Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Remember the good ol' days...
AMD share price continues to rise
Rollins, rollins, rollins
By Cher Price: Friday 12 November 2004, 15:53
* UPDATE It's $20.25 now, having risen $1.66 during the day's trading. INTC wobbled, rallied, and is now $23.39.
THE SHARE PRICE of AMD continued to rise steeply on Wall Street, after Wells Fargo rated the stock as a "buy".
At press time, it stands at $19.54, although it had reached $19.72 earlier on. On Wednesday this week it stood at $17.29. On that day, Lehman Brothers gave the firm the thumbs up.
Yesterday Merrill Lynch said there was room for the shares to grow, but Smith Barney reckoned the NOR flash memory market could drag things down.
Meanwhile, Intel (tick: INTC) opened at $23.03, down a little despite bullish results delivered by its major distributor Dell yesterday night.
No doubt an Infoworld story gave it a further boost, but whether Dell ever adopts the Opteron or not, such stories in the trade press do allow it to put pressure on Intel. Dell is Intel's biggest customer.
Over the last few years, AMD staff have routinely trekked into Dell's offices to pitch their processors, and Intel knows all about that. µ
"someone is going to buy them out "
Nobody is going to buy out AMD.
The AMD faithful predicted this very same fairytale in 1996 and 1997 when AMD went spinning into losses, and they repeated the fairytale again in 2001 and 2002.
The blind-faithful have predicted Motorola (now out of semis), IBM, TI, Samsung, Fujitsu, SUNW, etc.
None of this ever came to pass.
And it never well.
Here's a tip - there ain't no Santa Claus.
AMD can't make real money in a period where the blind faithful think that AMD has a clear product superiority over Intel at the same time that Intel has stumbled with product recalls, product cancellations, product delays, and complete roadmap changes.
If AMD couldn't solidify their gains when they (AMD} had a clear upper hand over Intel - as written about ad nauseum by the various intellegentsia in the technical news media and talking heads such as Brookwood and McCarron - then God help AMD when Intel gets its product roadmap corrected and into full swing later this year.
"The very example of a seagull poster.
10/29/2004
9/5/2004
7/21/2004
2/16/2004"
I am flattered - you follow my every post.
Now that I know how popular I am with you, perhaps I will find more time to post to you on this thread about the wonderful and magnificent goings-on at AMD and how the rose-coloured-glaases-crew on this forum continue to believe only fantasies about AMD and not the reality that is this sorry, sorry company.
Remember - Q4 of every calendar year is generally the "best" for semiconductor companies, and sales - and profits - head south in Q1 and Q2. So buckle up your seat belt, laddie, and prepare for AMD's come-to-Jesus confessional next week - and their impending spiral back into losses.
Yes, losses ahead, matey!
MirroBit MirrorBit On the Wall --- looks like profits are gonna fall !!!
" the Memory Group is anticipated to have lower fourth quarter sales and an operating loss. "
But I'm sure "ornand" will now accelerate AMD into profitable orbit in this very same quarter !!!
"Even Intel amazed at their ability to turn bull$hit into ca$h:"
Sorta dovetails with AMD's 35 years of turning ca$h into bull$hit !!!
AMD has zero net profit since it was founded 35 years ago.
Looks like AMD just preannounced a semi-disastrous quarter - in light of Intel's upside strong preannouncement froma few months ago.
Will niceguy767 withdraw all his pi-in-the-sky hype about AMD's Q4 2004 earnings now - or will just see blue skies ahead?
Press Release Source: AMD
AMD Updates Fourth Quarter Outlook
Monday January 10, 7:20 pm ET
SUNNYVALE, Calif.--(BUSINESS WIRE)--Jan. 10, 2005--AMD (NYSE:AMD - News) today announced sales for the fourth quarter ended December 26, 2004 are expected to be up slightly from the $1.239 billion reported in the third quarter. Fourth quarter operating income is anticipated to be positive, but down significantly from third quarter operating income of $68.4 million. The Computation Products Group (CPG) continued its upward trajectory and fourth quarter sales are expected to increase compared to the prior quarter. CPG is expected to be profitable. In a competitive and challenging NOR Flash memory market, the Memory Group is anticipated to have lower fourth quarter sales and an operating loss.
The company's prior guidance for the fourth quarter anticipated overall sales to increase, driven by processor sales that exceeded seasonal trends, and Flash memory sales that were flat to up.
AMD will report its fourth quarter 2004 results after market close on Tuesday, January 18, 2005. AMD will hold a conference call on this day for the financial community at 2:30 PM PT to discuss fourth quarter financial results. A real-time audio broadcast of the teleconference will be provided at www.amd.com and www.streetevents.com. The webcast will be available for ten days after the conference call.
