HPCwire: Changing the Game

The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing

HPCwire >> Blogs

Blog: From the Editor

January 15, 2009

Changing the Game

In my comments earlier this week on AMD's purpose-built "Fusion Render Cloud" supercomputer, I neglected to mention a possible downside for AMD's GPU business. In a nutshell, if this new supercomputer is going to be doing all the heavy lifting rendering-wise in the server, why do you need GPUs in the client?

The issue is probably more obvious when you realize that the supercomputer is being built with essentially the same "Dragon" chipset destined for high-end multimedia PCs. Specifically, it's the ATI Radeon HD 4800 GPU in the chipset that delivers all the nifty HD multimedia capabilities coveted by hard-core gamers and video enthusiasts. And it's not just for supercomputers and desktop machines. On Wednesday, AMD introduced a slightly less powerful offshoot of the HD 4800, the ATI Mobility Radeon HD 4000 series GPUs. These chips are aimed at the notebook market and promise to deliver "a home theatre-quality HD multimedia experience."

But if AMD's petaflop rendering monster (containing 1,000 Radeon HD 4800 GPUs) is truly able to deliver a cutting-edge multimedia experience to low-end PCs, then why buy the expensive box at all? And since the 1,000 GPUs in the supercomputer will probably be utilized more efficiently than in a 1,000 separate PCs, overall AMD will need to manufacture fewer of them to deliver the same computational performance.

Right now AMD is probably more focused on the upside of the Render Cloud. Since any device smaller than a notebook (netbooks, mobile phones, etc.) is not likely to house a CD/DVD drive for multimedia -- not to mention a discrete graphics processor -- the mobile market presents a natural opportunity for HD streaming, without the threat of cannibalizing current GPU revenue.

On the other hand, multimedia notebooks and desktops could get trapped in the crossfire. Users might decide to jettison the pricey GPUs (and DVDs) in favor of streamed multimedia content for the sake of convenience. Not only could gamers stop buying DVDs at $29.99 a pop, they also wouldn't have to upgrade their machine every time a graphics processor came out promising the latest whiz-bang special effects. Instead the cloud would get the upgrade, while the thin clients automagically pick up the new capabilities. So is there room for high-end GPUs on both the client and the server?

I guess the answer revolves around the pricing structure of content serving versus media ownership. Presumably content providers are planning to use some sort of subscription service to deliver games and other HD content from the Render Cloud. Since most online games and HD media on the Internet are currently available for free, there is a lot of pressure to keep prices low. But how low?

A model that already exists for something like this is Amazon's Video On-Demand, an online service in which digital video content can be purchased for lifetime ownership or merely rented for 24 hours. The content can be viewed offline (downloaded) or online (streamed) and can be run on a PC in a standard Web browser, or on other devices, even TVs. Using the Video On-Demand service, renting the "The Dark Knight" movie for a day costs $3.99, while buying it costs $14.99. Alternatively, if you want to own the movie as a DVD, the price is $20.99. The Video On-Demand means the DVD player -- the content renderer, in this case -- has become superfluous.

Obviously, content providers and chipmakers don't have the same interests. Game companies and other media developers will find a way to make money from their intellectual property, even when its form or distribution changes. But if AMD is going to serve up graphics computation on demand, it's risking making the client-side GPU hardware redundant. Of course, if the company can convince users that there is unique value for both client- and server-side GPUs, then problem solved. I guess that's why God made marketing departments.

Posted by Michael Feldman - January 15 @ 5:24PM

(Digg, Technorati, more)

Discussion

There are 1 discussion items posted.

Fusion Cloud Rendering
Submitted by gary@amd on 01/16/2009 - 2:13PM

The question on whether the Fusion Cloud Rendering supercomputer might decrease demand for client GPUs is a logical one, but our thinking is, not surprisingly, "no it won't." Why? A few reasons:

- Offline play and content viewing will still require strong GPU on client side
- Decompression on the client side will still benefit from a GPU
- Overall user experience improves with GPU horsepower being on the client side too
- Useage model could provide "cheaper" access to the apps and content if a strong client GPU lessens demands on the Fusion Cloud
- Increasing useage of the GPU to supplement the CPU for "non-traditional" GPU tasks. ATI Stream applications are now supported in Catalyst driver for this purpose.

