While running the workload we noticed that the unsymmetrify mapper job in the Mahout item-based recommendations was taking a long time to execute. The item-based recommendation approach calculates user preferences for an item based on user preferences towards similar items. When analyzing the profiles we noticed that one of the data structures is being recreated, and is allocating memory for every key value pair in the record. 

To fix this behavior, we developed a patch that initializes the data structure per record.  This pattern, which could be easily overlooked by programmers, can cause serious performance degradation in the performance of Mahout™ MapReduce jobs.  In the case of the workload we were studying, we measured on our AMD Opteron 4228 HE cluster more than a 4.5X speed-up in the job execution time with a two-line code change!

The following graph illustrates the improved execution time gained by using our optimization.

Performance improvements achieved by using this technique will vary by workload.  Gains depend on the number of key value pairs in the input record and the amount of heap being allocated.

The context.write method writes the data into intermediates and output files.  In MapReduce programming, when context.write is called the specified key value pair is guaranteed to be serialized and written.  Therefore, reusing the same object when possible will increase performance. 

The ASF Email dataset is publically available for download at http://aws.amazon.com/datasets/7791434387204566.  Our test cluster used eight data nodes and one name node.  Each data node has twelve hard drives and 64GB of RAM.  This system uses RHEL 6.2, Java version 1.7.0 update 4, Hadoop version 1.0.2 and Mahout 0.6 versions.

Bhaskar Devireddy is a Member Technical Staff in the Runtimes Team at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

But superhero movies aren’t the only highly anticipated events each year. What about when new products come to market?  Products with superhuman strength, speed and agility, like Dell PowerEdge Servers.  Dell’s commitment to innovation and reputation for bringing industry-first features into the data center has wowed its customers for years. When AMD FirePro™ professional graphics officially made its data center debut last year, Dell was the first OEM to offer AMD FirePro V7800P in a server platform, the Dell PowerEdge M610x blade server. The AMD FirePro™ V7800P is capable of tackling the most demanding compute intensive, data parallel tasks, and the PowerEdge M610x was the first complete Microsoft® certified RemoteFX® server solution for virtualized desktops.

Dell and its customers can appreciate the ultra-fast performance, visual quality and massive parallel processing power AMD FirePro™ professional graphics deliver in server environments. In March 2012 Dell introduced its new line of PowerEdge rack, blade and tower servers, and two of these new platforms are now available with AMD FirePro™ professional graphics. Harnessing the massively parallel computing power of graphics processors allows Dell to support new usage models and demanding workflows.

According to PC Pro, the PowerEdge R620 rack server “…sets new standards for server design, offering an unbeatable range of features and remarkable expansion potential.”  Each 1RU R620 comes with two PCIe® x16 slots with x16 bandwidth, as well as an optional third x16 slot with x8 bandwidth, and can accommodate two ATI FirePro™ V5800 graphics cards. For customers looking for a space efficient and affordable solution to deploy and support virtual machines, the R620 with one or two V5800s can support multiple virtual machines running common office applications with RemoteFX.

The Dell PowerEdge R720 is designed to push the limits of virtualization performance. According to ITPro, “As a general-purpose 2U rack server, the PowerEdge R720 looks capable of running any application that can be thrown at it. Its massive memory capacity and hypervisor redundancy make it ideal for virtualization, while the storage capacity, expansion potential and remove management are excellent.” The R720 is a high density solution ideal for larger VDI deployments, capable of accommodating up to four passively cooled AMD FirePro V7800Ps that can support up to 64 virtual machines, as each V7800P can support up to 16 virtual machines running common office applications.

The Dell PowerEdge T620 tower server offers high performance with built-in room for growth, and according to PC Magazine it’s “…custom made for fault-tolerance, redundancy and virtualization.” It features seven PCIe® slots and is capable of accommodating four actively cooled ATI FirePro™ V7800 professional graphics cards.

The Dell PowerEdge R620, R720 and T620 have received high praise from reviewers at technology publications, and I’m sure customers will be wowed by the increased application performance, availability and scalability these next-gen servers deliver.

Mitch Furman is a senior product manager, Professional Graphics, AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

While there are too many examples to mention, I’ve provided a sampling of applications here.  In medical imaging, multicore systems receive the streams of data from the sensors in MRI and CT scanners, quickly converting the data into extremely high-resolution images that can be analyzed by radiologists.  In video surveillance, many cameras capture images simultaneously which can then be quickly correlated and analyzed to identify and track items or subjects.  And in Homeland Security applications, vast sensor networks deployed in ports scan for hazardous or radioactive materials that someone might try to smuggle into the country.

While each of these high-performance multicore processor-driven applications will benefit from improvements in their real-time capabilities through increased performance, they also will benefit if they can be made more portable and more affordable.  For example, deploying medical imaging systems to field hospitals in impoverished countries can help improve healthcare in an entire region.  Similarly, more portable systems can help improve security in locations where high performance servers can’t easily be deployed for environmental or cost reasons. And more cost-effective and portable designs can translate into improvements in the nation’s security, ensuring better protection of our coastal ports, both large and small.

Developers of these current-generation multicore applications constantly strive for more performance, and squeeze as much performance as they can from their platforms.  But conventional sequential microprocessors and coding languages are reaching their limits for delivering the faster real-time performance, smaller system size and low-power demands of next-generation applications. CPU registers, data paths, caches, instruction-level parallelism, multicore processors, frequency scaling, ASICs and FPGA offload engines are all being stretched in their ability to deliver the required performance under increasingly stringent power budgets.

So developers are turning to new heterogeneous system architectures, which show good promise for delivering the performance they need while also enabling reduced systems costs, size and power consumption.  These new architectures leverage traditional graphics processors with their hundreds of parallel compute units to offload certain tasks from the CPU.  The GPU has evolved from a basic device rendering images on screen to an advanced, multi-purpose 3-D graphics processor that drives the rich visual interfaces of today’s multimedia displays as well as today’s compute-intensive, parallel processing-driven applications.

Along with these advances in GPU capabilities have come advances in GPU programmability.  Over time, these benefits have helped GPUs transition from driver-based programs to system based programming models for graphics.  Today’s GPUs are built around an advanced parallel processing architecture capable of operating on hundreds of pieces of data in parallel in a single cycle.  And by taking advantage of open and royalty-free programming standards for general-purpose computations on heterogeneous systems, such as OpenCL^TM, the GPU can be used for more than graphics processing.  Another typical use, for instance, is programming for system level applications.  Multiple GPU cards can already be ganged together to deliver combined graphics performance in PC applications.  And now these same techniques are used to build clusters for HPC (High Performance Computing) applications.

So it’s no wonder that developers of high performance, multi-core-driven systems are taking advantage of this new GPU technology.  And that’s a good thing.  In the future, we all might sleep a little more comfortably knowing better systems are helping guard against terrorists, that third world countries are getting better health care, and that our streets and facilities are more secure.

Cameron Swen is the Manager of Embedded Marketing at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

by Molly Swinney, IT Project/Program Manager

We often look at a job posting and see “excellent presentation skills” as defining criteria for job qualification.  But what does that mean?   Most of us recognize a good presenter when we see one, but it’s sometimes hard to put into words what it takes to be one.  Effective presentations break down into two aspects:  content and presentation delivery.

The content is the message of the presentation, so it warrants careful consideration. Here are a couple of content pointers:

Be Organized
First, tell the audience the main point up front and let them know the structure of your presentation.  This can be done simply through an introduction and an agenda, or by saying “Today I will be discussing X, Y and Z” and listing the top supporting points of the presentation.  This puts the audience at ease and gives them a framework for your presentation.

