Comment Feed for Channel 9 - DirectCompute Lecture Series 230: GPU Accelerated Physics http://ecn.channel9.msdn.com/o9/ch9/6/3/2/8/5/5/DirectCompute230GpuPhysics_512_ch9.png Channel 9 - DirectCompute Lecture Series 230: GPU Accelerated Physics Join Lee Howes, a Senior SDE at ATI/AMD, as he gives you a deep dive into GPU accelerated physics. This lecture explains how DirectCompute can be employed to create solutions simulating physical systems such as fluid and cloth systems. It also covers how to employ some of the libraries powered by DirectCompute and how to integrate these libraries into your application.  After completing the lecture, you will be able to understand the underlying algorithms of some of the physics systems written in DirectCompute, as well as how to make use of the libraries that implement them, such as Bullet.For more information on DirectCompute, download the PDC 2009 DirectCompute HOL, watch the DirectCompute Roundtable discussion, see the full DirectCompute lecture series, and download the slides for this lecture. en Wed, 22 May 2013 10:47:14 GMT Wed, 22 May 2013 10:47:14 GMT Rev9 Re: DirectCompute Lecture Series 230: GPU Accelerated Physics Excellent lecture on GPU Accelerated physics.

posted by johni2

]]> http://channel9.msdn.com/Blogs/gclassy/DirectCompute-Lecture-Series-230-GPU-Accelerated-Physics#c634136612440000000 Fri, 02 Jul 2010 09:54:04 GMT http://channel9.msdn.com/Blogs/gclassy/DirectCompute-Lecture-Series-230-GPU-Accelerated-Physics#c634136612440000000 johni2 Re: DirectCompute Lecture Series 230: GPU Accelerated Physics that is awsome Smiley

posted by aL_

]]> http://channel9.msdn.com/Blogs/gclassy/DirectCompute-Lecture-Series-230-GPU-Accelerated-Physics#c634139743800000000 Tue, 06 Jul 2010 00:53:00 GMT http://channel9.msdn.com/Blogs/gclassy/DirectCompute-Lecture-Series-230-GPU-Accelerated-Physics#c634139743800000000 aL_ Re: DirectCompute Lecture Series 230: GPU Accelerated Physics Neat, of course, but also terrifying in practice.

 

Simple graphical HLSL programs are a painful to debug, even with great tools like PIX.

 

Debugging this type of highly optimized code will be a nightmare. I'm not even sure how to write a reasonable unit test for this type of code. We can reason about the underlying machine and look at frame rates, but that's still comes short of the profiling capabilities we're used to on the CPU.

 

Hopefully we'll start seeing more cores on the CPU so that we can keep doing our physics calculations there, where our development tools are sufficient.

posted by bryanedds

]]> http://channel9.msdn.com/Blogs/gclassy/DirectCompute-Lecture-Series-230-GPU-Accelerated-Physics#c634172310240000000 Thu, 12 Aug 2010 17:30:24 GMT http://channel9.msdn.com/Blogs/gclassy/DirectCompute-Lecture-Series-230-GPU-Accelerated-Physics#c634172310240000000 bryanedds