CAUTIONARY STATEMENT
This release contains forward-looking statements, which are made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. Investors are cautioned that forward-looking statements in this release involve risks and uncertainty that could cause actual results to differ materially from current expectations. We urge investors to review in detail the risks and uncertainties in the company's Securities and Exchange Commission filings, including but not limited to the Annual Report on Form 10-K for the year ended December 28, 2003, and the Quarterly Report on Form 10-Q for the quarter ended September 26, 2004.
ABOUT AMD
AMD (NYSE:AMD - News) designs and produces innovative microprocessors, Flash memory devices and low-power processor solutions for the computer, communications and consumer electronics industries. AMD is dedicated to delivering standards-based, customer-focused solutions for technology users, ranging from enterprises and governments to individual consumers. For more information visit www.amd.com.
AMD, the AMD Arrow logo, AMD Athlon, AMD Opteron, AMD Geode, and combinations thereof, are trademarks of Advanced Micro Devices, Inc. Spansion and MirrorBit are trademarks of Spansion LLC. Other names used are for identification purposes only and may be trademarks of their respective owners.
--------------------------------------------------------------------------------
Contact:
AMD
Dave Kroll, 408-749-3310
Editorial
Email: dave.kroll@amd.com
or
Mike Haase, 408-749-3124
Investors
Email: mike.haase@amd.com
or
Ruth Cotter, 408-749-3887
Investors
Email: ruth.cotter@amd.com
"Intel Smithfield Dual Core could ship earlier than expected, in the first half of this year!"
I wonder if the Smithfield Dual Core will ship before AMD's
"Me-2"-Rion" Centrino copycat.
Transmeta - exit stage left.
Press Release Source: Transmeta Corporation
Transmeta Corporation to Increase Focus on Intellectual Property Licensing in 2005 - Translation - we're gettin' out of the CPU business !!!
Tuesday January 4, 5:00 pm ET
Company Will Complete Evaluation and Strategic Modification of its Business Model in January 2005
SANTA CLARA, Calif.--(BUSINESS WIRE)--Jan. 4, 2005-- Transmeta Corporation (NASDAQ:TMTA - News), the leader in efficient computing, today announced that it will increase its business focus on the licensing of its intellectual property and advanced technologies in 2005. Transmeta also announced that it will complete a critical evaluation of the economics of its current business model of designing, developing and selling x86-compatible microprocessor products, including its Crusoe® and Efficeon(TM) families, in January 2005.
ADVERTISEMENT
"Transmeta has spent a decade developing world class microprocessor technology, microprocessor products and related intellectual property," stated Matthew R. Perry, president and CEO, Transmeta Corporation. "By modifying our business model to focus more on our licensing opportunities, leveraging our substantial IP portfolio and our R&D capabilities, we would expect to reduce our cash needs substantially and to improve our results for our shareholders."
Over the last two years, Transmeta has worked to establish a revenue stream based upon the licensing of its proprietary technology and intellectual property. During 2004, Transmeta granted licenses to NEC Electronics and Fujitsu Limited to use Transmeta's proprietary LongRun2(TM) technologies for power management and transistor leakage control. Those licensing agreements include deliverable-based technology transfer fees, maintenance and service fees, and subsequent royalties on products incorporating the licensed technologies. Transmeta is engaged in active discussions with various industry leaders to license its advanced power management technologies in 2005. In addition, the company has entered into active discussions toward the potential licensing of other proprietary technologies, including its microprocessor designs.
The company's management is completing a critical review of its current business model, including an evaluation of its customer requirements and the economics and competitive conditions in the market for x86-compatible microprocessors. Transmeta is actively engaged in discussions regarding strategic alliance opportunities for leveraging the company's industry-leading microprocessor design and development capabilities to increase shareholder value. Transmeta has been working with Perseus Group LLC, a San Francisco-based investment banking firm, and other advisors, in that effort.
Transmeta management expects to complete its business model evaluation and planning with the company's Board of Directors in January 2005. The company presently expects to make an announcement and hold a conference call regarding its 2005 business model on January 21, 2005.
About Transmeta Corporation
Founded in 1995, Transmeta Corporation designs, develops and sells highly efficient x86-compatible software-based microprocessors that deliver a compelling balance of low power consumption, high performance, low cost and small size. The company's products are valuable for diverse computing platforms demanding energy efficiency, low heat and x86 software compatibility. Transmeta also develops advanced power management technologies for controlling leakage and increasing power efficiency in semiconductor and computing devices. To learn more about Transmeta, visit www.transmeta.com.