Second, review your presentation material and make sure that all of the content supports the main points.  If not, rework the content or remove it so that the main and supporting points remain clear.

Consider the Audience
Before developing your presentation material, consider the audience.  The age, position, technical and ethnic background of the audience may greatly influence what and how you present.  We all likely recognize that we would not present information about the solar system to a group of kindergartners in the same way we would to a group of senior scientists.  However, many people fail to consider that they may want to present the information about their work project differently to a group of executives than they would to a group of their scientific peers, or that other factors may influence their content.  Here are some  guidelines you can use when tailoring for the audience:

• Executives might  want to know things such as overall purpose, approach, timeline, cost and benefit.  You can prepare additional background material on these topics should they want to dive deeper into an area, or be prepared to provide a follow-up conversation or presentation on a specific aspect, if necessary.  An executive presentation should get to the bottom line quickly.  When discussing issues, present the situation, complication and proposed solution as succinctly as possible.

• Watch the acronyms!  Spell out acronyms in your presentation material and reference them with the acronym at least once.  For example:  Advanced Micro Devices, Inc. (AMD).

• If you have an international audience, slow down!  If you have ever listened to someone speak your language with a thick accent, you know what it must feel like to translate as you go.  Allow the audience sufficient pauses to be able to absorb the information that you are presenting.  Also, watch jargon and slang in your presentation.  It is hard enough to master another language without having to figure out local colloquialisms.

• Mixed audiences are the hardest because you have to satisfy many needs.  You will need to present information at a fairly basic level to bring everyone along, but you should also consider if there is a primary audience that needs to gains specific information.  You can sprinkle in more complex information for those that are interested or need the extra level of detail.  When presenting to mixed audiences, you may want to provide sources where those who are interested can get additional information.

Now that you’ve crafted stellar content, it’s time to focus on delivery.   We’ve all probably suffered through at least a couple of presentations where the speaker knew what he or she was talking about but had a very hard time getting the point across. Here are some tips that will help make your presentation the best that it can be:

Practice
Practice presenting your material out loud prior to the event.  No matter your best intentions, speaking something in your head doesn’t necessarily translate to how it will come out when you say it out loud.  It can be helpful to practice in front of a mirror to see what you look like, or, even better, record yourself on a video recorder to check your rate, volume, presentation style and motion.  Better still, ask a friend or co-worker to preview your presentation and offer pointers.

Survey the Venue
If you are not familiar with the room and equipment you will be using to present, go visit the room and try everything in advance.  If you cannot visit the site, contact the organizer and make sure you understand what will be provided.  It’s important to understand the presentation space (i.e., where you can move around and where you will be positioned in relation to the audience), audio-visual equipment, flip charts or white boards and seating.  The logistics of the room will determine how loud you must speak and how you might structure presentation materials.

Have a Backup Plan
If you are using electronics to present, consider what is key to your presentation and what might happen if something should fail.  Could you continue by simply talking about your material?  Or, could you provide handouts?  Is there a backup projector or another way to present (e.g., via Live Meeting)?  It’s helpful to save a copy of your presentation on a memory key in advance in case your hardware fails to connect properly or someone needs to make last-minute handouts.

Relax!
Remember, YOU are the expert and that is why you have been asked to present.  You have something of interest to tell, and others are there to receive your message.  Take a deep breath before you start, tell your story, and pause occasionally to check for understanding.

Although these pointers will not by themselves make you an expert presenter, hopefully they will help you bring your next presentation up to a new level.  So, next time you see “excellent presentation skills” in a job requisition, take heart that you know what to do to qualify…then go forth and deliver!

Molly Swinney is an IT Project/Program Manager for AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

1 x AMD Radeon™ HD 7970 graphics card based on our reference design

You’ll need one of our reference design boards for maximum compatibility. Once you remove the shiny chassis off the card, you’re ready to add some fancy accessories.

If water is your choice for cooling, you won’t be disappointed with the options available to you. Here are just a few on the market.

2^nd EK-FC7970 – Acetal+EN(Nickel)

EK-FC Bridge DUAL Serial

2 x EK-FC Link R48X0/58X0

Air Cooling

For those who think electronics and water don’t mix. You can still have some fun modifying your AMD Radeon™ HD 7970 GPU with one giant heat sink. (5 heat pipes + 3 fans)

Remember to check with your manufacturer of your card before you proceed as modifying your graphics card may void your warranty.

Happy Modding!

David Tjong is on the Technical Marketing team at AMD.  His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Download AMD OpenCL™ APP SDK 2.7 now from http://developer.amd.com/appsdk .

In addition to the above we have updated to gDEBugger, APP Profiler, Kernel Analyzer and APP ML, and there are numerous new and improved samples.  We are continuing to work on our samples and new samples will be posted on http://developer.amd.com/sdks/AMDAPPSDK/samples/Pages/default.aspx as they become available over the next few months.

The OpenCL™ 1.2 adds the following key capabilities

• Host access flags for memory objects enable more efficient buffer handling and provide added protection. For example, a buffer that is created as “write only” cannot be read from the host.
• Pattern based GPU buffer and image initialization can help eliminate need for certain buffer/image transfers
• Memory objects migration supports transfer of buffers prior to need
• New generalized image creation API
• Enhanced image/buffer map operations
• OpenCL  1.2 CPU device partition including partition of a CPU after addition to a context
• Generalized 1D and 2D images, image arrays,  and image<-> buffer interop
• Libraries support including the separation of compile and link phases and the ability to compile

The C++ Wrapper API provide the following new capabilities

• Defaults for platform, queue, device, … helping to significantly reduce  the amount of boilerplate code required.
• Improved simplified constructors for cl::Buffer and addition of cl::copy functions
• Additional support for events to functors

Notable C++ features that are supported by the OpenCL™ Static C++ Kernel language

• Kernel and function overloading
• Inheritance
  • Strict inheritance
  • Friend classes
  • Multiple inheritance
  • Templates:
    • Kernel templates
    • Member templates
    • Template default argument
    • Limited class templates (the “virtual” keyword is not exposed)
    • Partial template specialization
    • Namespaces
    • References
    • ‘this’ operator
    • with external symbols
    • Kernel reflection, the ability to query a kernel’s arguments
    • Support for printf as a built in function

Additional features supported in SDK 2.7 and the Catalyst 12.4 drivers include:

• Support for Asynchronous PCI transfers
• Video encode using VCE Encode (Win7)
• Open Encode update (12.4)
• Cl_khr_fp64 is now supported on AMD Radeon HD™ 7900 series devices (“Cayman”)
• Added OpenGL™ interoperability under Linux for  AMD Radeon HD™ 7000 series devices
• Stability Improvements
• Performance improvements
• Support for AMD Radeon HD™ 7000 series devices (“Southern Islands”) NPI
• Support for AMD’s Second Generation APUs (“Trinity”)
• Kernel Analyzer v1.12
• APP Profiler  v2.5

gDEBugger version 6.2; downloaded for use with this SDK from http://developer.amd.com/gDEBugger.

• Introducing Linux® Support
• New standalone user interface for both Linux® and Windows®, with enhancements for better navigation and ease of use
• Supports OpenCL™ kernel and API level debugging on AMD Radeon™ HD 7000 series graphics cards
• Supports OpenCL™ 1.2 beta drivers
• Automatic updater to notify and download new product updates
• Feature enhancements including support for static arrays, union variables and Find feature
• Stability improvements

APP KernelAnalyzer v 1.12

• Support for Catalyst revisions through 12.1 – 12.4.

