[Shootout-list] Keep the used compiler consistent?

[Ales Katona]

Hello, I'm writing to ask if you could please ether re-run the tests 
with Free Pascal Compiler v.2.0 (as you say on the main page that you 
use) or drop the old tests.

I've noticed that many benchmarks are still run with 1.9.6(atleast their 
results are from that version). It wouldn't be a problem if 1.9.6 was 
normal stable version but it was a beta.

I've also submitted a new benchmark for the sum-file. It seems it's a 
bit slower than the old one from 1.9.6 but the memory use shrinked from 
~30000kb to ~100kb. I'm also asking how are these things weighted? CPU 
speed is only about 10% less with the new benchmark but memory usage is 
1000%+ better and the old benchmark is still used.

Ales