Comments on: Popular Physics Engines comparison: PhysX, Havok and ODE http://physxinfo.com/articles Fri, 03 Sep 2010 16:19:18 +0000 http://wordpress.org/?v=2.9.1 hourly 1 By: Danelle Bealer http://physxinfo.com/articles/?page_id=154&cpage=1#comment-10 Danelle Bealer Sun, 06 Jun 2010 09:53:30 +0000 http://physxinfo.com/articles/?page_id=154#comment-10 Found your blog and decided to have a quick read, not what a normally do but nice one. Nice to see a blog for a change that isn’t full of spam and rubbish, and actually makes some sense. Anyway, nice write up. Found your blog and decided to have a quick read, not what a normally do but nice one. Nice to see a blog for a change that isn’t full of spam and rubbish, and actually makes some sense. Anyway, nice write up.

]]> By: Bebox http://physxinfo.com/articles/?page_id=154&cpage=1#comment-5 Bebox Tue, 08 Dec 2009 17:31:45 +0000 http://physxinfo.com/articles/?page_id=154#comment-5 Nvidia's policy of not supporting GPU acceleration for Physx unless the primary video adapter happens to use an Nvidia chip is likely to prove hurtful to its adoption in the long run. They would do well to revise this stance. On the other hand, Intel (and AMD)'s interest in keeping Havok an argument for buying quad core CPUs has left any kind of GPU support long overdue. And AMD's failure to deliver in the GPGPU area (No OpenCL support for anything less than RV770 class GPUs, no CS 4.x support at all) bodes ill for anything but the slightest lip service to plans in this area. Given the current state of things, developers and consumers alike should probably be rooting for Physx with its comprehensive hardware support, excellent developer support and liberal licensing conditions. Nvidia’s policy of not supporting GPU acceleration for Physx unless the primary video adapter happens to use an Nvidia chip is likely to prove hurtful to its adoption in the long run. They would do well to revise this stance.

On the other hand, Intel (and AMD)’s interest in keeping Havok an argument for buying quad core CPUs has left any kind of GPU support long overdue. And AMD’s failure to deliver in the GPGPU area (No OpenCL support for anything less than RV770 class GPUs, no CS 4.x support at all) bodes ill for anything but the slightest lip service to plans in this area.

Given the current state of things, developers and consumers alike should probably be rooting for Physx with its comprehensive hardware support, excellent developer support and liberal licensing conditions.

]]> By: Zogrim http://physxinfo.com/articles/?page_id=154&cpage=1#comment-4 Zogrim Tue, 08 Dec 2009 06:53:17 +0000 http://physxinfo.com/articles/?page_id=154#comment-4 "Although it looks like PhysX is pretty much about to conquer the world" Nowhere near. PhysX is just used in more PC games, but Havok is still preferred for AAA and console titles. “Although it looks like PhysX is pretty much about to conquer the world”
Nowhere near. PhysX is just used in more PC games, but Havok is still preferred for AAA and console titles.

]]> By: Hmm_Dubious http://physxinfo.com/articles/?page_id=154&cpage=1#comment-3 Hmm_Dubious Mon, 07 Dec 2009 22:33:06 +0000 http://physxinfo.com/articles/?page_id=154#comment-3 It is a bit misleading. Although it looks like PhysX is pretty much about to conquer the world, it is also due to the fact that the Unreal Engine 3 uses it. And the engine is very popular as it is supported on several platforms (Xbox 360, PS3 and PC). But still, any kind of physics is better than nothing at all :) It is a bit misleading. Although it looks like PhysX is pretty much about to conquer the world, it is also due to the fact that the Unreal Engine 3 uses it.

And the engine is very popular as it is supported on several platforms (Xbox 360, PS3 and PC).

But still, any kind of physics is better than nothing at all :)

]]>