Friday, 7 June 2013

AMD News Letter


Accelerated Apps
Photo Editing SDK by Aviary
Aviary SDK 
Aviary’s Windows 8 SDK has achieved significant performance improvements by implementing filters and effects through the new heterogeneous compute language C++ AMP. Computations are performed on the highly parallel GPU cores inside the AMD APU instead of the more serial CPU cores. Applying C++ AMP allows for processing of our complete range of filters and effects instantaneously – on average 16x faster than comparable processors according to benchmark studies conducted by AMD.

Update
hUMA diagram
Meet the Next Big Thing in Processors: hUMA
The HSA Foundation has released a new multicore architecture specification called hUMA or heterogeneous uniform memory access. hUMA allows CPUs and GPUs to share the same memory in a heterogeneous system architecture (HSA). hUMA makes it easier for developers to create apps that use the individual powers of CPUs and GPUS and is compatible with mainstream programming languages like Python, C++ & Java. You'll be able to see hUMA in action in AMD’s Kaveri processors, which are scheduled to appear later on this year.
HSA Foundation
HSA Foundation releases Programmer’s Reference Manual v0.95
This manual provides a standardized method of accessing available computing resources in HSA-compliant systems. The programming architecture detailed in the manual calls out features specifically exposed to programmers of the HSA architecture. HSA devices include a broad class of devices (i.e. GPUs, DSPs) and support hardware features that enable easier developer programmability including shared coherent virtual memory, platform atomics, user mode queuing and GPU self-queuing. Learn more.
Graphics Core Next
50 Awesome GCN Performance Tips 
Check out this resource with 50 Graphics Core Next (GCN) performance tips that AMD's @NThibieroz's tweeted out. @ElMarceltweets that this list is "incredibly valuable." For example, the first tip is "Issues with Z-Fighting? Use D32_FLOAT_S8X24_UINT format with no performance or memory impact compared to D24S8." See notes for this first tip plus get all 50 GCN performance tips.
Xbox One Logo
AMD APU to power new Microsoft XBOX ONE
Earlier this month, Microsoft announced that its next-generation XBOX ONE game console will be powered by a semi-custom AMD APU. The Microsoft XBOX ONE will use a single-chip, semi-custom AMD APU with components that were co-developed by Microsoft & AMD. The XBOX ONE is anticipated to be released later this year and “will merge social, gaming, and TV, glued together by the new Kinect 2.0” per this Tom’s Hardware article.
Laura Croft Head Shot
New Radeon SDK Direct3D 11 Samples: TressFX11 and Depth Bounds Test
When Steve Dent of engadget wrote the article “Because they’re worth it: game characters get AMD to do their hair”, he was referring to the new TressFX technology that’s used in Tomb Raider to give Lara Croft those luscious locks. Learn how to use AMD’s TressFX hair rendering technology in this sample or in the FMX 2013 presentation “Hair Rendering in Tomb Raider”. And while you’re perusing the Radeon™ SDK Direct3D11 Samples, you should review the DepthBoundsTest11 –which shows how to use the AMD Depth Bounds Test extension, a feature of the AMD GCN architecture, to limit rendering within a specified depth bounds.
Mark Papermaster
AMD's Mark Papermaster’s QA session at Jefferies 2013
At the Jefferies 2013 Global Technology, Media & Telecom Conference earlier this month, AMD’s CTO, Mark Papermaster, discussed the beginnings of the surround computing era and why AMD is excited about it. You can listen to the complete QA sessionfor free for a few more days before it goes behind a paywall.
Roy Taylor
Meet AMD's Roy Taylor, Corporate VP of Global Channel Sales 
Roy's first computer was an Armstrad 286 lug book. He got hooked on gaming after playing Doom with the OpenGL patch and a Voodoo card. He also has some interesting takes on online gaming & the cloud, R&D funding opportunities, AMD and mobile gaming, and more. Read Roy’s interview on KitGuru.

Education and Events
 Training Date Location
OpenCL Programming Course by AccelerEyesJuly 10-11 2013San Jose, CA
OpenCL Programming Course by AccelerEyesSeptember 4-5 2013Boston, MA
OpenCL Programming Course by AccelerEyesSeptember 25-26 2013Washington D.C.

Resources
OpenCL Software Development Downloads
OpenCL Training Online
Recommended Reading
University hcResources
Regional hcResources

No comments:

Post a Comment