# Random: Any way to calculate Transcendental numbers using abacus?

1. Mar 31, 2009

### Pinu7

For example is it possible to approximate cos(17) using a Japanese abacus?
Even if it takes a while. I'd imagine it would have to do with the Taylor expansion or something but I'm not sure.

I think it is important because one day calculators will turn against us in the 2014 robotic uprising.

2. Mar 31, 2009

### tiny-tim

Welcome to PF!

Hi Pinu7! Welcome to PF!

There's a work-in-progress series of lessons on the traditional abacus by Derrick Coetzee at http://moonflare.com/abacus/index.html which should soon have cosines …
… let's hope he finishes it before 2014!

3. Mar 31, 2009

### chroot

Staff Emeritus
Yes, you can calculate transcendental quantities with an abacus, to any precision you desire. An electronic computer is very similar to an abacus, in that it has finite states with a deterministic means of moving from one state to another. You could probably back-adapt the algorithms used by computers onto an abacus if you're so inclined.

- Warren

4. Apr 1, 2009

### HallsofIvy

Any decimal approximation to a number is a rational number and it is possible to calculate any rational number on an abacus. Thus, you can approximate any number on an abacus.

5. Apr 1, 2009

### Pinu7

Well, I am still unsure how exactly to calculate some of these basic numbers. I guess I could use the Maclaurin series for some of them. However, I can't find an algorithm for finding powers of numbers that are needed.

Perhaps, my dream of defeating our robotic overlords is doomed. :'(

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook