Big problems with BigIntegers
Back to General discussions forum
Hi friends,
I'm currently using C# to solve these wonderful problems. However, I noticed that the BigInteger type available with the System.Numerics namespace is not supported on CodeAbbey. My previous solution was obtained using my own computer with a version of the C# compiler that supports BigInteger. I'm curious, however: is it actually part of the challenge to solve the problem without using types that deal with very large integers?
Thank you for reading,
LordKuro
You didn't actually specify a problem. However, I cannot think of a Code Abbey problem where you would need to use a Big Integer library. Of the problems that I have contributed I have always ensured that the maximum size needed for integers is 2^63. That should be readily available to you. It is possible that you might find the use of Big Integers to be an advantage for a specific problem. Again I cannot think of one where that is the case. I think that you should consider how to solve your problem without Big Integers.
Problem 67 (Fibonacci Sequence) specifically warns solvers that some input values can be a hundred digits long, which is obviously well ouside the range that most static typing languages admit for integral values of fixed size. This wouldn't be a problem for languages like python. Are you aware of any language agnostic approaches to dealing with the problem of large numbers?
After having solved task 14 (modular calculator), you can solve 67-th using knowledge you have learned from 14-th.
But there is also nothing wrong with using BigIntegers for this problem if you want - I did using Java.
Please note that you don't have to use the C# option offered on CodeAbbey; it's perfectly fine to solve a task locally on your machine and then copy the result and source code over and submit it. For fun, I used a quantum annealing program for one of the problems - Rodion is still working on making this capability available on CodeAbbey ;).
@qwerty - I haven't yet solved Task 14 - thanks for the heads up!
@gardengnome - So cool! Did you have the opportunity to do the one-week D-Wave quantum programming training course? I remember trying to trick my PhD supervisor into letting me use my funding for it, but he was too smart for me. :P
Hi, no I have not done any formal D-Wave training, but they have a lot of good examples and demos, plus you can sign-up for an account with free time (though that comes with some strings attached). D-Wave's hybrid solver comes with a great python interface which abstracts from the underlying principles quite a bit, and that makes using their platform relatively straightforward.