Ways of finding out how code works

Today(or rather should that be this evening) I have made a big push in finding exactly what code (which hasn’t been commented!!!!) does for a section of my project.

The offending code is related to the Morimoto et al Oscillator. In which the code I am basing my project on has a bear bones implementation of the Morimoto et al Oscillator.

I have spent the night running test on the simulator getting output on various variables to decipher what code is linked to which equations.

I have nearly mapped out most equations to the code bar a few which are still puzzling me. A reason why is that the graphs (what would I do without gnuplot!) are sometimes more confusing the the uncommented code!  As shown below!

I’ll have a deeper look tomorrow and try to figure out the rest of the code. But progress tonight has been good 🙂

Posted in Research | Leave a comment

Progress…. finally!

Realised that I haven’t written anything for some time. I got a full time job, which has put some strain on my project but progress is being made.

In theory there are 3 sections to my work (ZRAM, Morimoto Oscillator and the combination of both). The ZRAM section is nearly done, with just the support polygon and the method to control the ZRAM point to deal with. The Morimoto Oscillator is thankfully just an add-on to work already done previously, maximum of 50 odd lines. The combination would be the easiest part which one one line of code to change 🙂

The next part would be to optimise the code via a neural net or GA. I have been thinking of the best way to do this. Previously we have used a grid computer to perform the process intensive neural nets/GA’s. But access is intermittent, I have been thinking of using my Playstation 3 which I can put Ubuntu on and leave it running 24/7. Progress with this will be detailed later…

That just leaves 12,000 words to write up for my report… 1000 words done so far…. time to put on the motivational music! 🙂

Posted in Research | Leave a comment