[Simh] Contributing to SimH

[Bob Supnik]

Christian Gauger-Cosgrove's comment about contributing to SimH triggered a thought I'd like to share.

Not surprisingly, there has been a lot of focus on the systems many of us used in the past, like the PDP-11 and VAX, and on improving SimH's usability. However, the purpose of the project is to preserve historic systems and software through simulation. More than 40,000 different computers were designed before 2010, and SimH supports a couple of dozen. So anyone can contribute to SimH by:

1. Writing a new simulator.
2. Extending an existing simulator with new IO devices or models.
3. Debugging beta simulators that are not fully tested.
4. Debugging software that is available but not tested.

All of these activities tend to be well isolated from the core control package and libraries. Writing a new simulator is unlikely to break SimH.

Older machines tend to be simpler to simulate than recent ones, although if you go really far back (like the IBM 650), you may run into issues like plugboard programming that are difficult to understand, let alone simulate. Old software is invariably in assembly language, and if that is out of your comfort zone, debugging old software can be problematic.

Some random suggestions:

New simulators: Bendix G15 (software is available); Datacraft/Harris 24b systems (software is available); DG 32b systems (software licensing is a problem); Prime 16b and 32b systems.

New IO devices or models: KL10 (porting Ken Harrenstein's KLH10 to the SimH framework); other IBM 14XX models (for example, the 1440).

Debugging beta simulators: Sigma series.

Debugging software: SDS 940 timesharing; CTSS for the IBM 7094.

I'm quite encouraged by recent developments, such as the HP 3000 simulator, the KA10/KI10 simulator, and the restoration of Unix "v0" (PDP-7 Unix). But there's always more to do.

/Bob Supnik