"Unlimited" Length Floating Point Numbers
I wrote this class a little while back and wrapped a calculator around it. I always intended to release the source once I had it at a decent working level (although it's still aesthetically displeasing). The class is called HugeFloat and it allows you to perform arithmetic with unlimited length (it's only limit is the end of memory) floating point numbers rapidly and accurately.
Please download it and give it the old once over. Take your time. There's no need to rush this. Then, once you've had the opportunity to marinate in your experience, share all of your emotions with me. Don't keep them bottled up inside. If you do, some day you might explode, and no one will be around to pick up the pieces. We all know you don't want that.
Download it here: http://www.geocities.com/lucky760/hugefloat/