MetaBall Duplivert Unwanted Movement Bug #41010

New Issue

Jimmy Gunawan · 2014-07-10T00:33:49+02:00

Jimmy Gunawan commented

2014-07-10 00:33:49 +02:00

System Information
Windows 7 64 bit

Blender Version
Broken: (example: 2.69.7 4b206af, see splash screen)
Worked: (optional)

Short description of error
Dealing with Metaball duplivert, parented to path with build modifier. Metaball keeps moving.

Exact steps for others to reproduce the error
Based on a (as simple as possible) attached .blend file with minimum amount of steps

It's really quite simple: I wanted to duplivert Metaball object, hoping it to create this "grow" kind of look. However, the Metaball result when combined with path + Build Modifier is really weird. The Mesh keeps moving around in 3D space. This should not be the default behaviour, I don't think.

**System Information** Windows 7 64 bit **Blender Version** Broken: (example: 2.69.7 4b206af, see splash screen) Worked: (optional) **Short description of error** Dealing with Metaball duplivert, parented to path with build modifier. Metaball keeps moving. **Exact steps for others to reproduce the error** Based on a (as simple as possible) attached .blend file with minimum amount of steps It's really quite simple: I wanted to duplivert Metaball object, hoping it to create this "grow" kind of look. However, the Metaball result when combined with path + Build Modifier is really weird. The Mesh keeps moving around in 3D space. This should not be the default behaviour, I don't think.

Jimmy Gunawan commented

2014-07-10 00:33:49 +02:00

Changed status to: 'Open'

Jimmy Gunawan commented

2014-07-10 00:33:49 +02:00

Added subscriber: @BlenderSushiGuy

Bastien Montagne commented

2014-07-10 08:34:45 +02:00

Added subscriber: @mont29

Bastien Montagne commented

2014-07-10 08:34:45 +02:00

Please always attach a .blen file illustrating the issue.

ronan ducluzeau commented

2014-07-10 14:23:10 +02:00

Added subscriber: @zeauro

ronan ducluzeau commented

2014-07-10 14:23:10 +02:00

I confirm that it gives something weird.
I am surprised that it gives something.
Metaball are not working with duplifaces but within this weird offset with dupliverts.

See attached.blend.
Mball_dupliblend.blend

@BlenderSushiGuy
I recommand you to always use particles for metaball.
It is easy to use a Curve Guide to make them follow a path.
Mball_CurveGuide.blend

I confirm that it gives something weird. I am surprised that it gives something. Metaball are not working with duplifaces but within this weird offset with dupliverts. See attached.blend. [Mball_dupliblend.blend](https://archive.blender.org/developer/F97317/Mball_dupliblend.blend) @BlenderSushiGuy I recommand you to always use particles for metaball. It is easy to use a Curve Guide to make them follow a path. [Mball_CurveGuide.blend](https://archive.blender.org/developer/F97318/Mball_CurveGuide.blend)

Jimmy Gunawan commented

2014-07-11 13:35:42 +02:00

@zeauro thanks for the CurveGuide example for Metaballs.

It can be nice to have Metaballs working with Dupliverts. If not too slow that is.

Can I also add snippet weirdness I found about Blender Particles?

When emitted from Vertices, let say just plane with 4 vertices, I am getting Particles emitted from vertices index: 0,0,0,0,1,1,1,1, 2, 2,2,2,3,3,3,3.

Why can't it be more like 0,1,2,3, 0,1,2,3,0,1,2,3.

I am experimenting with Blender Particles and Sverchok. I like more control over the order index.

