Version 3.00C
Early versions of the Nimbus PC featured 3 cartridge slots.
Details are a little sketchy and it appeared short lived, however details suggest 2 ROM slots and 1 software 'key' slot for protected software.
Research suggests that the ROM system was never put into production.
The softkey, however, does look to have had some degree of implementation, perhaps to secure a CAD application.

A ROM cartridge would allow for fast (read-only) access to an application. It has been suggested that this was developed for educational/comms terminal software use.

It appeared just like another drive allocation and even had the option of being bootable

Size options were 32k, 64k and 96K depending on total ROM IC's added to the cartridge

There was also provision for writeable EEROM cartridges using the same ports with a total of 24K, however at present that has not been tested

The Software Key looked to be an early adoption of a security dongle to protect software. Access to which was denied without the correct 'key'

The June 1986 Nimbus catalogue states:
Softkey is a specially designed software
key or dongle which will allow
developers to ensure protection of their
software copyright. The Softkey
contains a firmware code unique to the
user. The software contains the same
code and needs to be matched with the
softkey before it will run. In other
words, the licensed software can only be
run by a user in possession of the
appropriate unique softkey.

The cartridges looked to have a sliding mechanism, similar to a matchbox. Movement of which would expose the connection contacts.

The task of plugging and unplugging into the Nimbus mechanism would perform this action automatically one would assume.

The only known photo of a 'Softkey'

Example of the ROM cartridge

Showing the ROM Cartridge and the Software Key inserted

Speaking with members of the original Nimbus development team, the theme that develops is that nobody can recall there ever being a ROMPAC on the Nimbus!
This NGC (New Generation Computer) was being developed at a time of great change and knowing which way the market was heading was a challenge.
My theory is that the IBM PCjr was being developed at the same time as the Nimbus and was known to contain ROM Cartridges...If it was good enough for IBM...

A cartridge system was not new to RM, having used such a system on the RM 480Z, via the parallel port. Although again uptake was sketchy.


As time went on over the life of the Nimbus 186 the cartridge ports dwindled away.
My observations have shown combinations of:
1 Softkey
Then no cartridge ports at all, the PCB edge connectors were removed and the flap doors jammed shut.

Later, original style cases did away with the hole and just left a recess.
The redesigned 'slimline' Nimbus case and PCB removed the concept entirely.

The image above shows the inside view of Nimbus
with 1 ROMPAC and 1 SOFTPAC port

The implimentation of the ROMPAC was essentially another block device, like a hard disk.
It could be browsed like any other disk, albeit readonly.
This was somewhat unusual, other systems of the time placed the contents of the ROM directly into memory and executed it on boot up. The Nimbus boot screen would indicate the ROMPAC drive letters.

As there is no known ROMPAC existing and no documentation on their use, as far as I can tell the Nimbus went without a ROMPAC for its working life.
This led me to wonder if the mechanism for accessing such a device ever reached maturity....Hooking up a scope to the port looked promising...activity was seen on the port pins when attempting to view the contents. The schematics released from RM show that the cartridge uses multiples of the 27256 EPROM, giving either 32k, 64k or 96K switching ROMS via a couple of logic IC's
The port goes directly to the IO pins of the AY-3-8910 & MSM5202 music chips.

The feeling about now is why this never took off...the storage space just wasn't big enough and adding more ROMS would have made the cartridge enormous.

There is a documented use for the ports from the 'RM USER' magazine in 1985 speaking about them being used for diskless terminals (Nimbus TN)


The format here is based upon FAT12
Sector 0





00 E9 Jump Instruction
01 18
02 01
03 52 R
04 4D M
05 4C L
06 24 $
07 00 N/A
08 00 N/A
09 01 1=MS-DOS
0A 01 Partition Flag
0B 00

Bytes per Sector (little endian)

&H200 = 512

0C 02
0D 02 Sectors Per Cluster =2
0E 01 Reserved Sector Count = 1
0F 00
10 01 Number of FAT = 1 (no need for duplicates on read only disk)
11 70

Number of root entries, keep this small to avoid wasting space as each one takes up 32bytes, needed or not

&H70 = 112

12 00
13 40 Number of Sectors in logical image
&HC0 = 192 (2 sectors = 1k, so 96K total)
14 00
15 F9 Media Type Identifier
16 01

Sectors Per FAT
1 should be sufficient

17 00 N/A
18 09 Sectors per Track
19 00
1A-1F 00 N/A
20 40

Sectors Per ROM

&H40 = 64

21 00
The rest of the structure is per FAT12 Specifications

Directory listing from a newly made single ROM, ROMPAC

Recreation of the PCB

As there are no known images of the ROMPAC PCB, the following is my assumption on a workable solution

Circuit Diagram

Graphic Impression

Bill of Materials
M27256 3
1uf High Speed Ceramic Capacitor 5
4.7uf Electrolytic Capacitor 2
74ALS139N 1
74ALS74N 1

Hello World (1 ROM Needed)
MS-DOS 3.10 Boot ROMPAC (3 ROMS Needed)
Eagle Project

Finished PCB


I have found no bugs or issues with the ROMPAC system, it does exactly as expected.
The Nimbus can boot from either of the ROMPAC slots should the ROM carry boot files in the same way one would make a floppy disk bootable. Otherwise it can just be used as an additional read-only drive, accessed via a drive letter.

The main configuration of the system is held on the first ROM which contains Sector 0, the File Allocation Table and the root directory. The data then begins and continues on the other 2 ROMs if nessesary.
There is a setting on Sector Zero to indicate how many sectors are on each ROM then it knows when to jump to the next ROM.

My investigations suggest that an alternative cartridge can be configured via additional circuitry to use EEPROMS to make a writable drive, however I have not investigated this further.

The case design is something that I have no immediate plans to complete as it is superfluous to this endeavor

My only assumption to draw from these investigations would be that cost and small storage space were the reasons behind the withdrawal of the product, either at a very early stage or during development.
Anything to add, then please drop me a message

ROM Cartridges
  Main Menu