Game Developers Conference Europe 2011: more Intel session details

As you might know from our previous blog post Intel is going to attend Europe's biggest developer's conference, Game Developer Conference Europe (GDCE), which takes place from 15 to 17 of August, 2011, in Cologne, Germany. Intel is not only one of the big sponsors during GDCE but will also cover some hot topics regarding game development and show how you as a game developer can optimize your code efficiently. Just to make sure that your wonderful gaming titles will run best on existing PC platforms like "Sandy Bridge" and on future Intel architectures as well.

In order to share these information with you, Intel engineers will talk about different topics which have already been annouced. Here now a bit more information on the various sessions:

On Monday 5:40- 6:30 pm (CET), Intel software engineer Leigh Davies will talk about optimizing task-based games for multicore systems. He will show in depth how an efficient tasking system allows writing your game without worrying about the underlying hardware. Tasking decouples algorithmic parallelism from implementation, allowing your technology to scale to ‘N’ cores. Together with Leigh you will explore the parallelization of rendering, animation logic, and MLAA post-processing investigating the implementation and performance pitfalls we encounter along the way.

Sounds kind of cool, doesn't it?

One day later, on Tuesday, August 16, from 9:30 - 10:20 am (CET), Levent Akyil will enter the stage to give a lecture on „Hotspot Flops & Micro-Ops“. With his talk he will show how game developers understanding out-of-order execution of instructions as well as fully utilizing wide SIMD as this is a critical mission to obtain the most performance on modern CPUs. Moreover Levent will focus on instruction and SIMD level parallelism techniques using Intel's Sandy Bridge micro-architecture as an example. He will explain how instructions, including the new Advanced Vector Extensions (AVX), are decoded and executed out-of-order on the x86 back-end. In order to make his session more "lively" Levent will present sample workflows to identify bottlenecks in existing code, including architecture specifics, and validate changes to obtain optimal code performance.

Another session which is worth to be visited. And the best of it: I will be at GDCE 2011 and will do live bloging about all these different topics.  What people think about their favourite topics you will see and hear as well as we will video tape some interviews for you. And about the remainder of the Intel sessions you hear about tomorrow - stay tuned :-)
For more complete information about compiler optimizations, see our Optimization Notice.