Skip to main content

Still trying to figure out how to atan2 in assembly for #LoveByteTCC. Translated some c64 routines that gets me close, but still ways away.
#lovebytetcc
in reply to Devine Lu Linvega

tomas
tomas
• •
doing this with 16-bit integers seems like a very difficult problem

that said, it's also a very very cool thing

you could apply the same function for reading a joystick signal for example

or create little boids with steering behavious that fly around

atan2 is very cool
in reply to Devine Lu Linvega

Devine Lu Linvega
Devine Lu Linvega
• •
Now we're getting somewhere.
in reply to Devine Lu Linvega

Devine Lu Linvega
Devine Lu Linvega
• •
It's too hard for me, I've never been good with trigonometry and this one is kicking my butt. I'll wait for tomorrow's #LoveByteTCC puzzle.
#lovebytetcc
in reply to Devine Lu Linvega

Philip Nelson
Philip Nelson
• •
trif is the reason I didn't get the grade in calculus I needed (a C-) during school and I had to take it three times
in reply to Philip Nelson

Devine Lu Linvega
Devine Lu Linvega
• •
@blu3r4d0n the only time I had geometry classes was in high school and I wasn't paying attention 🤦‍♀️
@Philip Nelson
in reply to Devine Lu Linvega

テオ
テオ
• •
as much as i admire your ability to make cool stuff i have an even greater respect for your ability to openly and publicly admit your limitations. 🫡
in reply to Devine Lu Linvega

YRabbit
YRabbit
• •
it looked like a stereo picture and I even tried to squint my eyes, but nothing showed🙂
in reply to Devine Lu Linvega

Marquee Design
Marquee Design
• •
atan2 in (6502) assembly is tricky . x*constant / y can sort of approximates it a bit (it is used in the smaller sized variants of today's challenge)
in reply to Marquee Design

Devine Lu Linvega
Devine Lu Linvega
• •
@marqueedesign I went for something like this:

https://raw.githubusercontent.com/dustmop/arctan24/master/arctan.asm

where the screen is divided in 24 segments using bit shifting to find each segment of a quadrant, I could divide in much finer details (16 segments per quadrant). And I managed to draw something like each quadrant, but then I need to find the distance and.. honestly the code is so gross already that I might as well do something else-
@Marquee Design
in reply to Devine Lu Linvega

Lovebyte Demoparty
Lovebyte Demoparty
• •
Ouch, your environment is giving you a hard time! If you are willing to cheat: math.atan(y/x) is "approximately" the same as y/x (sizecoding trick courtesy of Pestis) . Of course it isn't a great approximation, but you'll end up with something looking like this:
in reply to Devine Lu Linvega

Lovebyte Demoparty
Lovebyte Demoparty
• •
More pointers from the pros:

"instead of r² = x² + y² leaving out the square root also works -> r = x² + y² when done well, the difference is almost not noticeable" - Hellmood

"atan2 is tricky to simulate... trust me i tried" - Superogue
in reply to Lovebyte Demoparty

Devine Lu Linvega
Devine Lu Linvega
• •
@lovebyteparty you might have given me a bit of a hint, lemme try this once more.

back to the drawing board
@Lovebyte Demoparty
in reply to Devine Lu Linvega

Devine Lu Linvega
Devine Lu Linvega
• •
@lovebyteparty I think that if I can test the distance against a sin table, then I should have something that looks like perspective! Thanks for the help with this ^__^
@Lovebyte Demoparty
in reply to Devine Lu Linvega

Lovebyte Demoparty
Lovebyte Demoparty
• •
Hopeing it helps to know that everyone agrees that what you are trying to do is -hard-, anything you manage to pull off is cool! 💪

Lo, thar be cookies on this site to keep track of your login. By clicking 'okay', you are CONSENTING to this.

⇧