Page 2 of 3

Affector?

Posted: Sat Jan 31, 2009 10:50 pm
by 3DModelerMan
Isn't this for nodes? I thought affectors were for particles. Why not an animator?

Posted: Sun Feb 01, 2009 8:12 am
by arras
Isn't this for nodes? I thought affectors were for particles. Why not an animator?
Right, animator, not affector.

Posted: Mon Feb 02, 2009 9:18 am
by fmx
hmmm
thanks for all your opinions, I think I know where I'm going with this now.

Irrlicht simplicity >>> Physics-API familiarity

That should be the defining slogan of this "project", and so its going to morph into an Animator or something just as daftly simple.

Update on its way ;)

Posted: Mon Feb 02, 2009 10:35 am
by geckoman
Yeeha :D

Posted: Tue Feb 03, 2009 10:39 pm
by fmx
I just updated the first post with the latest version of irrSimplePhysics v0.002 (link: http://www.mediafire.com/?mxyyy3tthhi )

its NOT an Animator, just a cleaned up Class that I was half-way through before I decided to make an animator from it.
I'm thinking of keeping the Class and instead providing additional Animator functions too, best of both worlds ;)

if anyone tries it then please do leave some feedback!

time

Posted: Sun Feb 08, 2009 6:39 pm
by 3DModelerMan
Does this have time based movement?

Posted: Sun Feb 08, 2009 7:00 pm
by fmx
its supposed to be fps-independant, but I think it might be messed up actually :oops:
I just figured out how to achieve TRUE fps-independance for physics (you'd think i would have know it already >_>) and why PhysX code has a different mechanism for invoking its update-function.

but anyway
give it a try and see how it works for you. I'll be updating this in a day or two anyway, so I'll add your requests and fixes etc, too.

What kind of physics do you need for your game?
Characters, vehicles, projectiles, terrains...?
maybe I could tailor it specifically for your game first, then branch it out later to make it more "generic" ;)


btw
geckoman / trivtn / others : what about you guys? what kind of physics do you require for your games?

collision

Posted: Sun Feb 08, 2009 7:16 pm
by 3DModelerMan
I only really need time what you have here but true frame reate independent. And collision detection.

Re: collision

Posted: Sun Feb 08, 2009 7:28 pm
by sudi
3DModelerMan wrote:I only really need time what you have here but true frame reate independent. And collision detection.
Use a physic engine. like newton/bullet/PhysX

Posted: Sun Feb 08, 2009 10:34 pm
by fmx
collision-detection might be a pain.
Sudi's right, don't give up trying real physics-engines until you find something which works :)
(ie, use this as a last resort, or PM me and i'll try find out why something isn't working for you)

have you tried Newton? IMO its by far the easiest and simplest Physics-engines to get going with Irrlicht.

Posted: Wed Feb 11, 2009 8:21 pm
by Pyritie
Noooo, this thingy is lovely :D

Posted: Wed Feb 11, 2009 10:42 pm
by geckoman
I need either a simple leightweight physics-engine that is capable of doing some explosion thing
or
something like irrphysx to get the extra-speed from hardware-using, but I'd prefer leightweight, as it's less work for me :D

Posted: Thu Feb 12, 2009 8:36 am
by JP
What do you mean by lightweight and why does it need to be lightweight? Not that i'm trying to force you to use IrrPhysx or anything, just curious!

Posted: Thu Feb 12, 2009 12:07 pm
by geckoman
By lightweight I mean, as fast as possible, without much overhead so that my game runs really fast on older computers
AND
I mean very easy to use and hopefully not much time needed to implement.

But as I said before, if I don't find such a thing, I will very likely use irrphysx :D

Posted: Thu Feb 12, 2009 12:54 pm
by JP
Hmm well physics calcs are often quite expensive.. i don't know if FMX's calcs are quicker than Physx or not but I'm guessing he might not have the same accuracy and complexity as Physx provides in their calculations. Have you tried IrrPhysx on any old computers to see how it runs?