[Date Index] [Thread Index] [Author Index]

Mathematica GeoDistance function performance ...oo

• To: mathgroup at smc.vnet.net
• Subject: [mg129358] Mathematica GeoDistance function performance ...oo
• From: mark at loc8te.net
• Date: Mon, 7 Jan 2013 23:05:00 -0500 (EST)
• Delivered-to: l-mathgroup@mail-archive0.wolfram.com
• Delivered-to: l-mathgroup@wolfram.com
• Delivered-to: mathgroup-newout@smc.vnet.net
• Delivered-to: mathgroup-newsend@smc.vnet.net

While the Mathematica "Geo*" functions are wonderful, I was surprised at how slow the GeoDistance computation is - admitting that I'm just getting started with the Geodesy stuff.  I have latitude-longitude pairs, as decimal degrees, and want the distance between successive pairs of them.  Haven't measured this but the computation seems slow relative to other computations.

I understand that the Great Circle computation is non-trivial.  Has anyone else found GeoDistance performance an impediment?  For example, do the three different methods available to the function make a performance difference?

I'm concerned that assuming a "flat earth" model would lead to cumulative errors in the successive differences computation.

Thank you,
--Mark

Chief Scientist
Loc8te
San Francisco, California
USA