RayFire Tool

#1161

Thank you sir!

#1162

Hi Mir-
I’m using version 1.43. There’s an “check for update” pull-down menu, but it says I’m up to date. I know 1.45 is out. How do I update?

Thanks
Alec

#1163

All new main versions are available via customer area.

#1164

We would like to thank everyone who was interested in RayFire at Siggraph this year!
You rock, you shatter, you blow away!

Talked to nVidia people briefly:
the 64bit physx SDK is on it’s way. PhysX64 is out by the end of the year hopefully :slight_smile:

Kind regards,
Anselm

#1165

End of the year, I was hoping for sooner. I hope they cook new features in it though :slight_smile:

#1166

Likewise it’d be great to ditch 32 bit max sooner than that.

Just as an aside folks, I’ve been working with rayfire for a few days now and I’m starting to get decent results - a few dumb questions would be handy for me though.

What are the really obvious things that kill physx / rayfire sims in terms of speed? I’m doing a lot of glass shatter stuff at the minute so as in Hristo’s videos I’m breaking with shapes first, then doing a few levels of fragmenting to get detail in the areas that the impact will happen. Is the amount of the objects the largest factor or things like size of pieces? For example I’ve a few setups with lots of very small bits (might be about 1700 objects making up the pane of glass) and I’ve fragmented them with voronoi and a rift. Am I best off using something like size threshold in the advanced options to stop my fragments getting too small? Also will things like the collision tolerance in physx make a big difference to speed? From Hristo’s video the bigger the value the less accurate the sim is - so if I had objects that were physically smaller than the collision tolerance value is it going to slow things down a lot? For a sim with say 2000 Fragments at what point do you guys start to think that the sim is running too slow and there must be something wrong? Minutes, hours?

Lastly how do you deal with multiple sims? For example I’ve got three different setups in my scene that are similar - is there a handier way to deal with sims / fragmenting than clearing the lists of objects and loading in the objects for each different setup?

Cheers in advance,

John

#1167

hi,
is there a way to set the amount of force needed for groups to break? a.e: i like the concrete settings with mass by material density. but with these settings the groups won’t break. with other settings they do break but their weight doesn’t look beliveable anymore…
would be cool to have an extra control value for that.

#1168

I think groups have solidity too, that you can use to do that.

#1169

thanks man, thats just what i was looking for! i aready changed that before but it seems i didnt go low enough…

ah and its in the custom properties too. nice!

#1170

Hey everyone!

I just thought I would share one of my greatest “rayfire ****ups” so far! haha yeah it sounds pretty bad… and I guess it is… but the result is so bad, it’s awesome!

The deal is, it was a simple test of the glue feature in rayfire 1.45 that turned out different than expected, simply because I cranked the glue strength up waaay too much. So when the objects fragmented and got glued, some of them twisted somehow and rayfire went crazy to fix it!

It might just be me… but I had a good giggle :stuck_out_tongue:

Enjoy!

http://thomaskc.dk/wip/rayfire/snake_box.avi

#1171

Hahah, yeah that was fun :smiley:

#1172

Hi
I am having a problem with a simple object while using PhysX.
When i simulate it seem that my object goes back to original object.
I have no idear why this is happeing, any assistance would b appreciated.
Thanks
Eaton

Here’s the scene file:

http://www.500mills.com/upload.max

Before simulation:

After simulation:

#1173

Previous versions of PhysX plugin used original, not transformed object’s state, current PhysX plugin resets objects transformation applying Xform modifier. it seems that after simulation it deletes this modifier and object get it original state.
Anyway, before simulate scaled objects it is beter to make Reset Xform and then collapse thei modifier stack. Should help.

#1174

Awesome, thank you very much.

#1175

Some R&D for a high-speed WWII aerial combat dogfight (using glue).

http://www.areagrey.com/other/fus01.mov
http://www.areagrey.com/other/fus02.mov

#1176

those are some amazing results with the glue! especially in the second MOV with more complex shape segments being glued together.

#1177

Nice :slight_smile: I dig the collision with the cylinder then the wall :wink:

#1178

Thanks :slight_smile: Avoid model interpenetrating faces/vertices and glue works good. The idea is to model both wings of the plane, and then have them hollow. Can then create internal parts like rivets, rods, bolts, cement chunks and circuitry to some degree for more destructive detail.

http://www.areagrey.com/other/fus03.mov

EDIT: Psycho - thanks dude. I usually start a RF sim at frame 10, and animate the impact group to smash into something from 10 to 20 (and put keys at 0 and 20). That way it has a little gravity in its motion.

#1179

thats gonna look sweeeet!!! i assume u dont let physx kick in until one frame before impact?

#1180

Using glue how?
is this a new addon to Rayfire?

Sorry to seem stupid, but it looks really cool.

David

EDIT: Never mind i see it is an add on…