On Jul 24, 2013, at 4:45 PM, check.nyah wrote: > In relation to my project I have developed a more efficient algorithm for > implementing the bn_mat_inverse and bn_mat_identity() routines. Since I > believe it will increase run time of my pull routine since it takes > bn_mat_inverse() seriously; and the current implementations are not good > enough. Just wish to know if its ok?
For core math functionality, you should demonstrate that yours is 1) correct and 2) better. Write a little test program that compares your new version with the previous version, feeding random matrices to it in a loop until it takes at least 60 seconds. Run it a few times to make sure the numbers are stable. Then replace your function with the existing bn function and compare the time. Cheers! Sean p.s. There is no bn_mat_identity() function. There is a MAT_IDN() macro, a bn_mat_identity global, and a MAT_INIT_IDN static initializer. ------------------------------------------------------------------------------ See everything from the browser to the database with AppDynamics Get end-to-end visibility with application monitoring from AppDynamics Isolate bottlenecks and diagnose root cause in seconds. Start your free trial of AppDynamics Pro today! http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk _______________________________________________ BRL-CAD Developer mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/brlcad-devel