APP Profiler v2.5 includes several key new features, including:

• Stability improvements

APP ML 1.8

• Support for real to complex FFT

New and updated samples

• Nbody: optimized for improved performance
• DeviceFission: a new version of this sample using OpenCL 1.2 Device Fission capabilities. The old version is still included but renamed as DeviceFission11Ext
• ImageOverlap and GaussianNoiseGL are two new OpenCL™ 1.2 samples
• DwtHaar1DCPPKernel: an additional version  of DwtHaar1D but modified to use the C++ kernel language
• MatrixMultiplicationCPPKernel: an additional version  of MatrixMultiplication  but modified to use the C++ kernel language. This sample supports multiplication of both int and float matrices through use of a template.
• TransferOverlapCPP: an additional version  of TransferOverlapCPP but modified to use the C++ wrapper API
• The URNGNoiseGL and HistogramAtomics samples have been modified to use the C++ wrapper API
• The FFT, MersenneTwister,  and EigenValue samples have been modified to use C++ kernel language
• There have been incremental improvements to a number of additional samples

1. Information is based on a published Khronos Specification, and is expected to pass the Khronos Conformance Testing Process. Current conformance status can be found at www.khronos.org/conformance.”

Mark Ireton is a Sr. Manager, Product Application Engineering at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Many embedded applications run serially on a standard CPU. Under these circumstances, parallelization occurs in multiprocessor systems or virtually via time-splicing control of the individual, relatively large processor. The situation is different with a GPU where tasks can be distributed over many, very small and highly-specialized engines enabling parallel execution in a single step (SIMD=Single Instruction Multiple Data), creating the classic parallel compute architecture.

These engines are linked with one another according to their respective tasks and execute the tasks in each clock cycle in parallel. What is special about leveraging a graphics processor for parallel processing is the fact that the individual processor tasks are not hard-wired (as is the case in the Vertex Shader Unit of a simple GPU).

Instead the particular tasks are freely configurable like they might be found in a network processor. Since task formulations are seldom purely serial, there are in practice enormous efficiency advantages that can be leveraged when using an APU, which can process both serial and parallel tasks.

As a result, traditional CPU performance benchmarks, which were used to compare the performance of platforms, are no longer meaningful because a one size approach does not fit every application. The APU provides the scalability to meet the wide range of compute demands of a variety of embedded applications.

Software developers can take advantage of APU programmability through standards such as DirectCompute (Microsoft) and OpenCL^TM. And the portability of OpenCL can allow programmers to preserve their source code investment across product generations.

OpenCL is a powerful programming environment with which computing tasks can be distributed across a variety of architectures involving a various number of hardware compute units. This can be of crucial importance in our customers applications such as portable medical imaging that tend to execute a variety of workloads and must do it efficiently from both a cost and power perspective.

congatec’s COM Express Type6 Module, the conga-TFS, is based on the AMD Embedded R-Series APU platform. The integration of AMD Embedded R-Series APU helps expand the COM Express standard with a new module that combines power efficient high-performance x86 processing, unprecedented discrete-class graphics performance and high performance parallel processing support. The conga-TFS enables multimedia advancements and an exceptional visual experience across applications.

congatec will offer three variants of the AMD Embedded R-Series APU on the TFS ranging from the dual-core AMD Embedded R-272F APU up to the quad-core AMD Embedded R-464L APU. The conga-TFS uses the AMD Embedded A70M Controller Hub to deliver a powerful compact 2-chip solution with support for up to 16 GB dual-channel 1600MHz DDR3 memory.

For our customers designing digital signage and casino gaming solutions, the integrated graphics core supports DirectX^® 11 and OpenGL 4.2 for fast 2D and 3D graphics. The third-generation hardware Unified Video Decoder offers seamless processing of H.264, VC-1, MPEG4 Part 2 and MPEG2 video streams. The conga-TFS offers a choice of graphics interfaces including VGA, 18/24bit single/dual channel LVDS as well as three DisplayPort 1.2, one HDMI and two Single link DVI interfaces enabling the direct control of three independent displays.

For data streaming and conversion applications that require high throughput I/O, the conga-TFS has seven PCI Express x1 lanes and a PCI Express x8 link. In addition, the combination of four SuperSpeed USB 3.0 ports and four USB 2.0 ports help enable applications such as Point of Sale that require a lot of USB for scanners, scales, card readers and more.

For storage centric applications the conga-TFS has four SATA 6Gb/s ports to provide high-speed disk access and one Gigabit Ethernet interface enables network connectivity. The congatec board controller features ACPI 3.0 power management and high definition audio to round out the features of the conga-TFS.

Dan Demers is an Area Sales Manager at congatec Inc. He holds a B.B.A. degree in International Business from Grand Valley State University, Grand Rapids, Michigan and an M.B.A. from Ashford University, Clinton, Iowa. Mr. Demers has more than 14 years of experience in embedded computing having worked with Fortune 500 companies in the medical, military, and communications markets.

Here at Advantech-Innocore we’ve been designing special purpose embedded systems for the casino gaming industry for more than 13 years. Indeed in some segments of the market our DPX series motherboards have become the de-facto standard used among many slot machine OEMs.

As we make the chipset selections for each new product we are looking for the right combination of platform features with special attention to those that are most important to our customers in the casino hardware industry. Key among those features is graphics performance and multi-display support. Longevity, cost and technical support are also very important considerations. When we heard about the upcoming AMD Embedded R-Series platform, we were initially skeptical of the claims of “discrete level graphics performance” from the integrated graphics in an embedded chipset.

Anyway we began our design and throughout the process we were given great support from the AMD team – a challenge made all the harder since Advantech-Innocore’s design teams are located in the United States, U.K. and Taiwan. Sure enough the benchmarks realized on the DPX-S430 with the AMD Embedded R-Series APUs are as good as or better than what we were expecting. The initial customer feedback has also been very encouraging.

With unique AMD features like AMD Eyefinity technology for up to four independent monitors, we feel that the DPX-S430 will allow many of our customers to bring to market eye popping new game content without the inconvenience of adding discrete graphics cards which add cost.

For the most demanding applications, the DPX-S430 also allows the flexibility to add a graphics card via a full PCIe x 16 v2.1 graphics card slot and the platform supports AMD Radeon™ Dual Graphics whereby the chipset graphics engine works in coordination with discrete cards such as the AMD Radeon™ E6760 and E6460 GPUs.

Overall we are excited to offer the DPX-S430 with AMD Embedded R-Series APU technology to our gaming industry customers. Advantech-Innocore continues to create a roadmap of exceptional quality using the latest technology combined with in-house manufacturing and local supply and service.

The DPX-S430 offers a marvelous package of performance at a price point suitable for our industry and with the backing of the AMD Embedded roadmap for the longevity our customers also need.

Customers who would like to upgrade from their existing DPX-S series platforms can do so with little integration work due to the consistent mechanical and software footprint and the support of our global support network.

Craig Stapleton is a Product Director at Advantech-Innocore. Advantech-Innocore is a leader in embedded computer products for the gaming industry. Advantech-Innocore designs, manufactures and sells technologically advanced PC products tailored to the specific needs of the gaming industry. With a large installed base worldwide, the DPX® range of products is proven in the harshest environments. Visit us at http://www.advantech-innocore.com for more information on our products and services.

So you get creative and innovative. Sometimes your innovation helps differentiate your products from the competition, and sometimes it simply enables you to meet your customers’ requirements.

Certainly, advances in processor technology have helped by ushering in reductions in semiconductor geometries while simultaneously helping to boost performance. The problem is that increases in performance came at a price: increases in the complexity of the core architectures and software development environments.

And while technological advancements like the transition from single to multi-core processors provide incremental increases in system performance with each new generation of processor technology, developers quickly figure out how to exploit all of the available performance of the low-power variants of these processing solutions.

