View Full Version : TP Follow Spline???

11 November 2005, 07:55 PM
Need some help here guys,

I need to try and recreate the following image;

Stars (

I know this has been discussed loads of times here. I've read through them
but I can't see where I'm going wrong. The star shapes need to be different sizes etc.

Here's where I am so far, where am I going wrong? :banghead:

Stars C4d (

I'm trying to pick up TP but it's kind of tricky. Wish I'd paid more attention in Maths class!



11 November 2005, 08:42 PM
On a first glance you need to check that in both the emitter and the follow spline options you put Particle Groups in "Group" and Objects in "Shape". You have that mixed up.

Go to Plugins-ThinkingParticles-Options to bring up the box where you can add Particle Groups. Then drag and drop "All" in all "Group" fields. Then drag and drop your star object in the "Shape" field of the Basic Emitter.

When I find some time I'll work on the file to give you different star sizes and orientation.

11 November 2005, 08:46 PM
If this doesn't have to have z depth, Trapcode Particular for After Effects is 99 times faster than TP. It's a brilliant, brilliant tool.

11 November 2005, 08:54 PM
I've had this problem. The way Srek built this was cool--as the particles near a point, they're attracted by the next point. That's how they follow the spline. But the thing that happens to me is that they tend to 'clump' around the points, then separate, then 'clump' again.

TimC has an example of another method on his site ( called 'Basic TP Flocking.' That's it in a nutshell.

The way its done is by building an invisible tubular poly object (use a converted sweep) and making that a collision object. It behaves like a 'pipeline' for the particles. Then send a null along that path of yours with an align-to-spline. Make the null a pgravity object or a pfollow (? I forget) in your TP settings. For additional variations you can connect a bunch of low-poly spheres to the pipeline--they'll act like rocks in the middle of a river rapid and keep it all from being too regular.

11 November 2005, 09:28 PM
I've had this problem. The way Srek built this was cool--as the particles near a point, they're attracted by the next point. That's how they follow the spline. But the thing that happens to me is that they tend to 'clump' around the points, then separate, then 'clump' again.

This should not occur if you increase the Precision value. If the TP Follow SPline setup does not fill the need, there are two other setups for something similar on my website

11 November 2005, 09:35 PM
Hi Tony,

I know this has been discussed loads of times here. I've read through them
but I can't see where I'm going wrong. The star shapes need to be different sizes etc.

Open the TP Basic Emitter Xpresso setup, click on the PStorm node and change the "Size Variation" Parameter to your likeing, maybe something around 80%.

Here's where I am so far, where am I going wrong? :banghead:

The "Point Velocity" value is to high, set it to 0.01 and see if it fits your needs better.

Hope this helps

11 November 2005, 09:58 PM
Yup It's official. I'm a dummy.

Tried to follow your instructions Creature, but still having trouble.

Shrek, your comments on Size Variation worked at least in the preview.

However, when I render a frame the geometry isn't rendered?

Help??.... :cry:

11 November 2005, 10:01 PM
Sorry, forgot that. You dropped the shape in the wrong linkfield of the emitter.

11 November 2005, 10:02 PM
Here's a fixed file which displays the geometry:

Look in the Emitter and Spline Follow parameters for what I've changed

11 November 2005, 10:33 PM
Thanks Creature,

One thing I do notice is how the particles seem to increase speed after the first few points
of the spline. Is this to do with the distance between points on the spline?
Or is it a parameter I need to change?

Once again thanks! :beer:

11 November 2005, 06:11 AM
This does not happen here if i apply the changes i mentioned. If you don't chenge the Point Velocity then this is completely normal, since the particles are drawn tightly to the shape of the spline in the first few frames.

11 November 2005, 09:41 AM
Hey Shrek,

Thanks for your advice. I now have my stars flying along the spline nicely.
I've even animated the point velocity so the stars grab onto the spline less as time goes by.

Just one more thing i'm having trouble with! I've tried changing the SPIN parameter
in PStorm thinking it would rotate the star shapes as they move along but it doesn't seem
to be working? Am I missing something?

What is a shame with Cinema is I'd like to add motion blur to the stars but some effects
don't work on higher resolutions... :rolleyes:

Thanks again,

TonyG :)

11 November 2005, 10:00 AM
The FollowSpline XPresso is placing the particles on the spline and aligns them to the spline every frame - so that overrides the Spin value of the Emitter.

11 November 2005, 12:51 PM
Hey Creature / Shrek!

Sorry to drag this on! :cry:

I've tried a different approach. The stars aligning to the spline didn't make the stars random
enough for me.

Check out this file if you've time please;

Stars on Spline (

I've animated the emitter size to spread the stars wider as they move along the spline
but I'd like to animate the star size to increase as the emitter gets closer to the camera.

I tried keyframing the endsize to increase at a certain point, but it effects the stars (particles) that have already been 'dropped'? Any ideas please?

Thanks for all your help.


11 November 2005, 02:10 PM
I've probably used up all the goodwill for people to help out here but I kind of solved the problem of star size myself. Looked at all the parameters when something occured to me that was staring me right in the face!

Just bring the spline points nearer the camera! Doh! :banghead:

There you go.

Thanks for all your help guys!


11 November 2005, 02:29 PM
I changed it back to Thinking Particles (because you have a lot more controll over TP) and put a file together which should do everything you want.

