Combinations and permutation limits on Calculators

[JaSon]

I've noticed that old Hp calculators and Kinpo(recent HP and TIs) can calculate combinations to the full range the calculator can display ie. any combination with an answer less than 9e100 or 9e499 (depending on the calc), But all Casio and Sharp calculators (even a graphing casio)cannot calculate any combination larger than 104C52 which eguals 1.583065848e30, I get an error message trying to calculate 106C53. I can however calculate 114C51=8.27e32, 124C50=1.49e35, 134C49=1.16e37... and so on increasing the number of objects by 10 and decreasing the number taken by 1. I don't see where there would be a similar limit in the calculation n!/((n-r)!*r!) for all these combinations.
Any Ideas?

[R Lion (Spain)]

I had noticed it some time ago, playing with a Casio fx400P and the 15C...
I suppose the algorithms are different and so the Casio gives the overflow error, but can't give you a sure answer.

Raul

[R Lion (Spain)]

I mean fx4000p

[hugh]

i've just tried out a bunch of casio machines and none can perform 106C53, presumably because the intermediate 106!/53! is too large when calculated as 106*105*...*54, is about 1e100.
naturally, all the 1e499 hps, like 20s, 32sii, 48g work. the 15c also does and also 134C49 too which is excellent.
looking at older hps, they dont seem to have it! i cant find "COMB" on the 41c nor 28c. is this missing or have i missed it, anyone?
older still, are there any with combinations on keyboard, 30's and 20's dont (is there one, i dont have all).
can someone try the 42s.
which ti's cope? the 89 simply printed out the whole integer which damn cool.

[Paul Brogger]

The 42s appears to have the same capabilities as the 32S/Sii in this respect. Either handles the examples mentioned, and up to 500C240 (7.83E148), which is as far as I've got time to go, right now.

[R Lion (Spain)]

The 32/42/48/49 can calculate: 5e6C97...