Version 5 plan

The question of what to put in version 5 is going to take a while. Version 4 has a nice feel to its operation. Perhaps the heart icon could play a role. I think I'm going to go for some more efficient button push way of building in 16 functions into the slide. I'll probably use my own particular extra eight. At the moment this is likely to include 4 functions relating to the higher maths I'm working on. (Exponential, inverse Lambert-W, Series sum from 1 to infinity of x to n over (n factorial n) which I call Q or the log complement function, and the sum of the natural log and the log complement, which is the Ei(x) function at it's natural origin (limit set)). Don't ask.

I think the other 4 functions will be the logarithmic integral at the natural origin and 3 which work slightly different with 2 inputs of 2 digit each (splitting the input into digit pairs) to a 4 digit result. The operations will be times, divide and harmonic parallel. I assume most people can add manually. The logarithmic integral provides a good estimate (not exact) for the number of prime numbers less than a number.

I may introduce other functions, including 8 hidden functions on the long press STAR and NUM. I will decide on this when I have optimized the scale and pre-scale of the functions already present. I expect sin to be easy, then cos, asin and acos. The extra four would be ...?

I may factor SET_ON and the number to string conversions into one magic display value routine. I think it may use slightly more power, but will reduce code size, as parameter value constant reduction can not be done using fixed libraries. Although this may be left until version 6 as I do not expect the slide additions to exceed the memory.

a little explanation of the exponential functions group

Put simply x*(df/dx) and I(f/x)*dx are better operators. The constant remains constant "constantly". I call this ratiometric calculus. To find out more look up the functions, and maybe "wedge product" and "Lie algebra differential equation". It appears that polynomial term, Q, log and exp terms multiplied may form a closed group under calculus operators. Assuming that division is an iterative multiplication process, and that powers of products of Q, logarithm, exponential and polynomial are also integrable in terms of such.

Put simpler Newton's calculus is a process of differentiation (throwing away constants), and integration (making constants up), and shit is isomorphic to constancy. Strange thing is the constant integration sequence becomes a polynomial of log in ratiometric calculus.