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

Re:simple equation system -- crashing M5?

• To: mathgroup at smc.vnet.net
• Subject: [mg44282] Re:[mg44256] simple equation system -- crashing M5?
• From: Rolf Mertig <rolf at mertig.com>
• Date: Tue, 4 Nov 2003 03:23:54 -0500 (EST)
• Reply-to: rolf at mertig.com
• Sender: owner-wri-mathgroup at wolfram.com

>Has anyone had such an experience?? Any ideas??
I do not know how to use the built-in Solve or related functions to solve 
your system (though I am sure it is possible somehow).

However, one can just implement a modification (using Collect 
and a modification of Factor for intermediate results) of the 
"High-School-Algorithm" to solve this type of equations.

I have done this some time ago in Solve3.m of FeynCalc (since this type of
equations come up frequently in Feynman diagram and integrals calculations).

On a decent Linux machine Solve3 gets a  
result in about 6 minutes. Using Simplify to get a shorter result takes 
not more than 67 MB.
Since the result is somewhat longish (LeafCount is 176202) I put it here:

http://www.feyncalc.org/download/examples/solve3.m
http://www.feyncalc.org/download/examples/solve3.out

Regards,

Rolf Mertig

---

Mertig Consulting
http://www.mertig.com