I discovered a relatively compact way of calculating a cube root on an ordinary calculator. The procedure is built on the formula for geometric series and on the calculator’s hidden “K” – constant – register.
By “ordinary calculator” I mean one of those inexpensive calculators that you can find in endless sizes, shapes, and styles, but all of which use identical calculating logic. They all include, among other things, an 8-digit display, square root key, and two clearing keys, AC and CE. They also all include an internal K register. I suspect that over 99% of people who use ordinary calculators have no idea what the K register is, or that it even exists. This is because I suspect those people who are geeky enough to figure out the K register don’t use ordinary calculators. Instead they buy and use fancy scientific/ programmable/ graphing calculators.
Now, to calculate a cube root. First enter the number, then press [√] [√] [×], then repeatedly press [√][√][=] eleven or twelve times at which point the number in the display stabilizes. That number is the cube root correct to seven digits. For example, take the cube root of 10:
Enter: See displayed:
[1][0] 10
[√][√][×] 1.7782793
[√][√][=] 2.0535247
[√][√][=] 2.1287512
… 7 more iterations …
[√][√][=] 2.1544339
[√][√][=] 2.1544342
[√][√][=] 2.1544342
The answer differs only in the last digit from the accurate approximation 2.1544347.
This algorithm depends on the geometric series for 1/3:
1/3 = 1/4 + 1/16 + 1/64 + … + 1/4n + …
Packaged into exponential form:
x1/3 = (x1/4)(x1/16)(x1/64) …
And calculated with the assistance of the calculator’s K register, which caches the factor x1/4, while the product builds up in the displayed result.