Big Bear Version 3 - First End-to-end Demo of an early prototype

The re-write and modernisation of Big Bear Model Rail software reach's a milestone with the first successful end to end, screen to model, layout demonstration.

Big Bear came into the world in 2005, starting as a simple program to help a friend control his model railway, and evolved into a fully-fledged application for controlling track and the trains. The original system has been popular with users, and the software was based loosely on the architecture used for medical imaging software I had previously been contracting for.

However, as with all software, it gets old, creaks, and is proving hard to extend, not helped by me losing the original graphics files :( So much for backups. So it was time to bite the bullet and start from scratch. Out goes the inflexible PNG graphics and in-comes SVGs that can be made as big or as small as possible. This was also a great opportunity to learn some new skills, so the front end is now based on ReactJ (even discovered in the process that Javascript isn't quite as bad as it used to be). With the help of style sheets and a smattering of DevExpress controls the user front end is taking shape nicely.

Real-time communications with a service are accomplished with SignalR. Oh, how I love SignalR. It's one of those technologies that does what it says on the tin and just seems to work so well. SignalR will be getting a valentines card come February. The service itself runs using .Net Core 3.1, which hopefully means I will be able to compile it for other platforms. In fact, the whole system is being designed with the view of running on tablets and other stuff that isn't Windows. The service supports all the functionality and utilises Core 3.1 plugins. After some pain dealing with reflection, got there in the end, and means adding functions will be the case of 'dropping in' DLLs.

So, what does mean for the model railway? So far we can create layouts and from the vblog, switching a railway turnout or point on the display will send messages to the service which in turn converts these into the signals required to send to the hardware controller that in turns switches the real live turnout on the layout. So here goes, a video demo.