Now there’s better news. New processing solutions based on heterogeneous architectures are emerging that enable low-power designers to significantly increase their system performance without adding significant cost or power to the system.

Heterogeneous processing itself isn’t new, but what make these solutions different are the open and royalty-free programming standards for general-purpose computations on heterogeneous systems that are developing around them, such as OpenCL^TM. Using standards like OpenCL helps programmers preserve their expensive source code investment and easily target and port code between multi-core CPUs, GPUs, and new APUs (Accelerated Processing Units) which combine both x86 processing cores and graphics processing units on a single die.

We are already seeing a variety of innovative applications of heterogeneous processing solutions in small form factor and low-power applications, such as:

• Smart cameras capable of running a full PC operating system while delivering up to 90 Gflops of performance (previous generations had just 2-3 Gflops of performance). See the Ximea case study.
• Medical imaging solutions in which parallel processing is accelerating the FDK algorithm (widely used for tomographic reconstruction of X-ray CT scanner data); similar approaches are being investigated for applications such as portable ultrasound devices.
• Digital signage and surveillance application analytics improvements; companies are creating algorithms that identify whether the subject or audience is male or female, child or adult, and are investigating parallel processing as a means of enabling the algorithms to run on lower-power and lower-cost processing solutions. If successful, they could easily integrate the algorithms into a camera or display.
• Programmable parallel processing solutions are in consideration as DSP replacements in telecommunications and mobile radar imaging.

For designers of low power solutions looking for innovative new approaches to best their competitors, or simply meet their customers’ exacting requirements, heterogeneous system architectures and the software to support them offer a solution that can meet, or exceed, these requirements. And with the introduction of the AMD Embedded R-Series platform this week, AMD is now able to offer a complete range of low power heterogeneous computing solutions with support for parallel processing.

The AMD Embedded G-Series platform announced in 2011 serves the needs of very low power applications with TDPs (Thermal Design Power) from 5.5 Watts to 18 Watts and calculated single precision floating point performance ranging from just under 50 GFLOPs to just over 90 GFLOPs. The AMD Embedded R-Series APU fits nicely on top of that with TDPs from 17 Watts to 35 Watts and calculated single precision floating point performance ranging from 172 GFLOPS to 563 GFLOPs. So now there are more options available to help you find that ideal balance between performance and power consumption.

Cameron Swen is the Manager of Embedded Marketing at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Phil Hughes is a Senior PR Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Myth No. 1: CPU is the brain of the computer.

Wrong. For a long time, when talking about PCs, people made the analogy that the CPU is the brain of the computer and the GPU is the eyes. The reason for this was obvious: CPUs do logical, numerical computation, while the GPU displayed images. As an analogy, the human brain is divided into two hemispheres – left and right. Popular opinion states that the left side is responsible for logical, numerical computation, while the right side is responsible for creativity, arts, emotions and visual attention. These two vastly different modes of thinking can both be considered as ‘processing’ the information, just that the data being processed is different.

Similarly, we can use this analogy when thinking about the various components of a PC, with the left hemisphere of the brain as the CPU and the right side as the GPU. However, for today’s applications, GPUs are absolutely as critical as CPUs. Not too many people need to run Excel spreadsheets faster, but many would like their photos to be more colorful, their HD movies smoother and their games more responsive. If the GPU you’re running cannot support this, then you might just be missing half of your brain, in a manner of speaking! You can read my full argument here: http://blogs.amd.com/home/2010/08/04/missing-half-your-brain/.

Myth No. 2: Manufacturing process is paramount.

Moore’s Law as applied to integrated circuits has accurately predicted that the number of transistors that can be placed on an integrated circuit doubles approximately every two years. We are so accustomed to Moore’s Law that we are now equating  new manufacturing technology with increased transistor density, which in turn, can lead to increases in performance or battery life. However, this is not necessarily true. Today, people care about the experience their device delivers and not just the manufacturing process. Just ask anyone who has ever used an iPad – what technology was the chip powering the iPad built on? If they don’t know, they shouldn’t feel bad about it. Most of us don’t give much thought to it.

Myth No. 3: Graphics (GPU) is only for gamers.

With the tremendous shift in people using their PCs for multimedia entertainment, application developers have started to change the way they code their programs in order to take advantage of the incredible compute power locked in the GPU. That has led to significant improvements in the responsiveness of leading web browsers such as Microsoft Internet Explorer 9, Google Chrome and Mozilla Firefox. Quickly managing photo editing via Adobe Photoshop CS6, video editing with Sony Vegas Pro 11, and reduced file compression times with WinZip 16.5, all of which are GPU accelerated applications, can vastly help improve the users’ experience.

Sometimes what we learned “back then,” isn’t always applicable to the here and now. Hopefully, my blog will help correct these misperceptions about GPUs and CPUs in the market and redefine them in their own unique ways as they shape a whole new generation of computing.

Leave me a comment below or share with your network of friends. Also, to find all our latest videos, photos, blogs and more for our AMD A-Series APU click here.

Sasa Marinkovic is the Sr. Manager, Industry Marketing Programs at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

We’ve been in close contact with our technology partners, not only with well known industry and technology leaders and brands, but also with exciting up-and-coming innovators and top minds and contributors to the open source community to enable new and enhanced performance and capabilities for end users. Some new products are launching today that have been specially tuned for our next-generation AMD A-Series APUs with exclusive features and enhanced performance:

• Arcsoft is launching Arcsoft Link+, which can become the users’ complete home media command center, enabling each device to share, stream or download  media to another device on their home network. With Arcsoft Link+, consumers can stream to and from one device to another   supported device, and with a PC with our AMD A-Series APU on the home network, video enhancements such as Arcsoft’s SimHD upscaling and AMD Steady Video stabilization technologies as well as necessary file conversions and can occur  in the background. See video demo.

• Cyberlink is releasing updates to its flagship products: PowerDirector 10, PowerDVD 12, MediaShow 6, and MediaEspresso 6.5 with optimizations for the new second-generation AMD-A-series APUs, providing users with hardware-accelerated HD playback, enhanced video editing, maximum rendering and transcoding performance.

• Fluendo is launching Moovida Universe, a 3D Media Center, providing users a new way to rediscover the content in their local media library. With its tailored 3D UI powered by a video game engine, Moovida Universe Interface is capable of displaying up to 13 concurrent videos that take full advantage of our next-gen AMD A-Series APUs power.

Here are just a few other examples of some recent support for accelerated user experiences that add to the already robust and growing list of optimized applications for AMD APUs, that include  products such as recently launched Adobe Photoshop CS6:

• Video playback: VLC media player, one of the top open-source based media players on the PC, will be including integrated support for AMD Steady Video technology 2.0, making it easy for users to turn on and off video stabilization from within the VLC user interface. Future work is continuing around OpenCL™ for de-noising, with an emphasis on “quality.” Here’s what VLC had to say about OpenCL acceleration on GPUs: “VideoLan is making it easier for consumers to enjoy smooth video playback with AMD Steady Video integrated into our VLC Media Player,” said Jean Baptise Kempf, president, VideoLAN. “Thanks to the versatility of OpenCL™, we can leverage the immense parallel processing capabilities of GPUs to help enhance everyday multimedia tasks and provide a rich video experience.”

• Image processing: When it comes to editing photos, we’ve worked with the team over at GIMP to enhance the photo-editing experience, “AMD has been an outstanding colloborator, working with the open-source community to enable GIMP to leverage the GPU capabilities using OpenCL,” said Victor Oliveira, GIMP OpenCL developer. “With millions of downloads, our open-source image editing tool now has 19 high-depth filters accelerated on the AMD A-Series APUs, among others features, allowing users to do their image manipulation tasks in a faster and smoother way.”

