The PB-2000C is one of the most unusual calculators made by Casio.

This curious beast is actually programmable in the C language. No kidding! You write your program in plain (pre-ANSI) C, save it from the built-in text editor, compile it, and run the result. This is true C we're talking about, ladies and gentlemen, not some emulation/simulation: you get the whole works, pointers and all, so it is in fact quite possible to lock up the machine for good with a bad pointer assignment, for instance. A small RAM-based file system implementation is used for storing program and data files. The machine also has the ability to interface with external peripherals. This is a surprisingly useful machine!

I received this machine as a gift from Paul Fox, author of the venerable CRiSP editor. Although Paul was unable to supply a manual, it was not very difficult to figure out the basic functionality of this computer. (Mind you, it'd be nice to get my hands on a reference manual for all the built-in library functions.)

In addition to C, the machine also provides limited "programmability" in the form of a single-formula storage memory where a formula can be stored, edited, and executed.

The C implementation is quite robust. It can even handle fairly obfuscated C code, such as the following single-line monstrosity computing the value of π to an arbitrary number of digits:

long a=10000,b,c,d,e,*f,g;
main(){printf("digits?");scanf("%ld",&c);c*=3.5;c-=c%14;f=malloc(4*c+4);
for(;b-c;)f[b++]=a/5;for(;d=0,g=c*2;c-=14,printf("%.4d",e+d/a),e=d%a)
for(b=c;d+=f[b]*a,f[b]=d%--g,d/=g--,--b;d*=b);}

(This code is not of my own creation. I downloaded it ages ago from the Internet. I have no idea as to the identity of its original author.)

Of course the machine isn't particularly fast. Computing 50 digits of π takes several minutes. But, it works.

I have, of course, also written a Gamma function program for the PB-2000C. Finally, I could do this in C! This beast computes the logarithm of the Gamma function for any real argument to 10+ digits of precision:

double lg(x)
double x;
{
	double g;
	double pi = 3.14159265359;
	int s;

	s = x<0;
	x = s ? -x : x;

	g = 2.506628283501;
	g += 92.20704845211 / x++;
	g -= 83.17763708288 / x++;
	g += 14.80283193078 / x++;
	g -= .2208497079533 / x;
	g = log(g) + (x-3.5)*log(x + .85) - x - .85;
	return s ? log(pi/(3-x)/sin(180*(x-3))) - g : g;
}

main()
{
	double g, x;
	printf("%lf", &x);
	g = lg(x);
	printf("lnG(%g)=%14.12g
", x, g);
	printf("G(%g)=%14.12g", x, exp(g));
}