Chris Crawford wrote:
>hi,
> i think we should be using the tangent to the fitted circle for the
>fast fit, instead of simply tracing the first segment. it will give us
>a lot more accurate results, as illustrated by this wild example (run
>3070, event 985)
>--chris
>
I wrote the code for using various methods for fast fit and tangent is
one of
the methods that I wrote. When I using MC, I did not see any big
improvement.
For some tracks, it was doing better but other tracks, it was doing worse.
Anyways, I will put tangent method as a default method and maybe, I will
make a flag so that users can choose a method out of three different
methods
for fast fit.
-T
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST