Why is this exciting. because the previous iteration of the code was slooooow - it took hours (even days to do this) and generated lots of temporary files.

Why is this important.? The LSST will discover ~5 million Solar System objects... &

.and to compare the observations to models of the Solar System populations we need to simulate millions of synethic small bodies from the model and figure out which ones would be detected/observed so we can compare them to the real detections/discoveries from the survey

I might need to run millions of simulations tweaking parameters from the model - if it takes days or weeks to run one simulation that is not going to be feasible -so we need to get down to running in minutes or as close to that as possible for ~10^4-~10^5 simulated asteroids

This is a big group effort @phbernardinelli Aidan Berres, @coc415 Carl Christofferson, Sam Cornwall, Siegfried Eggl, Grigori Fedorets, Matt Holman, @Lynne73 , @mjuric , Jeremy Kubica, Jake Kurlander, @javacitrus , Conor MacBride, Shannon Matthews, Steph Merritt, +

@phbernardinelli @coc415 @Lynne73 @mjuric @javacitrus Joachim Moeyens, Shantanu Naidu, Drew Oldag, Brian Rogers, @colinsnodgrass , Max West, Dave Young round out the team

The code's still in active development - the big new pieces have been added but in a quick way for testing they need to be be fully integrated. So still lots to do in the next few weeks of the incubator. But this a huge step forward for the project

I should note that sorcha (the name of the simulator) is python and open source. So once it's ready the team will be writing a paper and it will be available for the community to use

@megschwamb @Astrogator_Mike Can you say what the biggest 2-3 changes were to get this speedup?

@ted_dunning @Astrogator_Mike Generating the ephemeris, going from orbit to where on the sky the simulated planetesimals are on the sky and figuring out which telescope pointings the objects are in, handling 10^6 simulated objects quickly