February 14th, 2003


Debugging update

Things continue. I didn't write a report yesterday, because I was calling lilairen in the evening, and forgot to post stuff first. Yesterday, I fixed a couple of small bugs, and then discovered one that was annoyingly difficult to fix.

Collapse )

Today started with a bit of a digression. Another thing I've been needing to do is make the registration brochure for a conference that my advisor is hosting -- I've got earlier versions of that, so it's just an update with new material, but it still ended up taking all afternoon because it needed to be completely rearranged to fit on a 4-fold rather than a 3-fold brochure layout to include all the info.

Collapse )

Next up: While it may be working, it's not really solving the right problem -- it's solving for flow around a cylinder with extra viscosity added to make things behave better when it's not close to the final solution. I need to make it not need that; probably what will happen is having that in at the start of each run, and then slowly taking it out once things settle down to something close to a solution.

And then I need to try some more interesting sorts of flow coming in and see if I can get data, and then I need to generate some piles of data.

Then, phase two of this project....

Debugging, again.

It works! *bouncebounce*

Well, er, almost. It's supposed to have the error go away exponentially (i.e., if the error in the solution after 250 steps is 1/10th the error after 200 steps, then the error after 300 steps will be down by another factor of 10 from that), but on very fine grids, there's a small bit of error that goes away linearly -- which means it goes away very slowly. But it's a very small bit, and in general it works, and it now takes 20 minutes to do calculations that would have taken 20 hours before I started this bit of debugging and stuff.

Meanwhile, I go home now and spend the evening with suzanne. No more debugging of programs today.