F distribution on the 41C

[quantalume]

(12-18-2015 08:38 PM)Dieter Wrote:  For the record, here is my attempt at a F distribution program for the HP41. I hope there are not too many errors – try yourself and see what you get, I did not do any thorough tests.

Looks very good! Preliminary testing found no errors, and reduction in execution time compared to JMB's solution varies but is up to 50%. I made a few minor changes; I replaced the LBL 70 - 90 with LBL 02 - 05 (single byte). This reduced program length by 7 bytes and improved speed slightly. Your code is 74 bytes shorter than JMB's, although JMB's includes a standalone continued fractions module.

Dieter Wrote:The major limitation of this program is the Beta function which ... means that odd d.o.f. can be up to 70 and even d.o.f. up to 140.

I don't think anyone has any business doing problems with dof over 70 on a pocket calculator, let alone a vintage one. :-)

Thanks for your efforts!

David