3 ms·
Nice. Bonus is that it doesn't use self modifying code and another plus one for using ldx(labs0),Y - Looks like this one can also be optimized for repeated m
by djmips 3y ago
Nice. Bonus is that it doesn't use self modifying code and another plus one for using ldx(labs0),Y - Looks like this one can also be optimized for repeated multiplication by the same factor by skipping the first 4 instructions.
Did you write this one?
- dzdt 3y agoOh crap. The LDX (labs0),Y instruction doesn't actually exist! If you are cool with undocumented opcodes you can replace that with LAX (labs0),Y which does exist. Otherwise it would need another 2 cycles for TAX. Yes I wrote, based on mult65.a credited to Nick Jameson. That is one of the better ones; I took a look and thought about how it could be sped up a little.
- djmips 3y agoHaha, just like chatgpt - that's the same mistakes it makes