• Productivity: We’re not only accelerating video playback and photo imaging software through OpenCL, but also file compression and encryption. Corel’s WinZip 16.5 is able to leverage both CPU and GPU resources on our new second generation AMD A-Series APUs. Corel is just as excited as we are stating: “millions of business professionals and consumers count on WinZip to zip and deliver their files quickly,” remarked Patrick Nichols, President of WinZip. “The OpenCL accelerated file performance of WinZip 16.5 on AMD A-Series APUs lets our customers zip, share and secure even the largest file sets faster than ever.”

These are just a few of the more than 100 applications that leverage the immense graphics and parallel processing capabilities of the AMD A-Series APU;, the best entertainment and gaming experience. Check out more accelerated apps at www.amd.com/coolapps. Check out all our partners below for more:

• Adobe
• Arcsoft
• BlueStacks
• Codemasters
• Corel
• Cyberlink
• DICE
• DivX
• Eidos, division of Square Enix
• Fingertapps
• GIMP
• Handbrake
• Magix
• Microsoft
• Moovida Universe by Fluendo
• Musemage by Paraken Technology
• Nuvixa
• Sony Creative
• VLC
• vReveal by MotionDSP

Gabe Gravning is a Sr. Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

As we approach the six month anniversary of the award-winning GCN Architecture, we thought we’d take a minute to look at what has made this one of our most successful GPU architectures of all time, starting today with a high-level overview.

Now, when people ask us for the five-second version of the story, the GCN Architecture’s success really comes down to one thing: performance. Be it in absolute frames per second, visual fidelity, or efficiency—topics we’ll cover at length in the weeks ahead—every one of the many innovations in the GCN Architecture is all about giving you the best possible product.

Moving from last generation’s AMD Radeon™ HD 6970, the GCN Architecture’s radical re-design yields radical numbers: Up to 4.5x the compute performance, up to 2.5x the tessellation performance, and up to 95% lower idle GPU power consumption.^1-3

Massive Compute Performance

Enabling 350% better compute performance was a matter of building the right GPU at the right time.³ As open compute languages like OpenCL™ and C++ AMP take flight, Graphics Core Next helps AMD build the graphics cards destined to power the killer apps of tomorrow.

We achieved such a dramatic improvement by optimizing first for higher utilization of compute resources. At every opportunity, a product based on the GCN Architecture intelligently manages workloads in a way that yields more performance to applications.

Efficiency was a corollary goal. By transitioning to the industry’s first 28nm process, we could enable an architecture that not only packed more compute units into a given space, but more complicated and capable compute units. Any of our GPUs can offer more performance than analogous products from prior architectures, and our biggest dies offer a positively stunning 3.79 TFLOPS of compute performance with the AMD Radeon™ HD 7970.

Most importantly, we made this power easy to utilize. With support for major industry languages, like OpenCL™ and C++ AMP as I mentioned, we simplified coding for the GPU with standardized scheduling and optimizations. In other words, we strive to provide consistent and predictable performance to minimize the effort of designing GPGPU applications.

The proof lies in the incredible software being produced by AMD’s partners, each supporting an array of GPU-acceleration features.

From browsers to video editing, AMD Radeon™ products can leverage GPU acceleration in some of today’s best and brightest applications.

Built for ULTRA Settings

Today’s enthusiasts expect more of their GPU than ever before, and rightfully so. The last year has introduced a string of titles with fabulous graphics, like The Elder Scrolls V: Skyrim and Battlefield 3. When designing the AMD Radeon™ HD 7970, the ultimate form of the GCN Architecture, we knew we had to build a GPU that was primed to power games like these at ultra settings.

With new features like SSAA and MLAA 2.0, an industry-leading 3GB frame buffer, and blistering performance at resolutions like 5760×1080 (or higher), the AMD Radeon™ HD 7970 enables the experiences gamers deserve.

From 3GB of ultra-fast GDDR5 and a slew of new image quality features, our press agree that the AMD Radeon™ HD 7970 knows how to play the game.

The DNA of Efficiency

And who says a high-performance GPU can’t be green? Beginning with all of the 28nm AMD Radeon™ HD 7000 Series desktop products, we’ve designed and implemented a great feature called AMD ZeroCore Power technology.⁴

AMD ZeroCore Power allows a single GPU to virtually shut down when your monitor goes into the long-idle suspend state with an amber light. Instead of consuming 18-24 watts–as is the case with our past and competing products–AMD ZeroCore Power-enabled products consume less than 3W. That’s a reduction in idle power of up to 95%.³

AMD CrossFire™ technology users benefit even further. Upon exiting a game, AMD ZeroCore Power will dynamically migrate all secondary GPUs to the low-power AMD ZeroCore state. These secondary GPUs are available for activation at any time, but they consume virtually no power while they remain idle—even their fans stop spinning!

AMD ZeroCore Power technology is exclusive to AMD products, and reduces idle GPU power consumption by up to 95% versus past and competing products.

Just the Beginning

Today’s overview is just a taste of the cutting-edge technology woven into the fabric of the GCN Architecture. As we move through May into June, I’ll be exploring many and more of these topics in fine detail to fully illustrate how the world’s fastest and most advanced graphics leverages Graphics Core Next to enable truly next-gen experiences in gaming and compute.⁵

Never settle, friends.

Robert Hallock is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Substantiation:

1. Results based on average FPS measurements in SiSoftware Sandra 2010 Pro (AES256) benchmark at 1920×1080 comparing an AMD Radeon™ HD 7970 (23.1) vs. an AMD Radeon™ HD 6970 graphics (5.1) on the following system: Intel Core i7 (3.3GHz), MSI X79A-GD65, 16GB DDR3-1600 9-9-9-24, Windows 7 Ultimate RTM 64-bit. [GDRT-9]
2. Results based on average FPS measurements in Unigine “Heaven” benchmark at 2560×1600 set to extreme tessellation and no anti-aliasing comparing an AMD Radeon™ HD 7970 (80) vs. an AMD Radeon™ HD 6970  graphics (30.6) on the following system: Intel Core i7 (3.3GHz), MSI X79A-GD65, 16GB DDR3-1600 9-9-9-24, Windows 7 Ultimate RTM 64-bit. [GDRT-10]
3. Results based on average FPS measurements in SiSoftware Sandra 2010 Pro (AES256) benchmark at 1920×1080 comparing an AMD Radeon™ HD 7970 (23.1) vs. an AMD Radeon™ HD 6970 graphics (5.1) on the following system: Intel Core i7 (3.3GHz), MSI X79A-GD65, 16GB DDR3-1600 9-9-9-24, Windows 7 Ultimate RTM 64-bit. [GDRT-9]
4. AMD PowerPlay™, AMD PowerTune and AMD ZeroCore Power are technologies offered by certain AMD Radeon™ products, which are designed to intelligently manage GPU power consumption in response to certain GPU load conditions. Not all products feature all technologies – check with your component or system manufacturer for specific model capabilities.
5. Based on GCN Architecture, the world’s first 28nm graphics architecture, the AMD Radeon™ HD 7970 and HD 7950 graphics achieve 3DMark® 11 scores in excess of X2700 and X2200, respectively. These results are the two highest scores achieved by single-GPU graphics cards on stock platforms as of 26 January, 2012. System configuration: Intel Core i7-3960X (3.3GHz), MSI X79A-GD65 8D, 16GB DDR3-1600 (9-9-9-24) and Windows® 7 64-bit. Testing based on AMD Catalyst™ 8.921.2. The NVIDIA® GeForce® GTX 580, NVIDIA’s highest performing single-GPU graphics card, was evaluated using NVIDIA ForceWare 290.53 BETA and scored X2124. GRDT-4

The GCN Architecture and its associated features (PCI Express® 3.0, AMD ZeroCore Power technology, DDM Audio, and 28nm production) are exclusive to the AMD Radeon™ HD 7900, HD 7800 and HD 7700 Series Graphics.

