using intel assembler e.g multiplications and division on fractional numbers.
1) I see no way you can do this on any "fractional number" if those numbers are variables without invoking floating point.
2) floating point in assembler sounds like a lot of fun, but it will take you many months to do.
3) post back next year when you are done.

if you ARE going to do this
1) get a processor with mathematical capabilities beyond 8 bits
2) use C

Erik

PS: if the fractions are one or two constants you will be much better off with using fixed point and scaling; however with the resolution 23512/25523 indicates, even that will be a task.