shatter help


#421

Ruch,

Just use a stand in for your geo that you want to make passive. And turn your visibility off. Your current mesh might be too dense, so when you make it concave you may run into problems. Hope this helps.


#422

Hey tokanohanna thnx…it’s working fine now…Enjoy


#423

Hello PPL,last night i tried to find script to pre shatter by giving ready black & white texture but nothing found.So if anybody have this kind of script(not plugin) then plz put here.And if this kind of script is not Available then plz try to make if u r good @ scripting bcz this will save some more time as compare to voronoi script.Thanx


#424

hi ruchit, well am sure you can get some sweet mesh out of a trial blastcode plugin… delete the engine and use your image based broken pieces… although it is restricted but with patience you can manage some good stuff. as far as a script is concerned, am pretty sure its complicated as it would need to take care of some serious polygon mesh and not just a simple nurbs or a polyplane…this is for a texture/image based break … hope this is what you meant by black and white texture.

maybe you can also try some other techniques like polycutter and stuff…


#425

Hello i already told i don’t want to fall in Any plugin And blastcode is like bad output.Anyway thnx for ur reply.


#426

ruchitinfushion, without a plugin this isn’t possible. Well, it could be possible though the python API but it would be so slow it would be unusable. Sorry.


#427

Hi everyone,

I was not able to play a bit with the dg_voronoi since it got released, and finally today, I installed it and was really excited about it.
But since I have started using it, it never seems to complete the shattering of the full geometry.

I get a lot of “# Warning: Cannot perform boolean operation”.

Even with a fresh new cube, and cleaning up the locators (deleting the locators too near to a face) does nothing in bettering the result.

I am quite surprised and I am thinking that there is something that I am missing here. I don’t see anything I should have done in another way, would someone have a suggestion please?

Thanks!
Xavier.


#428

Allright so it seems to be working perfectly at home on my mac laptop, go figure.
I am using x64 linux at work.

This script is very nice, I like it!

I was thinking if there was a way of adding more octaves to the voronoi noise so that the pieces created could get more detail. It seems that the voronoi noise gives nice convex shapes and I guess I need to mix in another shattering technique to get concave pieces.


#429

New user story on fracture-fx.com

Read how Doug Holder completed a destruction shot with Fracture from tracking to final render:
http://www.fracture-fx.com/Doug_Holder


#430

Interesting read. I should really try to get into the beta for fracture!

Alex


#431

5000 pieces generated with dg_voroPy in 9h 25m 22s.


YouTube video HD720


#432

5000 pieces generated with dg_voroPy in 9h 25m 22s.

Is that 9 HOURS and 25 minutes for the simulation?


#433

No, it was 9h 25m for generation 5000 objects with python script dg_voroPy.
Dynamics simulation with Physx was 3-4sec per frame.


#434

No, it was 9h 25m for generation 5000 objects with python script dg_voroPy.
Ah. Thank you for the clarification.

FYI: shard generation in Fracture in practically instant. As in: on-the-fly :wink:

www.fracture.fx.com


#435

Just to show the difference in speed between a script vs a plugin, using the Fracture plugin for 5000 pieces only took 33 seconds. :buttrock:


#436

It slow because dg_voroPy is runining only on 1 core on my 8 core machine, because of python.

Here is time test on 2 machines. With python only 1 core working on both.

But, i mean it is possible to make it more fast with PyCUDA - http://mathema.tician.de/software/pycuda

  [img]http://img691.imageshack.us/img691/8486/dgspeed.jpg[/img]

And here is 50,000 pieces in just under 15sec as plugin.


#437

AtrusDni,

Not sure how you timed Fracture but my results are far quicker.

Anyways Fracture is only single threaded, at this time, but still only took 2.24 seconds for 5000 and 22.33 seconds for 50000.


#438

It could very well be the machine that I am on. It is a few years old, and I was running 2 Maya’s at the same time (>.<)

Needless to say, its FAST. Very very fast.


#439

I did notice it took the viewport about 30 seconds to refresh, tho Fracture calculates within a few seconds, and that may be what you are referring to.


#440

That could also be the reason. Needless to say Fracture beats the scripted approach hands down.