Full Circle

As you might have gathered from reading some of my previous posts, I'm doing some simulations to back up my arguments in my thesis. They're not the purpose of the thesis, but they can I suppose be likened to the bonus features on a DVD (although regrettably there's not going to be an author's audio commentary). Anyway, the simulations I'm doing at the moment are for one of my very early chapters. Oddly enough, I had never run any simulations for such a simple parameter set - the very first simulations I did back in my first year had an extra element which made them more complex than the ones I'm running now. The perhaps obvious way to have set these simulations up then, would be to remove the extra element from that first lot of simulations by doing a new version of the computer program and nullifying all references of that extra element. However, I decided to go down a different path.

Over the years of my PhD, the computer programs I have written have in a way undergone reproduction, or spawning. That is, the program for each particular task was based on the program for the previous task, but with some additional feature. This is reflected in how in a previous post, I talked about how my thesis structure felt a bit like a rainbow. Each chapter built upon the previous one. Thus I eventually reached one 'end' of the rainbow, but I wanted to get back to the start, the other 'end'. To do this, I took the final computer program that I had written, which was rather complex, and then created a special case of it by turning the variable parts that made it complex into more constant entities. This generated the conditions required for the earliest simulation. I set up a test to see whether this new program and the special case of the most complex program gave the same results. Thankfully, they did. Now I just have to wait for all those numbers to be crunched.