@zeauro thanks for the CurveGuide example for Metaballs. It can be nice to have Metaballs working with Dupliverts. If not too slow that is. Can I also add snippet weirdness I found about Blender Particles? When emitted from Vertices, let say just plane with 4 vertices, I am getting Particles emitted from vertices index: 0,0,0,0,1,1,1,1, 2, 2,2,2,3,3,3,3. Why can't it be more like 0,1,2,3, 0,1,2,3,0,1,2,3. I am experimenting with Blender Particles and Sverchok. I like more control over the order index.

ronan ducluzeau commented

2014-07-11 14:11:30 +02:00

It adjust amount of particles fora continuous emission from first index to last.
Using index faces, it can be used for desintegration with explode modifier from a side of a mesh to another.
A looped emission would not work well for that.
It would imply that user should be rigourous to create explosion by emetting as much particles than faces in geometry.
And users who do explosions are not rigourous.

I don't know Sverchok.

Originally, a goal of 2.5x particles system implementation should be to offer conversion of caches into action strips that could repeated or mixed.
But it was not reached. Lukas-t is rewriting current particles system implmentation using nodes since 2011.

As a workaround, you can be rigourous and use Trail count. It has a "length in frames" option.
So you can only emit 4 particles in a short time and adjust "length in frames" option according to trail count and emission time.
Of course, trail repetitions are not sensible to physical changes. It is just duplicates of first 4 particles, not real particles.
But lukas' system should be able to do that.

It adjust amount of particles fora continuous emission from first index to last. Using index faces, it can be used for desintegration with explode modifier from a side of a mesh to another. A looped emission would not work well for that. It would imply that user should be rigourous to create explosion by emetting as much particles than faces in geometry. And users who do explosions are not rigourous. I don't know Sverchok. Originally, a goal of 2.5x particles system implementation should be to offer conversion of caches into action strips that could repeated or mixed. But it was not reached. Lukas-t is rewriting current particles system implmentation using nodes since 2011. As a workaround, you can be rigourous and use Trail count. It has a "length in frames" option. So you can only emit 4 particles in a short time and adjust "length in frames" option according to trail count and emission time. Of course, trail repetitions are not sensible to physical changes. It is just duplicates of first 4 particles, not real particles. But lukas' system should be able to do that.

Jimmy Gunawan commented

2014-07-11 14:23:37 +02:00

Cheers for hint and suggestion.

Blend example if you tried Sverchok:
https://www.dropbox.com/s/rz5zfa4972p816q/particle_knot_nikolaus.blend

Anyway, trying to get certain Particle emission.

Cheers for hint and suggestion. Blend example if you tried Sverchok: https://www.dropbox.com/s/rz5zfa4972p816q/particle_knot_nikolaus.blend Anyway, trying to get certain Particle emission.

Bastien Montagne commented

2014-07-18 14:40:39 +02:00

Yet, there are some weird things going on with mballs + dupli… Will try to have a look at it.

Bastien Montagne self-assigned this 2014-07-18 14:40:47 +02:00

Bastien Montagne commented

2014-07-18 17:36:00 +02:00

DevNotes: So far, I have the feeling MBalls make make_child_duplis_cb() being called for each 'virtual' children generated by dupli, instead of just being called once per real child.

Code says everywhere MBalls have specific handling of duplis… funny part is, I can't figure out where it is currently! :P

DevNotes: So far, I have the feeling MBalls make `make_child_duplis_cb()` being called for each 'virtual' children generated by dupli, instead of just being called once per **real** child. Code says everywhere MBalls have specific handling of duplis… funny part is, I can't figure out **where** it is currently! :P

blender-admin commented

2014-07-19 00:41:13 +02:00

This issue was referenced by e0d8e62f2b

This issue was referenced by e0d8e62f2bb97d75088f7f1fd9f0fab578a02410

Bastien Montagne commented

2014-07-19 00:42:55 +02:00

Changed status from 'Open' to: 'Resolved'

Bastien Montagne closed this issue

2014-07-19 00:42:55 +02:00

Bastien Montagne commented

2014-07-19 00:42:55 +02:00

Closed by commit e0d8e62f2b.

Closed by commit e0d8e62f2b.

Sign in to join this conversation.

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

MetaBall Duplivert Unwanted Movement Bug #41010