Safe Harbor Statement
This release contains forward-looking statements made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. Such statements speak only as of the date of this release, and we will not necessarily provide updates of our projections or other forward-looking statements. Investors are cautioned that such forward-looking statements are subject to many risks and uncertainties, and may differ materially or adversely from our actual results or future events. Important risk factors that could have material or adverse effects on our results include general economic and political conditions and specific conditions and volatility in the markets that we address, the rescheduling or cancellation of significant customer orders, market acceptance and adoption of our new products by our present and future customers and end users, difficulties in developing or manufacturing new and existing products in a timely and cost effective manner, our dependence on third parties for sourcing materials and providing manufacturing services, intense competition and competitive pressures, the ability to enter strategic collaborations or raise financing on satisfactory terms, the risks and costs inherent in any efforts that we may undertake under a modified business model, our ability to recognize significant licensing revenue in the future, patents and other intellectual property rights, and other risk factors. We urge investors to review our filings with the Securities and Exchange Commission, including our most recent reports on Forms 10-K, 10-Q and 8-K, which describe these and other important risk factors that could have an adverse effect on our results. We undertake no obligation to revise or update publicly any forward-looking statement for any reason.
Transmeta, Crusoe, Efficeon and LongRun2 are trademarks of Transmeta Corporation. All other product or service names mentioned herein are the trademarks of their respective owners.
--------------------------------------------------------------------------------
Contact:
Transmeta Corporation
Mark R. Kent, 408-919-3000
or
Mozes Communications
Kristine Mozes, 781-652-8875
Kristine@mozescomm.com
Re: " Is there anything happening at TMTA?"
Oh yes, bubby.
The end is finally near - Transmeta will be exiting the CPU business in a few short weeks, after burning through hundreds of millions of dollars of investors' money.
Looks like the reamining investors want a halt to the cash burning.
Read on, poor soul.....
Press Release Source: Transmeta Corporation
Transmeta Corporation to Increase Focus on Intellectual Property Licensing in 2005
Tuesday January 4, 5:00 pm ET
Company Will Complete Evaluation and Strategic Modification of its Business Model in January 2005
SANTA CLARA, Calif.--(BUSINESS WIRE)--Jan. 4, 2005-- Transmeta Corporation (NASDAQ:TMTA - News), the leader in efficient computing, today announced that it will increase its business focus on the licensing of its intellectual property and advanced technologies in 2005. Transmeta also announced that it will complete a critical evaluation of the economics of its current business model of designing, developing and selling x86-compatible microprocessor products, including its Crusoe® and Efficeon(TM) families, in January 2005.
ADVERTISEMENT
"Transmeta has spent a decade developing world class microprocessor technology, microprocessor products and related intellectual property," stated Matthew R. Perry, president and CEO, Transmeta Corporation. "By modifying our business model to focus more on our licensing opportunities, leveraging our substantial IP portfolio and our R&D capabilities, we would expect to reduce our cash needs substantially and to improve our results for our shareholders."
Over the last two years, Transmeta has worked to establish a revenue stream based upon the licensing of its proprietary technology and intellectual property. During 2004, Transmeta granted licenses to NEC Electronics and Fujitsu Limited to use Transmeta's proprietary LongRun2(TM) technologies for power management and transistor leakage control. Those licensing agreements include deliverable-based technology transfer fees, maintenance and service fees, and subsequent royalties on products incorporating the licensed technologies. Transmeta is engaged in active discussions with various industry leaders to license its advanced power management technologies in 2005. In addition, the company has entered into active discussions toward the potential licensing of other proprietary technologies, including its microprocessor designs.
The company's management is completing a critical review of its current business model, including an evaluation of its customer requirements and the economics and competitive conditions in the market for x86-compatible microprocessors. Transmeta is actively engaged in discussions regarding strategic alliance opportunities for leveraging the company's industry-leading microprocessor design and development capabilities to increase shareholder value. Transmeta has been working with Perseus Group LLC, a San Francisco-based investment banking firm, and other advisors, in that effort.
Transmeta management expects to complete its business model evaluation and planning with the company's Board of Directors in January 2005. The company presently expects to make an announcement and hold a conference call regarding its 2005 business model on January 21, 2005.
About Transmeta Corporation
Founded in 1995, Transmeta Corporation designs, develops and sells highly efficient x86-compatible software-based microprocessors that deliver a compelling balance of low power consumption, high performance, low cost and small size. The company's products are valuable for diverse computing platforms demanding energy efficiency, low heat and x86 software compatibility. Transmeta also develops advanced power management technologies for controlling leakage and increasing power efficiency in semiconductor and computing devices. To learn more about Transmeta, visit www.transmeta.com.
