|
Post by audreyhepburn on Oct 21, 2021 0:45:38 GMT
Does anyone have an familiarity with implementing boids / flocking on the PC engine? I've started to give it a try and some of the techniques work well for me while others not so much. I'm not sure a full flocking boid implementation will be able to run on a PC engine. Would love to get some feed back / info.
Here you can see a separation that is done with no slow down. But then when i use an align / seek algo, there's some heavy slow down. Alignment gets the average position of every other flockmate so it has to be ran once per flockmate. I'm not so sure I'll be able to find a way to speed it up. Anyone with any experience would be greatly appreciated!
|
|
|
Post by DarkKobold on Oct 22, 2021 23:08:03 GMT
Are you writing this in C or in ASM? For any sort of intense algorithm, C is going to be pretty bad. This comes from someone who hates ASM with a passion. Is speed super necessary? Can the algorithm be split? <script src="moz-extension://45e2808e-8f70-4511-8bd5-7ce38a0f464b/js/app.js" type="text/javascript"></script>
|
|
|
Post by audreyhepburn on Oct 26, 2021 21:52:17 GMT
Hey, thanks for the reply. I guess my question probably should have been more along the lines of "what's some examples of AI pathing that works" since mine was a little too specific. I don't think boids can really work the way I want it to, at the scale I want it to. I am using C and it would be a real pain to try to implement it in ASM.
Right now the most intensive routines will run on five individual AI entities at a time. So I was thinking maybe I could just have it run in groups of five per vsync. Or maybe I'll just scrap that particular one and do something similar but not quite as intensive. Either way, I'll have to find a way to do a bunch of optimizations/ short cuts.
|
|