Game Development Community

Mercury Particle Engine vNext and GUI

by raa brubb · in Torque 3D Professional · 02/10/2014 (3:40 pm) · 5 replies

Hey all,

Lately, I've been looking for a great particle system to improve Torque3D. After almost no search I stumbled upon Mercury Particle Engine vNext. It works just like PhysX particles but it is open source. I was thinking we could integrate it into the project dll, merge it with the wind emitter (for wind effects) and add it to the particle editor. I don't know yet if it's possible but let's be honest. We need a better particle engine. Anyways, what is your opinion, should anyone spend time trying to integrate this?

#1
02/10/2014 (4:26 pm)
Go for it! More options are always better than fewer options!
#2
02/10/2014 (7:04 pm)
Is it a rendering engine or a particle physics/behavior engine? Or both? I watched the demo video and looked at the github but it seems to lack any kind of explanation of what it is as an entirety and what it's goals are.

I was just thinking, we're working on PhysX 3.3 particles, but while doing so perhaps it might be a good idea to make the particle system work similar to how the physics system works. A plugin system allowing easy switching between regular torque driven particles and a third party particle controller. This would allow for physx, bullet, or perhaps this, to take the place of torque driven particles as well. This really only covers behaviors, interactions, positions, etc. though.
#3
02/10/2014 (10:36 pm)
1. There is no license on the Mercury Particle System GitHub page.
2. It sure looks a lot like that is 2D only, might be wrong on this I just didn't find any examples of 3D-particles.
@Andrew I'll send you an update on that soon btw.

Oh yeah, the rendering system of the particle system is fine btw, it's just the simulation I guess..
The only thing these "New" particle engines does better than T3D's are that they are offloading the work to the GPU or integrate it directly into the physicssystem so that it can do a few more tricks.

Edit: Correction, that is definetly only for 2D systems.
#4
02/11/2014 (9:30 am)
End thread, Lukas has best comment.