With our upcoming launch of “Trinity” (now shipping!), we are excited to report that, in addition to an outstanding multimedia experience delivered by the AMD HD Media Accelerator, users can finally game on an ultrathin-and-light notebook with great battery life.   We don’t just mean play any old game at minimal-quality settings, either—you can play today’s hottest new titles, like Battlefield 3 from DICE and DiRT Showdown from Codemasters, in stunning high definition. “Trinity” powers a brilliant gaming experience, with an increase in graphics performance of up to 70% over the competition, thanks to the latest AMD Radeon graphics technology.[i] Suffice it to say, our partners at Codemasters are excited for “Trinity” as well!

“Graphics power today’s entertainment lifestyle and with the AMD A-Series APUs, customers will have a richer visual experience when playing DiRT Showdown,” said Clive Moody, Executive Producer, DiRT Showdown. “The arrival of the Second-Generation AMD A-Series APUs delivers a superior visual quality and outstanding performance with a full panel DirectX 11® gaming experience that’s second to none.”

If you’re like Codemasters and you can’t wait to get your hands on “Trinity” then we highly encourage you to take a look at the new HP ENVY Sleekbook, just announced by HP (pictures below). As thin as 19.8 mm and starting at less than 4 pounds, the new, 15.6” HP ENVY Sleekbook, powered by “Trinity,” delivers an incredible gaming and multimedia experience with outstanding battery life. For more on the HP ENVY Sleekbook, stay tuned to the HP ENVY home page, as the HP ENVY Sleekbook with AMD “Trinity” is expected to be available in the United States on June 20 with an MSRP starting at $599.99 per HP’s recent major announcement.

For years, PC gamers have clamored for an affordable, ultrathin-and-light notebook with great battery life that is capable of playing the latest games. With “Trinity,” gamers’ dreams are finally coming true!

Neal Robison is the director of ISV Relationship Management at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

[accordion_pane title="Get Dirt Showdown Free"]

Buy a select AMD product and receive a copy of Dirt® Showdown™ FREE!

• CyberPower
• iBuyPower
• Maingear
• NewEgg
  • Graphics Cards
  • Accelerated Processors
• Smooth Creations
• SuperBiiz
  • Graphics Cards
  • Accelerated Processors
• VelocityMicro

[/accordion_pane]

[accordion_pane title="Overview" current="yes"]
>> Play the demo on Steam!
DiRT Showdown is a brand new dive in and drive adrenaline rush of speed, style and destruction from the creators of the multi-award winning DiRT series. Players jump into a new world of arcade racing with pick up and play controls, speeding, tricking and smashing their way to ‘Showdown’ finals to compete against rivals in front of thousands of fans in a vibrant festival atmosphere.

There are three broad categories to DiRT Showdown’s stunning world of action-sport racing. Players will use nitrous to blast past rivals and negotiate courses filled with ramps, multiple routes and obstacles in racing events. Gaming’s most advanced damage engine is pushed to its limits in demolition derby events, where players smash and crash their way to victory in jaw-dropping, bone-jarring style. Finally, in Hoonigan events gamers can demonstrate their freestyle driving skill in huge free-roaming stunt parks with new accessible controls.

In the career mode, globe-trotting Showdown players will travel from Miami to San Francisco, London to Tokyo and other famous locations earning the adulation of the crowd at hyper-energised, frenzied, unsanctioned race events. Over 50 different events across four championships challenge gamers in a variety of conditions – sun, snow, and rain -- through the day and under the floodlights at night.

With an exciting mix of licensed and bespoke cars and powered by the EGO Game Technology Platform for phenomenal graphical performance and stunning damage, DiRT Showdown will be the new standard for arcade driving delirium.

www.youtube.com/watch?v=k7kqc69nVv0

[/accordion_pane]

[accordion_pane title="Features"]

• The talent that brought the world Colin McRae DiRT, Colin McRae DiRT 2, The BAFTA Award winning Race Driver GRID™ and DiRT 3 (Metacritic: 87) is developing DiRT Showdown. Using the award winning EGO Game Technology Platform known for its killer visuals, weather, and damage system, DiRT Showdown sets the new technical benchmark for arcade racing thrills.
• Accessible, pick up and play controls combine with spectacular events and stunning graphics to deliver high octane, dive in and drive thrills from event one.
• Ken Block’s all new Gymkhana 4 HFHV Ford Fiesta debuts alongside a varied selection of vehicles across a range of classes. From saloons and muscle cars to pick-up trucks, hearses and vans.
• Courses are littered with obstacles, pinch points, multiple-routes and ramps.
• Players will crash, smash and bash their way through a range of demolition derby themed events.
• A new accessible handling system lets players let loose in free-roaming, freestyle stunt parks. Perform trick runs to competitive stunt events, explorations challenges and more.
• Issue Showdown Challenges so players can test themselves against their friends whether they are online or not.
• From San Francisco to Miami, Nevada to Michigan, Tokyo to London, each of DIRT Showdown’s locations delivers a unique atmosphere for a wide range of events.
• Whenever a large crash happens, players can see ‘kill cams’ from several angles showing how the carnage unfolds. In single player events, gamers can also upload clips direct to YouTube.

[/accordion_pane]

[/accordion]

Developed by: Codemasters

Published by: Codemasters

© 2012 The Codemasters Software Company Limited (“Codemasters”). All rights reserved. “Codemasters”®, the Codemasters logo and “DiRT”® are registered trademarks owned by Codemasters. “DiRT Showdown”™ and “EGO”™ are trademarks of Codemasters. Ford Motor Company Trademarks and Trade Dress used under license to Codemasters. All other copyrights or trademarks are the property of their respective owners and are being used under license. Developed and published by Codemasters.

Why You Should Attend
AFDS is not a marketing event. It’s not about selling one brand or component. AFDS is where you discover the latest technology advancement from the experts who are redefining where the industry is headed.

Who Attends AFDS
The best in the industry converge at AFDS to share what they are working on and to collaborate with you. Here’s just a small sample of companies expected to attend AFDS. Microsoft, ARM, Square Enix, Nixxes Software, Corel, ALTELL, ArcSoft, MulticoreWare Inc, Adobe, ViVu, Sony Computer Entertainment, and Nuvixa.

What You’ll Learn
Get actionable, hands-on training in:

• OpenCL™
• Multi-GPU
• OpenGL
• GPU/APU

Claim your seat at our most popular sessions for gamers

Compute in the Future of Gaming
With DX11 capable hardware now the norm for gamer spec PCs and the next generation of consoles on the horizon, AAA games development is about to take a very interesting turn towards compute-based solutions.  This talk will offer an insight into some of the techniques now possible with compute and what this means for games over the next 5 years.

See David Perry, Chief Executive Officer & Co-Founder, Gaikai deliver his keynote Delivering Console Quality Video Games from the Cloud

Perry will present Gaikai’s progress on trying to help grow the video game industry by using virtual CPU and GPU cycles in the cloud.  This talk would be of interest to people interested in the topics: video games, high performance clouds, custom servers, broadband usage, IPTV, Facebook, Tablets, proximity networks, low latency experiences, low friction experiences & accessibility.

