Re: Big numbers

I would have thought that the automatic conversion to float on overflow is a bad thing - it would be much better to generate an error at this stage, then at least the software has chance of catching the problem and dealing with it. There are lots of things that can go wrong other than the bitwise operations, for example, eventually i++ doesn't do what you expect when i=10^300. In all scripts which are suffering from the problems you describe, the authors really want the number to remain as an integer. My guess is that not many, if indeed any scripts would break if you stopped the conversion to float from happening. Cheers -- Phil Driscoll Dial Solutions +44 (0)113 294 5112 http://www.dialsolutions.com http://www.dtonline.org