I need a lot more precision in my program than doubles can provide. I''d like to use a class that can handle arbitrary lengths and perform almost all operations standard ints can. Anyone have any advice? TIA, Avalon

Curiously enough, I was thinking about this exact problem earlier today. The best advise I could give you is to learn how floating-point numbers are structured (which is like scientific notation, I believe), and using assembly probably wouldn''t hurt either. I don''t believe it''ll ever be as fast as standard floating-point types since the processor based floating-point operations are design for numbers of a certain size, but I could be wrong. Here''s an example of what I would probably do.

  class real {private:   long int value;   char     exponent;public:   real();   real(val, exp);   real(float);   real(double);   ~real()      // operator functions, etc...};  

The danger from computers is not that they will eventually get as smart as men, but that we will meanwhile agree to meet them halfway - Bernard Avishai

Curiously enough, I was thinking about this exact problem earlier today. The best advise I could give you is to learn how floating-point numbers are structured (which is like scientific notation, I believe), and using assembly probably wouldn''t hurt either. I don''t believe it''ll ever be as fast as standard floating-point types since the processor based floating-point operations are design for numbers of a certain size, but I could be wrong. Here''s an example of what I would probably do.

  class real {private:   long int value;   char     exponent;public:   real();   real(val, exp);   real(float);   real(double);   ~real()      // operator functions, etc...};  

The danger from computers is not that they will eventually get as smart as men, but that we will meanwhile agree to meet them halfway - Bernard Avishai