Learn more and Register

As a reminder, here are the sites you’ll need to check out to find all ten:

• http://developer.amd.com/afds/pages/default.aspx
• http://developer.amd.com/pages/default.aspx
• http://blogs.amd.com/fusion
• http://www.facebook.com/AMD/app_282531588467018
• http://semiaccurate.com/2012/04/16/two-new-keynotes-announced-for-afds/

For round two, we’re focusing on the AMD Facebook tab, the AFDS site and right here on the blog. Without further ado, here are the five clues to finding the remaining five “easter eggs”:

• Head over to the home page of the AMD Fusion blog. Keep your eye (and cursor) towards the header of the page. Clicking on a key word may remind you of “doing a barrel roll”. But, keep in mind, one click doesn’t always do the trick.
• Read about why you’ll want to attend AFDS, making sure to go through each of the key reasons, taking you all the way to the bottom of the page. There you’ll find one of AMD’s greatest social media resources.
• Check out the AFDS Facebook tab. Three times is the charm when you spot a key logo at the top of the page. Here, you’ll receive some additional clues that will help lead you to the remaining Easter Eggs.
• Also on the AFDS Facebook tab, look to “sign up and stay informed” to find the next hidden treasure. Submitting nothing at all may be a surprisingly smart move.
• In the same location as the previous clue, you’ll need to try something else to find the final egg. Where you are asked to provide your email address, you may want to enter your favorite open source programming language “For The Win”.

There you have it – combined with the first five clues, you should now have a clear map to help you find what you need in order to complete the AFDS Virtual Scavenger Hunt. Don’t worry, if you aren’t the first to find all ten, we will be giving away a Hewlett-Packard dv6 laptop with VISION Technology from AMD for second place, and an AMD  Radeon^TM HD 7970 series GPU for third place.

Questions? If you’ve got ‘em, leave them in the comments or you can find us on Facebook and Twitter. Good luck!

Travis Williams is a Public Relations Representative at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

NO PURCHASE NECESSARY TO ENTER OR WIN.  A PURCHASE OF ANY KIND WILL NOT INCREASE YOUR CHANCES OF WINNING.  Open to residents of the 50 US States (DC) and Canada (excl. Quebec), 18+.  Contest begins April 23, 2012, and ends May 21, 2012.  For complete contest rules and details, visit here.

Riley Lewis is like a lot of kids today in that he spends hours in front of a computer screen.  What makes him unique is the professional grade software and hardware that is a regular part of his busy schedule.  Already an avid inventor, Riley started using SolidWorks® at the age of 11, when he used the program to do a failure analysis of his team’s entry in the prestigious Tech Challenge and helped make the design into an award winner at the event.  His participation there caught the eye of SolidWorks’ CEO, and in a blink of an eye Riley was headed to the annual SolidWorks users’ conference to learn more about his new-found passion.

Since then, with the help of AMD and other contributing companies, Riley has worked with his proud father and peers to establish a well-equipped 3D design lab for his charter school, and established an elective 3-month course on the school’s curriculum.  The lab includes 3D printer technology that allows Riley and other students to take their designs to actual prototypes during the course.

At the center of this unique learning environment are the workstations using AMD FirePro™ professional graphics cards that the students use to execute their designs in SolidWorks.  With two AMD FirePro cards in the lab, Riley and his friends have taken advantage of the tremendous graphics processing horsepower at their disposal.

“In the class we’ve designed speaker enclosures to make them sound better,” explained Riley recently.  “As part of that we did a lot of rendering work to see how it would finally look.  The AMD cards rendered things faster than anything we had used before.”

Antoine Reymond, AMD strategic alliances manager, has worked with Riley since first meeting him at the SolidWorks event and came away impressed with his tenacity to obtain the best hardware available, and his drive to share his knowledge with others his age.

“I was immediately impressed by Riley’s capacity to understand engineering problems when I first met him.  To do the kind of engineering work he is doing in SolidWorks while considering design for manufacturability or rapid prototyping is pretty amazing at any age,” says Reymond.  “It’s very rewarding to help young people who are passionate about technology in the context of sustainability and responsibility for in the world we live in.”

Next up for Riley, in-between focusing on making it through eighth-grade, is a full slate of engagements heading into the summer as he works to get others his age enthusiastic about science, design and inventing.  Future plans include a degree in mechanical engineering and at least two start-up businesses he is already planning.

We wish him all the luck and will continue to make sure that AMD FirePro professional graphics are ever-ready to back up his vision!

Bahman Dara is senior manager, worldwide marketing for AMD Professional Graphics. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

1. CodeAnalyst application runs as a native 64-bit application on 64-bit Windows OS flavors.

2. We have made several optimizations to help improve overall CodeAnalyst experience a pleasant one.

3. We have fixed quite a lot of bugs to help improve overall stability of CodeAnalyst.

We hope that this release helps you with your work efficiently and pleasantly. If you encounter across any bugs or see where AMD CodeAnalyst could be improved, please reach us through our forums or by replying this blog page.

OpenCL and the OpenCL logo are trademarks of Apple Inc. used with permission by Khronos.

Read the story here:

Try to remember how your PC looked 10 years ago. Mine was grey, beige and had a green button for power. Today, the computer’s physical appearance and performance (eg. cooling) is just as important as the computational performance you get on screen. With an AMD Radeon™ HD 7970 graphics card you’re getting amazing hardware that only deserves a matching setup.

In this article I’m going to look at a couple builds that are definitely not dull grey boxes. We’ve taken our AMD CPU, RAM, and Video Cards and installed them in 3 different chassis. I just need to throw in a hard drive and wire everything together and I’ve got myself a LAN party.

All images are for illustrative purposes only and no endorsement of AMD or by AMD is implied.

Common Hardware:

Processor:  AMD FX 8150 Processor with Liquid CPU Cooling System

RAM:  AMD  Radeon™ Memory Entertainment Edition 4GB DDR3 x 2

Video Cards:  AMD Radeon™ HD 7970 graphics card x 2

Motherboard:  ASUS Crosshair V Formula

For all the following builds I’ll be using AMD hardware. This includes everything from the AMD Radeon™ HD 7970 Graphics card to the AMD branded Liquid CPU Cooling System.

On to our first setup…

Setup A:
Chassis: Cooler Master Cosmos II
PSU: Seasonic Platinum Series 860W

When the ATI Radeon™ HD 5970 GPU was released 2009, I was in awe. The ATI Radeon™ HD 5970 GPU was a beast; not only in sheer performance, but also in physique, spanning just over 12 inches. This graphic card was definitely no lightweight and was troublesome (if not impossible) to fit into certain chassis.

Fast forward to 2012, we have the AMD Radeon™ HD 6990 GPU and the AMD Radeon HD 7970 GPU. I measured these boards to be roughly 12 and 11 inches in length respectively. So for the moment the largest cards haven’t gotten bigger, but do we still have the same trouble fitting these graphics cards into a chassis?

With our first setup, there is definitely no issue fitting the AMD Radeon™ HD 7970 GPU. The race-car inspired Cosmos II is a self-proclaimed Ultra-Tower. Supporting up to 10 fans and 13 HDDs, as you can see below the chassis also comfortably fits 2 of our AMD Radeon HD 7970 GPU’s in AMD CrossFireX™ technology mode.

Setup B:
Chassis: Enermax Fulmo & Enermax Fulmo GT
PSU: Enermax Revolution87+ 850W

Sure, extra space is great in chassis’ for routing cables and better airflow, but if you’re like me and living space is a bit more limited, opting for a Mid-Tower over a Big Tower is a reasonable compromise.