Post #1

Join the Discussion

You must be a registered user in order to comment on HPCwire stories.

Members enjoy:

	Access to Special Content, including Podcasts, Webcasts, and Whitepapers and more!
	Ability to take an active role in the discussions that are shaping the HPC technology discipline!
	Special offers available only to HPCwire members!

Become a Registered User Today!

Registered Users Log in Here to Comment

Email Address:	Password (case sensitive)

Remember me	Forgot Password?

Michael Feldman

Michael Feldman is the editor of HPCwire.

More Michael Feldman

Feature Articles

The Week in Review

The ACM Turing Award goes to the creator of the modern personal computer; and Voltaire announces a mid-range InfiniBand switch and new technology that accelerates distributed applications. We recap those stories and more in our weekly wrapup.
Read More...

Florida State Gives Virtual SMPs a Spin

The prospects for virtual SMP technology got another boost last month when Florida State University announced it had installed a new HPC system from 3Leaf Systems. The servers are being housed at the university's HPC facility and will be used across a range of scientific disciplines.
Read More...

HPC Powers Bobsled Team to Olympic Gold

For the first time in 62 years, the four-man Olympics bobsled team from the US captured the gold medal, setting a course world record in the process. The winning bobsled had some state-of-the-art engineering behind it, including CFD software from Exa Corporation. As it turned out, that software may have proved to be the margin of difference in the race.
Read More...

Read more HPCwire features...

Top Headlines

GP-GPUs: OpenCL Is Ready For The Heavy Lifting

Mar 11 | Linux Magazine | CUDA may be the rage, but OpenCL is a standard that has some features you may need. Read more...

Can Free Software Drive the Fourth Paradigm?

Mar 09 | Free Software Magazine | Data-driven computing will need open software. Read more...

Graphics Card Maker Turns to High-Performance Bioinformatics

Mar 09 | Bio-IT World | Tahoe Informatics founder eyes GPUs, CUDA software. Read more...

CFD: Light at the End of the Tunnel?

Mar 08 | Sporting Life | Formula One engineers differ on benefits of CFD. Read more...

AMD Tries to Draw Intel Into Chip Battle

Mar 08 | InfoWorld | AMD offers up 48-core server prize. Read more...

Read more headlines...

Featured Whitepapers

Virtualization for Aggregation And The vSMP Architecture™

Jan 12 | | In-depth look at vSMP Foundation server virtualization technology, technical implementation, use cases and capabilities. The technical whitepaper provides an architectural overview and details on the three vSMP Foundation products: vSMP Foundation for SMP, vSMP Foundation for Cluster and vSMP Foundation for Cloud.

Copper Cable Technologies for High Performance Computing

Jan 18 | | This white paper discusses Gore’s copper cable assemblies, and how they continue to exceed the standards for providing reliable, cost-effective solutions for high-performance computer applications.

View the White Paper Library

Multimedia

Webcast: Virtualized Data Center Roundtable

Join this online panel discussion for live Q&A with leading industry experts, analysts, and end-users to discuss the latest innovations, best practices, barriers to implementation, and measurable benefits of server virtualization with a particular focus on today's real world solutions.

Webcast: Watch SC09 Birds of a Feather Video: Scalable Fault-Tolerant HPC Supercomputers

Learn about scalable fault-tolerant architectures and examples of energy efficient and scalable supercomputing clusters using dual QDR InfiniBand to combine capacity computing with network failover capabilities with the help of programming languages such as MPI and a robust Linux cluster management package.

Webcast: High Performance Computing for a Smarter Planet

LIVE@SCO9: The IBM team discusses new innovations in hardware, software and services that help clients better understand their workloads and get insight from their R&D efforts. Technology demonstrations include the soon-to-be-released Power7 HPC processor, the DCS990 system with 2.4 petabytes of storage, the xCAT management tool, secure HPC cloud computing and more. Winners of two HPCwire Readers' and Editors’ Choice Awards! Take the IBM virtual tour at SC09 or more information go online to: http://www-03.ibm.com/systems/deepcomputing/sc09.html