nesdev.org

One algorithm I know for drawing (approximately) circles is Minsky's circle algorithm, which works like: for(;;) { y-=epsilon*x; x+=epsilon*y; plot(x,y); } The value epsilon is less than one. If it is 1/256 then you do not need to implement multiplication/division; you can just use the carrying (yo...

Here's code (6502.org) for a simple parabolic approximation of a sine wave.

If you run two instances 90 degrees apart that will give you something like a circle.

Of course, it's incremental not random access.

There's an example here (AtariAge) in Batari Basic.

or something in between
use an LUT

one way determine your deltas dx, dy and what their signs are decide which has the larger magnitude develope an index using the signs and which has the larger magnitude ie three bits two for signs and one for x or y having the larger magnitude that will define the octant you're moving in and can be ...

I have written a number of division routines in 6502 assembly, and I'm posting them here for other people to use. :) These routines start with any value (0-255) in the accumulator and finish with the integer division result in the accumulator. They are all constant cycles and do not use X or Y. Mos...

I rather like the Batari BASIC rand16
which justs shifts the two bytes of a
16 bit LFSR in opposite directions then
EORs them together.

One of the simplest PRNGs to implement is a linear feedback shift register. http://en.wikipedia.org/wiki/Linear_feedback_shift_register Here's the 8-bit LFSR I used in my coltrane.nes project. Any starting value can be used for _prng_seed, and it will output all 256 possible values in a fixed seque...

I tried difference of squares years ago, and it took about 90 cycles and a lot more bytes than the 8x8 long mul that Thwaite uses. It does take a lot of bytes both for tables and on zp but it's a lot faster than 90 cycles. in the version attributed to George Taylor (who said he got it from someone ...

I've used Lex and Yacc in the past. I wrote my first several compilers with them. Then I read the excellent article series Let's Build a Compiler by Jack Crenshaw. Although dated (with example code in Pascal) it's still the most effective way to write a compiler, in my opinion. Jack even addresses ...

For a pendulum you wouldn't be using this kind of calculations anyway, you'd be using a sin/cos table and then just determine the point in the circle where you want the pendulum to be... d = x * LUT(y/x*256) That first multiplication and that division look awfully expensive... how about a hakmem - ...

The code effects a jump to an address selected from a table that's inlined. the table is indexed by (the contents of) a. You would presumably do something like lda index jsr SELECT_TARGET . . SELECT_TARGET jsr JUMPENGINE target0 target1 target2 etc, if the targets return with rts But the jsr JUMPENG...

We've been here before. I can't speak for Blargg of course but I assume he's pointing out a general form for one way to do it and expecting you to be able fill in the details. And I don't see it in the Wiki. If you've got a few bytes in RAM to devote to self-modifying code, you could do something li...

Right no true scotsman would use a crappy generator :P I'm not sure how this was a "no true scotsman" argument. If faced with a choice of XORing two 16 bit generators, or one equivalent 32 bit generator (or even a 24 bit generator), I believe you would find with empirical testing that the...

A common ploy is to combine two different types of PRNGs in the hopes that they'll hide each others deficiencies. Say, an LFSR and an LCG XORed toghether I would not recommend this. Not sure I'd go so far as to recommend it, but it certainly is a common ploy. A better quality generator will be supe...

Well yeah, unless I can find a decent 32-bit polynomial (decent meaning maximal-length, as few 1s as possible, ideally with them spread out and away from the most significant 8 bits, while also allowing a good random distribution), it's probably a lot easier to just use a traditional LFSR instead o...