Safe Harbor Statement
This release contains forward-looking statements made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. Such statements speak only as of the date of this release, and we will not necessarily provide updates of our projections or other forward-looking statements. Investors are cautioned that such forward-looking statements are subject to many risks and uncertainties, and may differ materially or adversely from our actual results or future events. Important risk factors that could have material or adverse effects on our results include general economic and political conditions and specific conditions and volatility in the markets that we address, the rescheduling or cancellation of significant customer orders, market acceptance and adoption of our new products by our present and future customers and end users, difficulties in developing or manufacturing new and existing products in a timely and cost effective manner, our dependence on third parties for sourcing materials and providing manufacturing services, intense competition and competitive pressures, the ability to enter strategic collaborations or raise financing on satisfactory terms, the risks and costs inherent in any efforts that we may undertake under a modified business model, our ability to recognize significant licensing revenue in the future, patents and other intellectual property rights, and other risk factors. We urge investors to review our filings with the Securities and Exchange Commission, including our most recent reports on Forms 10-K, 10-Q and 8-K, which describe these and other important risk factors that could have an adverse effect on our results. We undertake no obligation to revise or update publicly any forward-looking statement for any reason.
Transmeta, Crusoe, Efficeon and LongRun2 are trademarks of Transmeta Corporation. All other product or service names mentioned herein are the trademarks of their respective owners.
--------------------------------------------------------------------------------
Contact:
Transmeta Corporation
Mark R. Kent, 408-919-3000
or
Mozes Communications
Kristine Mozes, 781-652-8875
Kristine@mozescomm.com
Intel's 64 bit Xeon/Nocona is already outselling AMD's Opteron.
"About 60,000 Nocona-based servers and 52,000 Opteron servers were sold, IDC said. "
IBM extends lead in server market
By Stephen Shankland
http://news.com.com/IBM+extends+lead+in+server+market/2100-1010_3-5465301.html
Story last modified Tue Nov 23 21:00:00 PST 2004
IBM increased its lead in a growing server market, nibbling away share from rivals Hewlett-Packard and Sun Microsystems, according to figures to be released Wednesday.
Sales of servers worldwide increased 5.5 percent to $11.5 billion in the third quarter of 2004 compared with the year-earlier quarter, according to market researcher IDC. IBM's sales grew faster, though, with revenue increasing 6.3 percent to $3.4 billion, giving Big Blue 31.7 percent of the market.
No. 2 HP grew slower than the market, increasing sales from $3 billion to $3.1 billion. Dell and Sun were essentially tied for third place, with Sun at $1.18 billion and Dell at $1.17 billion--but Dell's sales increased 14.1 percent compared with Sun's 0.1 percent, IDC said.
The overall market growth has been under way for six consecutive quarters, but only recently has it become strong, IDC analyst Vernon Turner said. Information technology spending now includes more new projects rather than just maintaining the status quo, he said.
"For a while the growth was close to inflation. Now we're seeing growth beyond inflation, which shows it is back to strategic thinking instead of just maintenance thinking," Turner said.
IBM and Dell rising while HP and Sun lose share of revenue has been a common theme in server market share statistics in recent quarters. So has the fifth-place position of Fujitsu and its European arm, Fujitsu-Siemens, whose sales grew 4.4 percent to $714 million.
In other findings:
• Linux server sales grew 42.6 percent to $1.06 billion--the ninth consecutive quarter of growth greater than 10 percent. HP leads the market with $287 million in sales, with IBM in second at $219 million and Dell third at $185 million, Turner said.
• Mainstream "x86" processors with 64-bit extensions to accommodate more memory, such as Intel's "Nocona" model of Xeon and Advanced Micro Devices' Opteron, are catching on. About 60,000 Nocona-based servers and 52,000 Opteron servers were sold, IDC said. About 60 percent of the Opteron systems run Linux, which supports the 64-bit extensions today, whereas most Nocona systems run 32-bit Windows, which won't get 64-bit x86 support until 2005.
• Overall, x86 server sales grew 18.3 percent to $5.4 billion, the third quarter in a row that they've been greater than sales of servers using reduced instruction set computing, or RISC, chips such as IBM's Power, Sun's UltraSparc and HP's PA-RISC.
• Unix servers, which most often use RISC chips, declined in sales 2.3 percent to $4 billion, though unit shipments increased 8.3 percent.
This quarter--a seasonally weak one for longtime leader Sun--Hewlett-Packard was No. 1 in the Unix market at $1.27 billion, Sun No. 2 at $1.14 billion and IBM No. 3 at $1.05 billion. IBM revenue increased 2 percent, but Sun dropped 2 percent, and HP lost 6 percent.
• Revenue from servers running Windows grew 13.3 percent to $3.9 billion, essentially tying Unix server revenue.
• More than 75,000 blade servers were sold, a $287 million market for the quarter, with IBM keeping its top spot with 44.2 percent of revenue. Turner predicted blade servers will be a $1 billion business for the full year of 2004.
Blade servers are getting more powerful. "End-user spending for blade servers in the third quarter exceeded our expectations as we saw a shift in mix away from single-processor blades to dual- and quad-processor blade systems," IDC analyst Jessica Yang said in a statement.