Our 2 x AMD Radeon™ HD 7970 GPUs fit quite well in Fulmo (Mid Tower) chassis. For those of you still debating whether to go mid or big, Enermax has made it easier (or more difficult) to make the decision with a nearly identical bigger brother Fulmo GT (Big Tower).

Setup C:

Chassis:  Thermaltake Level 10 GT

PSU:  Thermaltake Toughpower Grand 1200W

A big focus on this article has been on fitting the components inside the chassis’, but what if you’re looking for maximum style points? For those who want the red and black color scheme and curvy style inside and out. The Level 10 GT is a definite option.

Another thing to point out is that I’ve paired each of these chassis’ with a matching power supply unit. All selected PSUs have also gone through our AMD Gold Certification to ensure they are up to the task of running our graphic cards in AMD CrossFire™ Technology mode, but I’ll throw out some brownie points to Thermaltake with their Toughpower Grand 1200W. The design complements our AMD hardware quite nicely.

It also comes in white. (Snow Edition)

These enthusiast chassis’ definitely had little issue fitting our AMD hardware, and look great while doing so. So which setup is your favorite?

David Tjong is on the Technical Marketing team at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

The sequel to Rebellion’s highly-acclaimed 2005 game Sniper Elite, Sniper Elite V2 continues to offer gamers the most authentic World War II sniping experience by putting them in the middle of war-torn Berlin amidst the frenzied final battle between German and Russian forces.

With an emphasis on learning to handle the anticipation and pressure of the hunt, players are forced to use skill, patience and cunning to achieve their missions as they find themselves facing life-altering choices.

The celebrated “bullet cam” from Sniper Elite makes its return and will be more gratifying than ever. Skilful shots are rewarded with a slow-motion sequence of the bullet running its full trajectory from barrel to target as well as featuring graphic detail of the effects of the bullet entering and destroying a target.

www.youtube.com/watch?v=TRJ4LkrZSeo

Features:

• Detailed sniping simulation with advanced ballistics, taking into account gravity, wind, velocity, bullet penetration, aim stability and more.
• Amazing “kill cam” technology showcases what really happens when a bullet enters an enemy’s body.
• Authentic World War II Berlin locations as well as vehicles, weapons, and uniforms modelled after the original versions.
• AMD Eyefinity support gives you full awareness of your surroundings.
• DirectX 11 support provides the ultimate visual treat.

Screenshots:

Official Website: SniperEliteV2.com

©2012 Rebellion. The Rebellion name and logo and the Sniper Elite name and logo and the Sniper Elite Eagle are trademarks of Rebellion and may be registered trademarks in certain countries. All rights Reserved. Uses Bink Video. Copyright © 1997-2012 by RAD Game Tools, Inc.

Since the launch of the AMD Radeon™ HD 6000 series, AMD introduced support for DisplayPort 1.2, but it hasn’t been until CES 2012 where the technology has been demonstrated with the AMD GPUs.

Let’s take a look at some of the display technologies that were shown at CES this year.

Multi-Display Technology
Notebooks are great on the road: Portable, Wireless, and Convenient; but when you bring it home, why limit yourself to just with just one monitor? When you can have 4?

Don’t worry about messing with multiple cables. We’re talking about a single cable from your notebook. DisplayPort 1.2 enables this with Multi Stream Transport (MST) Hubs and Daisy-Chain monitors. Don’t get these confused with the DisplayPort 1.1 MST Hubs already in the market. DisplayPort 1.2 MST Hubs have support for monitors of different size, resolution, and orientation. (eg. 2 x FHD in portrait & 1 x WQXGA in landscape)

There’s no magic going on here. As you can see we have 3 monitors each individually connected to our DisplayPort 1.2 MST Hub. The only cable that is connected to the notebook is the connector on the DisplayPort 1.2 MST Hub.

Don’t get too excited just yet. This illustrated image of the DisplayPort 1.2 MST Hub is just a mockup of the possible future design. The current prototype is significantly larger and requires an external power adapter. It is expected that this technology would get smaller as the product is finalized, or perhaps even disappear if monitor manufacturers embed this technology directly into the monitor. (ie. Daisy Chaining)

4K2K resolution

Looking to move onto the next generation of display resolution?

At CES, one of the buzzwords floating around was 4K2K. 4K2K loosely refers to 4060×2160 or 3840×2160 (QFHD) resolutions, and various TV manufacturers showed their 4K2K TVs. It’s now the waiting game for them to be released. If you’re eager for a 4k2k monitor today you can pick up the Eizo Radiforce RX840 for a roughly $38,000. Although this model requires 2 DisplayPort 1.1 cable inputs.

With DisplayPort 1.2 bandwidth is effectively doubled, so why do 2 cables when we can do 1? At CES, VESA demonstrated a modified version of the monitor that accepted a single DisplayPort 1.2 cable input. That’s 4096×2160@60Hz resolution with a single cable. Needless to say, that’s a lot of pixel going through a single cable.

HDMI (Yes, we haven’t forgotten about you)
Haven’t jumped on the DisplayPort bandwagon yet? Or do you want to cling onto your HDMI cables a bit longer? The AMD Radeon™ HD 7000 series has your answer for that too. Supporting the optional implementations in the HDMI specification, the AMD Radeon™ HD 7000 series supports up to 4096×2160@24Hz. You’ll also need a TV/Monitor that supports the same optional implementation in the HDMI specification. For example, available in the market is the Sony VPL-VW1000ES 4K2K projector that will support 4096×2160@24Hz over HDMI.

Keep in mind, with a maximum refresh rate of 24Hz, it’s recommended only for media consumption such movies or videos. If you’re looking for a smooth gaming experience you’ll want at least a 60Hz refresh rate.

So what Display Technology are you most looking forward to?

David Tjong is on the Technical Marketing team at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

As part of AMD’s commitment to developer tools for heterogeneous compute platforms, I am excited to introduce you to gDEBugger 6.2.

gDEBugger 6.2 is a key milestone that adds Linux® support and a new standalone user interface that is available for both Linux® and Windows®. The major highlights of this release are:

• Support for Red Hat®, Ubuntu® and OpenSUSE™ Linux® distributions
• New standalone user interface with enhanced GUI for ease of use and better navigation
• Support for OpenCL™ kernel and API level debugging on AMD Radeon HD 7000 series graphics cards
• Support for OpenCL™ 1.2 beta drivers
• Stability and feature enhancements along with updated Microsoft® Visual Studio® Plugin

You can visit gDEBugger landing page to get more details and download it.

We value your input. If you have suggestions on how to improve our tools or if you experience any issues, let us know through our forums or comments to this blog.

Milind Kukanur is a Sr. Manager, Product Management at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

AMD’s second generation architecture of  core processors  codenamed “Bulldozer” was released in 2011 (products included  AMD Opteron™ 6200, AMD Opteron 4200 and FX  Processors).  These processors include a more advanced hardware prefetcher which is able to pick up more varied data access patterns than previous generation processors. 

Oracle JVMs by default use software prefetch instructions to prefetch heap memory when allocating new objects on the heap. (Java applications tend to do a lot of heap allocations).  Sometimes the software and hardware prefetching can get in each other’s way. We discovered that for most Java workloads the best performance on Bulldozer family processors could be achieved by just using the hardware prefetcher and disabling  the software prefetching.

AMD worked with Oracle to get this no software prefetch strategy into the Java 7 release (released in July 2011) as the default for these Bulldozer processors.  For those still using Java 6 or earlier releases, the same effect can be achieved by explicitly using the following option on the java command line: –XX:AllocatePrefetchStyle=0.

Tom Deneau is a Senior Member Technical Staff in the Runtimes Team at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.