Flash Physics Engine

SnakeO · 04-21-2004, 03:13 PM

I am currently working on a flash physics engine. It is in its beginning stages. If anyone would like to help, please email me :

jake AT hybridstudio DOT com.

Thanks
-Jake

jbum · 04-21-2004, 06:08 PM

Hiya Jake,

Before sowing seeds of doubt I should probably say that the activity of writing your own Physics engine is a really cool thing to do, and you will probably learn a lot from it. That reason alone is enough to make it a worthwhile project.

The following comments are more about the utility of such an engine.

I read an essay at gamasutra about the tradeoffs involved with general-purpose physics engines. I can't find it at the moment, but basically the gist of the article was that using a general-purpose Physics engine often makes your code less efficient (slower) than directly incorporating the physics into your code.

There are two reasons for this. One is that general-purpose physics engines tend to not 'cheat', and cheating tends to make code faster. For example, if you know that you have a fairly constant frame rate, you can usually factor time (dt) out of most equations, without any noticeable artifacts.

The other is that a general-purpose physics engine doesn't know about your game the way you know about your game, and will fail to optimize the code as well as you would optimize it if you wrote it yourself. For example, if you call two different methods in a physics engine, they might compute quantities separately which are the same, and could be reduced to a single computation.

I think that for Flash, these observations are particularly relevant, since Flash isn't a speed-demon to begin with.

I've been screwing around with making a Spring physics class (mostly as a learning exercise), but I'm not likely to use it for a simple springy-tween on a button, 'cause it's overkill.

Finally, I wanted to ask you what kind of applications you envision your Physics engine being used for? A physics engine used for games might differ heavily from one intended for making springy widgets, or one intended for physics demonstrations.

What kinds of physics do you want to offer?

SnakeO · 04-21-2004, 06:36 PM

I emailed you directly.

Please check your emails. I got the link from your site.

Thanks
-Jake

The Xing · 05-05-2007, 06:52 AM

Im also starting a physics engine are you interested in working together?

shipstern · 05-06-2007, 05:24 AM

Hi Jake,

What sort of integrator and collision detection are you using. Are you talking 3D with 6 DOF or only 2D?

04-21-2004, 03:13 PM	#1
SnakeO Junior Member Join Date: Nov 2003 Posts: 18	Flash Physics Engine I am currently working on a flash physics engine. It is in its beginning stages. If anyone would like to help, please email me : jake AT hybridstudio DOT com. Thanks -Jake

04-21-2004, 06:08 PM	#2
jbum Senior Member Join Date: Feb 2004 Location: Los Angeles Posts: 2,920	Hiya Jake, Before sowing seeds of doubt I should probably say that the activity of writing your own Physics engine is a really cool thing to do, and you will probably learn a lot from it. That reason alone is enough to make it a worthwhile project. The following comments are more about the utility of such an engine. I read an essay at gamasutra about the tradeoffs involved with general-purpose physics engines. I can't find it at the moment, but basically the gist of the article was that using a general-purpose Physics engine often makes your code less efficient (slower) than directly incorporating the physics into your code. There are two reasons for this. One is that general-purpose physics engines tend to not 'cheat', and cheating tends to make code faster. For example, if you know that you have a fairly constant frame rate, you can usually factor time (dt) out of most equations, without any noticeable artifacts. The other is that a general-purpose physics engine doesn't know about your game the way you know about your game, and will fail to optimize the code as well as you would optimize it if you wrote it yourself. For example, if you call two different methods in a physics engine, they might compute quantities separately which are the same, and could be reduced to a single computation. I think that for Flash, these observations are particularly relevant, since Flash isn't a speed-demon to begin with. I've been screwing around with making a Spring physics class (mostly as a learning exercise), but I'm not likely to use it for a simple springy-tween on a button, 'cause it's overkill. Finally, I wanted to ask you what kind of applications you envision your Physics engine being used for? A physics engine used for games might differ heavily from one intended for making springy widgets, or one intended for physics demonstrations. What kinds of physics do you want to offer? __________________ Last edited by jbum; 04-21-2004 at 06:15 PM.

04-21-2004, 06:36 PM	#3
SnakeO Junior Member Join Date: Nov 2003 Posts: 18	jbum I emailed you directly. Please check your emails. I got the link from your site. Thanks -Jake

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

05-05-2007, 06:52 AM	#4
The Xing Pushing the boundaries>> Join Date: May 2007 Location: South Africa Posts: 8	Im also starting a physics engine are you interested in working together?

05-06-2007, 05:24 AM	#5
shipstern Senior Member Join Date: Nov 2003 Posts: 524	Hi Jake, What sort of integrator and collision detection are you using. Are you talking 3D with 6 DOF or only 2D?