21
Apr
2009
granularity
I’ve been telling people that turning up the analysis resolution on their Ecotect models is a bad idea, but I’d never stopped to figure out just how bad.
This table shows the rate at which the numbers grow as you go up through the resolution steps.
| Sky subdivision |
Lowest |
Low |
Medium |
High |
Highest |
|
| Grid subdivisions | Number of divisions |
144 | 324 | 1296 | 2700 | 8100 |
| Low – 1 | 1 | 144 | 324 | 1296 | 2700 | 8100 |
| Medium 5 x5 | 25 | 3600 | 8100 | 32400 | 67500 | 202500 |
| High – 10 x 10 | 100 | 360000 | 810000 | 3240000 | 6750000 | 20250000 |
| Full – 25 x 25 | 625 | 225000000 | 506250000 | 2025000000 | 4218750000 | 12656250000 |
For those of you who are a bit rusty on your powers of ten (like me), that bottom right number is 12.5 billion! That’s for each panel, so if you have a 5×5 subdivision on a grid, that’d be into the trillions of calcs for your analysis. No wonder it takes a long time.
I thought I’d check to see if I was supposed to be using British billions, or American billions, and it turns out (Long_and_short_scales) that all billions have been the same in the English speaking world since 1974!