Copyright ©1995-2004 CNET Networks, Inc. All rights reserved.
Intel's 64 bit Xeon/Nocona is already outselling AMD's Opteron.
"About 60,000 Nocona-based servers and 52,000 Opteron servers were sold, IDC said. "
IBM extends lead in server market
By Stephen Shankland
http://news.com.com/IBM+extends+lead+in+server+market/2100-1010_3-5465301.html
Story last modified Tue Nov 23 21:00:00 PST 2004
IBM increased its lead in a growing server market, nibbling away share from rivals Hewlett-Packard and Sun Microsystems, according to figures to be released Wednesday.
Sales of servers worldwide increased 5.5 percent to $11.5 billion in the third quarter of 2004 compared with the year-earlier quarter, according to market researcher IDC. IBM's sales grew faster, though, with revenue increasing 6.3 percent to $3.4 billion, giving Big Blue 31.7 percent of the market.
No. 2 HP grew slower than the market, increasing sales from $3 billion to $3.1 billion. Dell and Sun were essentially tied for third place, with Sun at $1.18 billion and Dell at $1.17 billion--but Dell's sales increased 14.1 percent compared with Sun's 0.1 percent, IDC said.
The overall market growth has been under way for six consecutive quarters, but only recently has it become strong, IDC analyst Vernon Turner said. Information technology spending now includes more new projects rather than just maintaining the status quo, he said.
"For a while the growth was close to inflation. Now we're seeing growth beyond inflation, which shows it is back to strategic thinking instead of just maintenance thinking," Turner said.
IBM and Dell rising while HP and Sun lose share of revenue has been a common theme in server market share statistics in recent quarters. So has the fifth-place position of Fujitsu and its European arm, Fujitsu-Siemens, whose sales grew 4.4 percent to $714 million.
In other findings:
• Linux server sales grew 42.6 percent to $1.06 billion--the ninth consecutive quarter of growth greater than 10 percent. HP leads the market with $287 million in sales, with IBM in second at $219 million and Dell third at $185 million, Turner said.
• Mainstream "x86" processors with 64-bit extensions to accommodate more memory, such as Intel's "Nocona" model of Xeon and Advanced Micro Devices' Opteron, are catching on. About 60,000 Nocona-based servers and 52,000 Opteron servers were sold, IDC said. About 60 percent of the Opteron systems run Linux, which supports the 64-bit extensions today, whereas most Nocona systems run 32-bit Windows, which won't get 64-bit x86 support until 2005.
• Overall, x86 server sales grew 18.3 percent to $5.4 billion, the third quarter in a row that they've been greater than sales of servers using reduced instruction set computing, or RISC, chips such as IBM's Power, Sun's UltraSparc and HP's PA-RISC.
• Unix servers, which most often use RISC chips, declined in sales 2.3 percent to $4 billion, though unit shipments increased 8.3 percent.
This quarter--a seasonally weak one for longtime leader Sun--Hewlett-Packard was No. 1 in the Unix market at $1.27 billion, Sun No. 2 at $1.14 billion and IBM No. 3 at $1.05 billion. IBM revenue increased 2 percent, but Sun dropped 2 percent, and HP lost 6 percent.
• Revenue from servers running Windows grew 13.3 percent to $3.9 billion, essentially tying Unix server revenue.
• More than 75,000 blade servers were sold, a $287 million market for the quarter, with IBM keeping its top spot with 44.2 percent of revenue. Turner predicted blade servers will be a $1 billion business for the full year of 2004.
Blade servers are getting more powerful. "End-user spending for blade servers in the third quarter exceeded our expectations as we saw a shift in mix away from single-processor blades to dual- and quad-processor blade systems," IDC analyst Jessica Yang said in a statement.
Copyright ©1995-2004 CNET Networks, Inc. All rights reserved.
Here's a very high profile win for Itanium - in the financial trading sector.
I wonder why the mainstream media doesn't report this?
"The Chicago Mercantile Exchange, for example, is replacing the infrastructure that runs its iLink gateway, which traders use to access the exchange's Globex electronic trading platform. The infrastructure, built on Sun Microsystems servers running Sun's Solaris version of Unix, is being migrated to Intel servers running the Itanium processors and Red Hat Linux. The exchange is experiencing five times the performance of Sun/Unix at one-fifth the price, says Charlie Troxel, the exchange's chief technology officer. It originally planned to replace 20% of the Sun servers by the end of the year, according to Troxel. Now it's on track to replace 50%. "
S1 Embraces Linux
Financial-services companies find Linux's lower operating costs attractive, but they need scalability and security assurances.
