That's right, drawing 8x8 sprites is actually FASTER than drawing text.  Not only because the routine is faster, but also because it is drawn to the buffer instead of directly to the screen.  This is a paradigm you'll just have to get used to.

I had to do that in Pyoro for the scores.  The main reason was because regular font will not display correctly in the grayscale engine I made, and to make drawing as fast as possible in the least amount of memory, aligned 8x8 sprites are the way to go so I had to make my own number font to fit this size.  It was 7x8 instead of the standard 5x7.

I think a nice feature would be to expose the graph buffer as an array or pic or what have you. Could be useful for more complicated effects, if people want to write their own routines

randInt will come eventually.  The best way to get random numbers between A and B inclusive is like this: A+rand^(B-A+1) and this will basically be what that function will do.  So if you need a number 1-5, then simply do this: rand^5+1

Passing parameters is unnecessary complexity.  Saving into variables in the main program and then recalling them from the subprogram is about as efficient as it can get and is much simpler to understand for the programmer.  But like I said, I'm not sure if I will allow program modification tools.  You can embed an external program inside the compiled program and call it from there, but you will most likely not be able to call an arbitrary external program from the compiled code.  That would require error handling, something which I will add later, using the "Theta" variable.  0 means no error and all the other values represent different types of errors so you can check if some of the higher level commands executed properly.

ah ok, well thanks that beats my method which was rather crude and ineffective.

Mmmm since it is not possible to compile archived programs, do you think you could remove them from the compile menu?  Just a little nag

Hmm, the headers could work, and the Axioms idea. I had been thinking for a while about having an include statement so you can split your file into smaller more managable chunks for development.
How would these Axiom work? 

The include statement will just be inserting "prgmNAME" into the code somewhere.  I'm not sure how the Axioms will work, but they definitely won't be available on any time soon.

Just to give everyone a heads up, I am making a major change in the next version.  I ran into a problem.  When you display stuff, how does the compiler know in what format to display it?  How does it know if you want to display a string, a number (and if so, in what base?), or a character?  All three of those have a number as an argument, either as a pointer for strings, or a number for the ASCII value of the character and the plain old base 10 representation.  Therefore, I must add some way to tell how you want to display that number.

So in the future, you can display a string by omitting any additinal notation, as you have already been doing, but to tell it that that number is not a pointer to a string, you have to use >Dec at the end to tell the Disp or Output command that you intend to display that number as a base 10 representation.  Similarly, I might use >Frac to display the ASCII value of the number.

Sorry if this makes things a little inconvenient, but it will only reduce the size of the compiled program and add even more flexibility.  Its a very simple thing to fix to make older programs compatible with the next version.