particle deflection experiment - windows 13/02/2004

User-contributed CVS development builds. Please test and give feedback!

Moderators: jesterKing, stiv

leon
Posts: 103
Joined: Mon Oct 20, 2003 7:14 am

particle deflection experiment - windows 13/02/2004

Postby leon » Fri Feb 13, 2004 11:36 am

hi,

Hopefully this will work: I've never tried publishing a compiled version, so I'm not sure whether or not anyone else will be able to run this!

But anyway: I've been playing with particle deflection and if anyone's interested, here's a tuhopuu with the current code:

http://reblended.com/www/leon/thp_pdef.zip

You will need to unzip into a folder containing all the usual files, so make a copy of your tuhopuu folder, then overwrite the .exe with the file in the zip. If you're on another platform and can compile and patch yourself, there's a patch file here.

I've added another panel "Particle interaction" to the edit buttons for meshes, which can be seen when you are in object mode. It has two buttons: an on/off for particle deflection, and a slider for "surface damping" which controls how high particles bounce when they hit the surface (around 0.5 - 0.6 gives nice results depending on the angle the particle hits).

Any particles which hit the mesh should be deflected: but there's plenty still to do, so things don't work perfectly by any means! In particular....

- It really only works well for large squarish planes so far, and even then particles sometimes leak through. Animated surfaces will not work yet...

- Particles leak especially badly at joins between objects / faces....

- The surface normal is important: if particles don't deflect as expected, try flipping the normals of the face.

- You need to set the "particle keys" as high as possible (32) to get good results (and keep the particle life as low as possible)

- If you move the deflector or change the surface damping, you will need to hit "recalc all" in the particle effect.

- Try something simple first - like particles falling on a large plane. Experimentation is needed to get things working nicely!!!

Cheers

Leon

Monkeyboi
Posts: 561
Joined: Tue Nov 26, 2002 1:24 pm
Location: Copenhagen, Denmark
Contact:

Postby Monkeyboi » Fri Feb 13, 2004 12:47 pm

Interesting! It did seem quite buggy though, and as you say particles often fall through surfaces. They also often behave odly. I did a test where the particles started to get stuck scrubbing strangely up and down a surface untill they die. But it certinately is an interesting first try with this feature that has been requested again and again.

About the Particle Interaction panel: It takes up valuable Editbuttons space, and its size doesn't fit with other panels. There is so little in it, too.
I was thinking about other places to put it when it struck me it would be very logical if it were under Effects>>Particles. There could be a dropdown to flip between the following particles effects: Emitter and Deflector. In my oppinion that would seem more logical to the user, and also be cleverer spacewise.

-William

leon
Posts: 103
Joined: Mon Oct 20, 2003 7:14 am

Postby leon » Fri Feb 13, 2004 12:57 pm

Hi monkeyboi,

Thanks for taking a look! Can you post your test blend (the one with the particles scrubbing up and down the surface) somewhere?

I will spend some time this weekend on it, and strange problems like that sometimes make the bugs a little clearer.

I was thinking about the effects panel, but for now I'll leave it where it is - my idea was that in the future you'd have other settings as well, possibly attraction / repulsion etc.

Cheers

Leon

Guest

Postby Guest » Fri Feb 13, 2004 1:10 pm

This is awsome!!!
A huge step foward for the blender animators out there!
I am going to look closely each of your post :D

Keep up the good work !!!!

Guest

Re: particle deflection experiment - windows 13/02/2004

Postby Guest » Fri Feb 13, 2004 5:58 pm


Willy wrote:hi,

It's the first time I post here, :D
I've just tried your tuhopuu, for the "Particules Interaction";
it works good :)

I rendered a little animation that you can see at :
[url]http://www.zoo-logique.org/3D.Blender/animations/bounding_balls_[DivX].avi[/url]

However, your tuhopuu's size is 6.62Mo
against 4.11Mo for the 2.32.

Also, I find this good because it's enough hard working with the particules

I hope this message will appears like I want
and the little video will please you a little bit, leon

Friendly
Willy

dotblend
Posts: 72
Joined: Thu Oct 17, 2002 2:18 pm

Postby dotblend » Fri Feb 13, 2004 9:18 pm

indeed very cool!!!

i will try to post some sample file's and bugs
here @

http://www.dotblend.com/blenderbugs.htm

kencanvey
Posts: 104
Joined: Mon Jun 02, 2003 7:54 am
Location: Smethwick West Midlands UK

Postby kencanvey » Fri Feb 13, 2004 10:02 pm

Hi :shock: WOW works great for me, simple to get going as well, Well done, this really should get into the BF builds, great work mate.
:D

Ken

Eric
Posts: 192
Joined: Wed Oct 16, 2002 1:01 pm
Location: Sweden, Norrköping

Postby Eric » Fri Feb 13, 2004 10:24 pm

Dotblend: he said it will not work for animated faces yet. (looking at your sample 'n bugs page)

tmt
Posts: 13
Joined: Tue Oct 29, 2002 8:06 am

Postby tmt » Fri Feb 13, 2004 11:37 pm

First of all,it is nice to see that there is developement under the particle system to gain more realism. I like what I see. :D

Right, I gave couple test spinns to your deflector system, and it indeeds seems to work, but what I gathered it works like a baseball bat, which hits the particles back to the direction they came from, am I correct?

What I mean is in example smoke or fire would try to seek way around the obstacle and not be stroken back, so I gather I am asking are you developing particles to work like that?

But nice work, I hope to see updates on this projects. :D
At elYsiun I am known as Hippie

leon
Posts: 103
Joined: Mon Oct 20, 2003 7:14 am

Postby leon » Sat Feb 14, 2004 12:51 am

thanks for all the testing!

Dotblend - like Eric says, animated deflectors aren't supported yet (although you can animate the particle source). BTW, I had a look at the scene you made, and you will get much better results with the "keys" set to 32 rather than 8 as you have.

tmt - yes, that would be nice to have. I'm not sure of the maths for doing it though, but I'll have a think....

cheers

leon

nozzy
Posts: 41
Joined: Sun Oct 13, 2002 7:42 pm
Location: Stockholm, Sweden
Contact:

Postby nozzy » Sat Feb 14, 2004 9:25 am

Great work Leon, its works a treat, the particle leakage is very minimal you just had to keep an eye on
the the particle normals speed. Here's som duplivert cubes flowing nicely :)

Image

/nozzy

Guest

Postby Guest » Sat Feb 14, 2004 10:36 am

Great stuff! Keep up the good work:)


Return to “Testing Builds”

Who is online

Users browsing this forum: No registered users and 0 guests