By Steven Marlin, InformationWeek
Nov. 16, 2004
URL: http://www.informationweek.com/story/showArticle.jhtml?articleID=53200113
S1 Corp., a maker of online banking and other front-office applications for financial institutions, plans to offer Linux versions of its software by 2006. The company, whose products power customer-facing functions at 4,000 banks and insurance companies, wants to stay ahead of the curve as banks are intrigued by the open-source operating system's possibilities for dramatic operating cost reductions.
A lot of S1's overseas customers are already moving forward with Linux-based initiatives, and domestic customers are beginning to follow suit, says Rob Young, VP of product management and strategy at S1. Linux is being viewed "as a viable platform for heavy-duty production environments," he says.
Migrating S1's software, which is based on Java 2 Enterprise Edition, to Linux doesn't present a formidable technical challenge, Young says. But the vendor wants to be sure that S1 applications on Linux meet customers' security and scalability requirements. "The actual physical port isn't a major undertaking, but we want to be sure we've got everything rock-solid," Young says. S1's teller applications will be the first to be migrated to Linux, according to the company.
S1 isn't a Linux neophyte. The vendor already utilizes several versions of Linux in its own data center, which hosts S1 software for many banks. And Mosaic Software, an automated-teller-machine application company S1 recently acquired, already offers Linux versions of its products. It plans to implement a Linux-based system at a Midwestern bank early next year.
Linux is already making inroads in the financial-services industry, particularly in Wall Street trading applications. The Chicago Mercantile Exchange, for example, is replacing the infrastructure that runs its iLink gateway, which traders use to access the exchange's Globex electronic trading platform. The infrastructure, built on Sun Microsystems servers running Sun's Solaris version of Unix, is being migrated to Intel servers running the Itanium processors and Red Hat Linux. The exchange is experiencing five times the performance of Sun/Unix at one-fifth the price, says Charlie Troxel, the exchange's chief technology officer. It originally planned to replace 20% of the Sun servers by the end of the year, according to Troxel. Now it's on track to replace 50%.
The exchange will continue to run its historical database on Sun servers, however. That database is used to recreate trades in the event of inquiries from customers or regulators. That system, which uses database technology from Oracle, has reliability and scalability requirements that aren't easily addressed by Linux, Troxel says. He adds that the number of database entries has spiked from 2 million per week two years ago to 500 million last week.
This won't make Scott McNealy very happy.
"The Chicago Mercantile Exchange, for example, is replacing the infrastructure that runs its iLink gateway, which traders use to access the exchange's Globex electronic trading platform. The infrastructure, built on Sun Microsystems servers running Sun's Solaris version of Unix, is being migrated to Intel servers running the Itanium processors and Red Hat Linux. The exchange is experiencing five times the performance of Sun/Unix at one-fifth the price, says Charlie Troxel, the exchange's chief technology officer. It originally planned to replace 20% of the Sun servers by the end of the year, according to Troxel. Now it's on track to replace 50%. "
S1 Embraces Linux
Financial-services companies find Linux's lower operating costs attractive, but they need scalability and security assurances.
By Steven Marlin, InformationWeek
Nov. 16, 2004
URL: http://www.informationweek.com/story/showArticle.jhtml?articleID=53200113
S1 Corp., a maker of online banking and other front-office applications for financial institutions, plans to offer Linux versions of its software by 2006. The company, whose products power customer-facing functions at 4,000 banks and insurance companies, wants to stay ahead of the curve as banks are intrigued by the open-source operating system's possibilities for dramatic operating cost reductions.
A lot of S1's overseas customers are already moving forward with Linux-based initiatives, and domestic customers are beginning to follow suit, says Rob Young, VP of product management and strategy at S1. Linux is being viewed "as a viable platform for heavy-duty production environments," he says.
Migrating S1's software, which is based on Java 2 Enterprise Edition, to Linux doesn't present a formidable technical challenge, Young says. But the vendor wants to be sure that S1 applications on Linux meet customers' security and scalability requirements. "The actual physical port isn't a major undertaking, but we want to be sure we've got everything rock-solid," Young says. S1's teller applications will be the first to be migrated to Linux, according to the company.
S1 isn't a Linux neophyte. The vendor already utilizes several versions of Linux in its own data center, which hosts S1 software for many banks. And Mosaic Software, an automated-teller-machine application company S1 recently acquired, already offers Linux versions of its products. It plans to implement a Linux-based system at a Midwestern bank early next year.
Linux is already making inroads in the financial-services industry, particularly in Wall Street trading applications. The Chicago Mercantile Exchange, for example, is replacing the infrastructure that runs its iLink gateway, which traders use to access the exchange's Globex electronic trading platform. The infrastructure, built on Sun Microsystems servers running Sun's Solaris version of Unix, is being migrated to Intel servers running the Itanium processors and Red Hat Linux. The exchange is experiencing five times the performance of Sun/Unix at one-fifth the price, says Charlie Troxel, the exchange's chief technology officer. It originally planned to replace 20% of the Sun servers by the end of the year, according to Troxel. Now it's on track to replace 50%.
