Pages: [1]
Print
Author Topic: What new features to expect in PhysX 3?  (Read 10003 times)
MelvinEng
Newbie
*
Posts: 8


View Profile
« on: April 27, 2010, 11:29:50 am »

Hi,

I find myself anticipating the release of PhysX 3...and wonder if the features below are on the cards?
- implicit mesh generation for liquids
- universal finite element fracture model for rigid/deformable bodies(just like the DMM engine)


Cheers.  Smiley
Logged
Zogrim
Administrator
Hero Member
*****
Posts: 564


Expert Scrounger


View Profile WWW
« Reply #1 on: April 27, 2010, 11:54:38 am »

Quote
I find myself anticipating the release of PhysX 3...and wonder if the features below are on the cards?
Details weren't revealed to public yet - release date, features, compability, etc.

All I can say, SDK 3.0 is almost complete rewrite of current PhysX SDK  Grin

Quote
universal finite element fracture model for rigid/deformable bodies
All destruction/fracturing stuff in PhysX SDK will be based on APEX Destruction module

Talk from latest GDC 2010 "Authoring Physically Simulated Destruction with NVIDIA APEX" gives a good glimpse on current state of APEX D.
Video record is available at GDCVault - just search for "apex"

As for FEM fracturing, it isn't supported yet, but at the and of the session they added that some king of "stress system" is in to-do list
Logged

PhysXInfo.com Team
MelvinEng
Newbie
*
Posts: 8


View Profile
« Reply #2 on: April 27, 2010, 12:47:30 pm »

Thanks for the insights...

I also hope they improve the deformable body system:
- triangle/tetrahedral-level self-intersection avoidance?
- triangle/tetrahedral-level collision against other objects(rigid bodies, deformable bodies, fluid particles) instead of just particle-level collision?
And more on the fluid system:
- correct buoyancy behaviour of deformable objects in fluids? (do even rigid bodies have correct buoyancy behaviour in fluids?)
Logged
Zogrim
Administrator
Hero Member
*****
Posts: 564


Expert Scrounger


View Profile WWW
« Reply #3 on: April 27, 2010, 01:43:37 pm »

Quote
- triangle/tetrahedral-level self-intersection avoidance?
- triangle/tetrahedral-level collision against other objects(rigid bodies, deformable bodies, fluid particles) instead of just particle-level collision?
Sadly, I don't know such specific data.
You probably should post it at PhysX dev. forums
The'll hardly spoil such things, but at least it's worth a try

P.S. it's right time to do it - Quan Chen (PhysX support engineer) is massively answering on questions now, and gonna probably "fall asleep" for another month or two after than.


Quote
correct buoyancy behaviour of deformable objects in fluids ?
Dunno about SPH, but that's supported by Heightfield solver. You can try out demo by yourself
Logged

PhysXInfo.com Team
MelvinEng
Newbie
*
Posts: 8


View Profile
« Reply #4 on: April 28, 2010, 07:02:08 am »

Thanks for the suggestion...just did that.
Hope they do get back though...
Btw any idea what's the eta for v3?
Logged
Zogrim
Administrator
Hero Member
*****
Posts: 564


Expert Scrounger


View Profile WWW
« Reply #5 on: April 28, 2010, 08:48:42 am »

Quote
Btw any idea what's the eta for v3?
Probably early next year.
Logged

PhysXInfo.com Team
MelvinEng
Newbie
*
Posts: 8


View Profile
« Reply #6 on: April 28, 2010, 09:43:52 am »

Oh wow....that's really a long wait... Sad
Logged
Zogrim
Administrator
Hero Member
*****
Posts: 564


Expert Scrounger


View Profile WWW
« Reply #7 on: April 28, 2010, 03:53:37 pm »

There will be another transitional SDK version before 3.0, afaik.

However, I don't know nor release date, nor feature list.
Logged

PhysXInfo.com Team
Pages: [1]
Print
Jump to: