For students who finished the assignment early.
Crowdsourcing experimental data .
If you would like to participate in an informal experiment contributing to the body of knowledge in this area, read on. I'm inviting students to use their own simulator to obtain a number of data points that can be pooled to make a larger data set. The data points represent a hopefully converged state (for DI you must go through at least 100,000 cycles, others may converge earlier) for a certain grid size and density, and a random but reproducible start. Download this spreadsheet : it contains instructions and the sheet that you put your results on.
There's a paltry mark or two in it for you, but hopefully it's an interesting exercise as well. Pooled results will be publicised: I don't know whether any trends will become apparent (especially in the PI period), but it's worth a shot. Although the BML area has been quiet for the last few years, there have been quite a few research papers with no more data basis than what we're doing this week.
I'll accept contributions up to noon Saturday
Resource created Thursday 19 May 2016, 08:39:56 PM.