The exchange will continue to run its historical database on Sun servers, however. That database is used to recreate trades in the event of inquiries from customers or regulators. That system, which uses database technology from Oracle, has reliability and scalability requirements that aren't easily addressed by Linux, Troxel says. He adds that the number of database entries has spiked from 2 million per week two years ago to 500 million last week.
Egyptian American Bank opts for Intel Itanium 2 processor-based HP Integrity servers for its core banking system
Egyptian American Bank (EAB), one of Egypt's leading joint venture banks, has standardised its enterprise-wide core banking system on Intel Itanium 2 processor-based HP Integrity rx8620 and rx7620 servers to enhance customer service and to underpin the Bank's future growth.
The new solution will run all internal banking transactions. It will also provide a powerful, scaleable platform that will enable EAB to grow its customer base and revenues, by supporting the rapid development and roll-out of new financial products and services.
The centrally-managed enterprise solution hosting EAB's FLEXCUBE* banking application will provide more than 900 employees across 32 branches with enhanced access to customer financial records, product and service information to enable improvement in overall customer service.
As part of its strategy to deploy a new, centralised IT environment, EAB required a server platform that would deliver enhanced processing performance, support for its core banking application, and scope for future growth. Extensive evaluation and benchmarking demonstrated that the open-standards based Intel Itanium 2-based HP Integrity servers delivered increased availability of data and reduced management costs. Additionally, the platform demonstrated the flexibility to enable EAB to efficiently add extra processing resource to the existing system as business demand increases.
'As the central hub of our enterprise-wide IT systems, it is essential that our server environment delivers best in class reliability and performance now, while giving us the flexibility and scalability to grow and meet future customer demand,' said Jamie Hall, Executive Vice President and Chief Operating Officer, Egyptian American Bank. 'The Intel and HP solution provides us with the data-intensive processing performance to support day-to-day operations, while enabling us to fast-track the development of future products and services to maintain our competitive advantage.'
'Intel works closely with a wide range of server manufacturers and software developers to ensure delivery of efficient, tailored solutions, whatever the industry sector,' said Tom Garrison, Director of Enterprise Marketing, Intel EMEA. 'We are delighted to have worked jointly with HP to develop the robust and reliable Intel Itanium 2-based HP Integrity server platform that enables EAB to enhance customer service now while setting the foundations to extend their leadership in the future.'
'Through our extensive experience and expertise in delivering information technology solutions that meet the specific needs of financial services institutions, HP understands that customers in this industry require the highest levels of performance and flexibility to run mission critical applications and, ultimately, provide superior customer service,' said Rudi Schmickl, Vice President and General Manager, Enterprise Storage and Servers, HP EMEA. 'HP Integrity servers deliver the performance, high availability and scalability that customers such as EAB need to minimise downtime and adapt quickly and easily to meet their business needs.'
"introduction of its innovative MirrorBit™ technology.."
This technology was "introduced" over two years ago.
How long is AMD going to keep hiding the fact that this product is over two years behind schedule?
Will AMD always re-introdcuce mirrobit to hide the fact that the volumes are so low after two years of prodcution and 6 years of development ?
Monday May 13, 12:02 am Eastern Time
Press Release
SOURCE: AMD
AMD Introduces First Flash Memory Device Based On MirrorBit Architecture
New Technology Enables Highly Reliable, High Density NOR Flash
SUNNYVALE, Calif.--(BUSINESS WIRE)-- May 13, 2002 --AMD (NYSE:AMD - News) today announced it has been sampling the company's first MirrorBit(TM) Flash memory device. General availability of the 64 Megabit product is scheduled for the end of this quarter. MirrorBit technology is a breakthrough that allows a Flash memory device to hold twice as much data as standard Flash without compromising device endurance, performance or reliability. The company is on schedule to deliver 128 Megabit and 256 Megabit devices in the second half of the year.
The MirrorBit architecture uses a true multi-bit Flash memory cell that delivers the same high levels of performance and reliability as standard NOR Flash. NOR Flash is commonly used to store and execute operating system software for many of the world's electronic devices. MirrorBit technology allows for the lowest cost production of ultra-high density NOR Flash devices. MirrorBit Flash memory devices enable customers to develop advanced products such as cell phones, personal digital assistants, car PCs, telecommunications equipment, and TV set-top boxes that offer higher levels of features and functionality.
"The commercialization of MirrorBit technology is a defining moment for AMD's memory group," said Bertrand Cambou, group vice president of the company's Memory Group. "By delivering a cost structure that is two generations ahead of comparable standard Flash technologies, MirrorBit Flash serves the needs of both the code and data storage markets."