All important controlls are added as a HUD:

11 November 2005, 03:10 PM
Creature, you really are a know-it-all!!!

That's brilliant thanks a lot. I'd like to know how you created the hud controls.

I've taken a look and edited the limits (min/max) that you provided using edit-userdata
but creating them would really help with animation!

The reason I had the stars following the spline originally, was that I need to create realistic
motion blur as if the stars we zooming along the spline. There's no way to incorporate this into this scene is there?

Once again thanks a mill. If there's anyway I can help you out. Let me know! :applause:


11 November 2005, 03:29 PM
When you click on a Null-Object (for example - this works with every other object) you can see the User-Data in the Objectmanager (where you edited the limits of my UD). Just click User-Data > Add UserData to add a new one. To add an existing UD field to your HUD just rightclick on existing data fields and choose "Add to HUD". You can move the new hud field around in your viewport by CTRL-clicking on it.

I don't think that it is possible to put MotionBlur on the stars with this setup. Since the emitter is following the spline the stars don't have a real velocity that can be used to calculate the Motion Blur.

If you want MotionBlur on the stars I guess we need to start working on a different solution ;) Emitting particles in an invisible tube (shaped by the spline) acting as a deflector object should be a promising aproach I guess.

11 November 2005, 05:42 PM
Hey Creature,

Had a play around with some of the stuff I've learn't from you guys in the past while.

I'm amazed I've got to where I am now. Gotta do a bit more but hey, that's how you learn I suppose!

Check it out;

Tube (

Cheers, :beer:


11 November 2005, 06:19 PM
OK, I took a look at the tube file - you set it up correctly and I played around a bit with it but I think it does not give us the result we are after.

So here is something different which looks quite good imo. I used several SplineFollows to get a wider "beam" and to catch particles which go astray (they tend to loose contact to the spline when the values in one SplineFollow get too low)

Take a look:

But be carefull when you edit settings. This setup is quite sensitive to any changes of the values of anything. You won't call this stable or elegant - but it works.

11 November 2005, 09:05 AM
Hey Creature,

Found your file v. interesting. I've another question for you!
I tried something similar a while ago (though not as successful).

Your Group object SPLINE FOLLOWS with 5 TP FOLLOW SPLINE groups, could you create a similar effect with 1 TP FOLLOW SPLINE but all 5 of the SPLINE objects connected?
Or have you created it with 5 Splines so you can ajdust the point velocity on each spline?

I think i'm slowly starting to get me head wrapped around this... I think!


11 November 2005, 10:24 AM
Joining the multiple splines into one and using just one TPsplineFollow is an interesting idea - haven't thought of that. I just tried it out and it works to some extend. But the particles are more likely to loose contact to the path this way. I did not get a good result which I can post here but I guess it just lies in fiddling with the settings of the TPsplineFollow.
Edit: BTW - what I did is joining the 5 splines together to one. Select all 5 splines and use Functions > Join. I put this new spline in the TP

The only reason I set it up with multiple Follows is because it was just a quick thought that it might work catching runaway particles. And simply dublicating the TPsplineFollows was a quick way to test it.

But I guess an advantage of having multiple SplineFollow objects and not just one is that you can create a center spline which has lower precision values and "boundary-splines" with higher settings to prevent particles from loosing track.

11 November 2005, 11:32 AM
Hey Creature,

I'm still here! ;) I have the weekend to work on this.
(Until the girlfriend has me taking down more wallpaper!)

The TP stuff is starting to click now. I'm starting to get my head wrapped around
initiators, generators and conditions, I think!

Anyhow, I was playing around and came up with this;

test2 (

I basically started afresh after playing around with TP generators.
I added PScale attached to an AGE variant, so the Stars Size increases as time passes.
I added PSpin which allows the stars to Spin on one axis, which provides the randomness
I needed for the final image.

What I can't understand is, if the the stars are following the spline, I wonder if there is a way to have them banking as they turn the bends in the spline? I can't understand why this doesn't happen?

I'd like the stars to disperse more when the get past the final bend in the spline.
Keyframing the point velocity will affect all of the stars, I'd like to attach a Particle
Age condition but I can't seem to get it too connect between the terminals.

This new version is closer to the original image I need to create and also allows me to use motion blur but I'll worry about that later.

Just wondered what you thought?


TonyG. :thumbsup:

11 November 2005, 12:10 PM
I have no idea about banking. Seems a banking function isn't there in the pre-made TP Follow Spline so I guess you have to do it on your own :) An approach could be to get the Z- and X-Position and change the banking of a particle when both positions change. Would mean a bit of manual coding but I think it could work.

Same with spreading the beam of stars wider when they finished the final turn. Get the Z-Position of particles and once they are beyond a certain value let them be affected by something. Maybe some Wind could stir them up so that they aren't holing on to the spline as thightly as they did before? Or cou could transfer the particles which cross a certain z-position into a new group. Dublicate the TP Follow Spline, only set new Point Velocity and Precision values and let if affect only your new Particle Group. That should work I guess. (make sure your old TP Follow Spline does not affect "All" Particle Groups. Make a custom group for "before final bend" too. So that the two TP Follow Splines are clearly seperated from each other)

CGTalk Moderation
11 November 2005, 12:10 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.