal S100 usage of pin 67 for the PHANTOM signal and would of course disable any RAM board built to the IEEE696 S100 standard.

I have not run across any indication that memory does not have to be contiguous - if it did, I'd think it was an error. The "68K Board Family" manual and all the versions of the Cromix Plus "Sys Admin Manual" I've come across seem to clearly indicate multiple RAM boards should be configured to provide a contiguous memory space.

If you can produce a scanned copy of the manual and schematic- yes, please send it.