The backend has been updated to perform stability checks using a robot that periodically scans the library. Unstable fits are marked as such on the the database view.
User EricFDiaz posted a useful essay about the importance of turning on integration before posting the system on the backend. The post has been made "sticky" on the forum. » Read
Backend bugfix
Happy new year! The "inf" χ2 has been fixed for all fits, and the database should now reflect the corrections. Thanks for all the reports!
EricFDiaz says:
It is a tough system, Tom. But hey...at least you improved on my Chi^2 value. Some systems are like that; no matter what you do they just come out with incredibly high Chi^2 values. Eric
tomdickens says:
OK, apparently Markk owns this system!!! I uploaded it twice and both times, as you can see, it was assigned to Markk! He probably doesn't want to be blamed for my lousy fit...
tomdickens says:
This one is tough. The perturbations are so strong it is hard to get the Levenburg-Marquardt polishing started in the right local minimum. Plus it's very slow.
EricFDiaz says:
Hi Chris, It's interesting to note that you and I got results for this system which are extremely close--not identical but extremely close. I didn't know until I looked at your fit and the others that I hadn't done one for this system. So, I decided to do one. Can you imagine my surprise when I got practically the same result as you? And, both our fits are not that different from Greg's fit either. Maybe we have something here. I wonder what would happen with the introduction of a fourth planet...