Particle systems

[chris]

Here's a first attempt at making a meteor:

http://www.youtube.com/watch?v=O72u4Jxb2No&NR=1

I think it's quite a bit better than what has been possible in Celestia previously, but there's still room for improvement.

--Chris

[fsgregs]

Chris:

This is all wonderful. Your meteor contrail opens LOTs of educational possibilities, as does the potential for exhaust gases from rockets. I think kids would love to see meteors actually burning up as they streak through the atmosphere.

I am sure you are planning for a variety of transparencies in the plume effects in the ssc file. The ones in the videos are rather dense but I could see a need for some particle effects and smoke that would be mostly transparent gas.

Thanks again for your work.

Frank

[ElChristou]

Here's a first attempt at making a meteor...

Nice!
Could this system being used along with the comet code to render the tails?

[Cham]

Chris,

with your particle systems for Celestia, do you think we could design something like this (with animation) ?

BH.jpg

[chris]

Chris,

with your particle systems for Celestia, do you think we could design something like this (with animation) ?

BH.jpg

Not in its current state--there's no force model that would result in a spiral path for particles. The only force model I've implemented is a constant force along a fixed axis. For a spiral path, there need to be two forces: a central 1/r^2 force and a drag force opposite to the velocity vector. Is there an analytic solution to the equations of motion for such a system? That's a requirement of my particle system implementation (I don't see a way around this constraint for particles in Celestia.)

--Chris

[chris]

I am sure you are planning for a variety of transparencies in the plume effects in the ssc file. The ones in the videos are rather dense but I could see a need for some particle effects and smoke that would be mostly transparent gas.

Yes, the opacity is adjustable. Here's my attempt at reproducing the blue glow of an ion thruster:

http://www.youtube.com/watch?v=S5JgkD3H_Ac

In this example, I'm using a 2D Gaussian distribution for the initial positions of the particles.

--Chris

[BobHegwood]

For what it's worth, I think this is much better.
Still looks to be a bit thick in space, but what do I know?

Thanks, Brain-Dead

[chris]

For what it's worth, I think this is much better.

Well, the thrusters on Dawn are completely different than the ones on Cassini. So, the video isn't supposed to look 'better' than the Cassini better, just different.

Dawn uses ion thrusters powered by solar panels, whereas Cassini uses conventional chemical propulsion. An ion thruster generates thrust by ionizing atoms, accelerating them to high velocites either magnetically or electrically, and ejecting them from the spacecraft. More information and an image here: http://dawn.jpl.nasa.gov/mission/ion_prop.asp

--Chris

[ElChristou]

Nice!

[BobHegwood]

An ion thruster generates thrust by ionizing atoms, accelerating them to high velocites either magnetically or electrically, and ejecting them from the spacecraft.

Yes, I know... Exactly my point here.

[t00fri]

An ion thruster generates thrust by ionizing atoms, accelerating them to high velocites either magnetically or electrically, and ejecting them from the spacecraft.

Yes, I know... Exactly my point here.

Aha! I seem to remember you as a well-known expert on rocket models before becoming a Celestian!?

Cheers,
Fridger

[Fenerit]

Beautiful! Perhaps the Saturn's rings...

[BobHegwood]

An ion thruster generates thrust by ionizing atoms, accelerating them to high velocites either magnetically or electrically, and ejecting them from the spacecraft.

Yes, I know... Exactly my point here.

Aha! I seem to remember you as a well-known expert on rocket models before becoming a Celestian!?

Cheers,
Fridger

Perhaps I should elucidate on the comments above here...

What I meant was that the Dawn thruster looked better than the reproductions I had seen before. They always looked a bit too much like chemical reactions to me.
Also, I still think that this depiction is not quite what it should be, despite NASA's images. An ion thruster - it seems to me - would emit not a blue flame
expanding outward from the engine, but would look much more like light particles extending from the rear of the spacecraft.

Do you understand what I mean? Sorry to butt in again. I just wanted to make certain that you understood what I meant here.
Thanks, Brain-Dead

[t00fri]

...
Perhaps I should elucidate on the comments above here...

What I meant was that the Dawn thruster looked better than the reproductions I had seen before. They always looked a bit too much like chemical reactions to me.
Also, I still think that this depiction is not quite what it should be, despite NASA's images. An ion thruster - it seems to me - would emit not a blue flame
expanding outward from the engine, but would look much more like light particles extending from the rear of the spacecraft.

Do you understand what I mean? Sorry to butt in again. I just wanted to make certain that you understood what I meant here.
Thanks, Brain-Dead

Bob,

thanks for explaining. Unfortunately for this topic all I can refer to is my common sense. Not a bit more . You sure have more detailed know-how about this than me.

Fridger

[Chuft-Captain]

I am sure you are planning for a variety of transparencies in the plume effects in the ssc file. The ones in the videos are rather dense but I could see a need for some particle effects and smoke that would be mostly transparent gas.

Yes, the opacity is adjustable. Here's my attempt at reproducing the blue glow of an ion thruster:

http://www.youtube.com/watch?v=S5JgkD3H_Ac

In this example, I'm using a 2D Gaussian distribution for the initial positions of the particles.

--Chris

Sweet!



FWIW, I saw a video of a test-firing of a real ION rocket in a test-rig once. I'm a bit short on time right now, but if you'd like, let me know, and I could try and track it down again. (Or you could just try a Google search)

[buggs_moran]

Beautiful work Chris. I think this system (along with non linear path development) is critical in Celestia. The visualization possibilities are endless. Those videos on Youtube are beautiful. Will we be able to do non glowing particles as well?

BTW, if you only have one dimensional force motion, how did you do Tvashtar?

I wish I had more time to play, I've been away from Celestia so long...

[ElChristou]

Chris, you think to implement this in 1.6 or later?

[chris]

Chris, you think to implement this in 1.6 or later?

Later, I'm afraid. There are a lot of new features in 1.6.0 already, and it's time to focus on bug fixing and getting the release done. I will make a preview available soon and start soliciting feedback on ways to improve particle systems.

--Chris

[chris]

Beautiful work Chris. I think this system (along with non linear path development) is critical in Celestia. The visualization possibilities are endless. Those videos on Youtube are beautiful. Will we be able to do non glowing particles as well?

By non-linear paths, I take it that you mean animation support?

Non-glowing particles are possible. The smoke in the Buran liftoff is one example. For realism, very important thing will be the ability to specify some scattering properties for the particles. It could be as simple as specifying a phase parameter.

BTW, if you only have one dimensional force motion, how did you do Tvashtar?

In that example, each particle is affected by a constant force along the local z-axis of the particle system, resulting in parabolic motion. The motion of the particles would be more accurately modeled by using a central 1/r^2 force, but I don't think that anyone would notice the difference visually.

--Chris

[chris]

Here is another experiment with Tvashtar. The blue color has been toned down and I've tried to approximate the forward scattering of the plume matter by using a phase function to control particle brightness. Notice that the plume becomes much brighter when the Sun is nearly behind it:

http://www.youtube.com/watch?v=nJqkdbkOTLE

In fact, it probably becomes *too* bright. The phase function yields values much brighter than the maximum pixel. Without HDR, some sort of normalization term is probably required.

--Chris