"MirrorBit technology represents a paradigm change in the world of Flash memory," said Kevin Plouse, vice president of technical marketing and business development of AMD's Memory Group. "With its unique combination of high-density, high-performance and low cost, MirrorBit Flash will offer our customers a strategic competitive advantage and spur the creation of a new set of highly innovative products."
"With its innovative architecture, rigorous testing and advantageous cost structure, AMD's MirrorBit technology will make an impressive entry and be a forceful competitor in the Flash memory market," said Brian Matas, vice president, market research for IC Insights. "There's good reason to believe MirrorBit technology will allow AMD to capture additional Flash market share."
"AMD's MirrorBit technology offers industry-leading cost structure without compromises," said Gerry Duggan, Executive Vice President, Future Electronics, a major distributor. "We share AMD's vision of delivering only Flash memory that offers the highest levels of quality and reliability. We believe that this technology is the future of Flash memory and we will dedicate very significant resources to ensure that customers everywhere have access to MirrorBit products."
In AMD's MirrorBit cell, code or data is stored in two discrete and independent locations. By physically separating each bit and maintaining its individual integrity, AMD's MirrorBit devices are inherently more stable and reliable than competing multi-level cell (MLC) devices. MLC devices store fractional levels of charge in one location making them inherently less reliable, and slower to program, read or erase.
Over 350 billion bits of MirrorBit technology were subjected to AMD's rigorous and extensive qualification procedures. As a result, AMD is able to continue to deliver industry-leading levels of data retention and endurance. MirrorBit Flash memory devices are designed to deliver endurance of at least 100,000 program/erase cycles and 20 years of data retention at 125 degrees C.
MirrorBit products also offer customers an easy to use solution by maintaining complete pin-compatibility with existing AMD Flash families. As a result, current customers can easily replace their current AMD Flash device with a MirrorBit device without having to change their system design. The packages and pinouts provide customers with an easy migration path all the way to 1Gigabit Flash.
Features and Packaging
The Am29LV640M and Am29LV641M MirrorBit products are designed to offer a number of advanced features and specifications:
High-density, 64 Megabit Flash memory
Single 3.0 Volt power supply
Asynchronous access times as fast as 90 nanoseconds
Page mode access times as fast as 25 nanoseconds
16 word performance enhancing write buffer offers programming times as fast as 6 microseconds
Small 64kB sectors with erase times as fast as 100 microseconds, ten times faster than competing products
Minimum of 20 years' data retention at 125 degrees C
Minimum of 100,000 erase cycles per sector
Available in:
64-Ball Fortified Ball Grid Array Package
63-Ball Fine-Pitch Ball Grid Array Package
48-Pin Thin Small Outline Package
56-Pin Thin Small Outline Package
Pricing
The Am29LV640M MirrorBit product is priced at $7.95 in 10,000 piece quantities.
AMD Flash Memory Awards
AMD's Flash memory products have received numerous customer service, quality and technology awards. Cisco Systems, Nortel Networks, Samsung, the Bosch Group, and Volkswagen have all awarded AMD their top supplier awards.
About AMD Flash Memory Devices
AMD's technology is employed by the world's largest producer of Flash memory devices, Fujitsu AMD Semiconductor Ltd. (FASL). AMD Flash memory products encompass a broad spectrum of densities and features to support a wide range of markets. AMD Flash memory customers represent leaders in the automotive, networking, telecommunications, and handheld mobile terminal markets. AMD offers many Flash memory products, such as the award-winning Simultaneous Read-Write (SRW) product family; Super Low Voltage 1.8V Flash memory devices; and Burst- and Page-Mode devices. AMD developed the robust Known Good Die (KGD) program and the patented negative gate erase technology, and developed highly reliable Fine-pitch Ball Grid Array packaging.
About AMD
AMD is a global supplier of integrated circuits for the personal and networked computer and communications markets with manufacturing facilities in the United States, Europe, Japan, and Asia. AMD, a Fortune 500 and Standard & Poor's 500 company, produces microprocessors, Flash memory devices, and support circuitry for communications and networking applications. Founded in 1969 and based in Sunnyvale, California, AMD had revenues of $3.9 billion in 2001. (NYSE: AMD - News).
AMD on the Web
For more information about the Am29LV640M product, please visit our virtual pressroom at http://www.amd.com/news/spotlight. Additional press releases and information about AMD and its products are available at www.amd.com/news/news.html.
AMD, the AMD Arrow logo, and combinations thereof, and MirrorBit are trademarks of Advanced Micro Devices, Inc. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.
Note: A Photo is available at URL: http://www.businesswire.com/cgi-bin/photo.cgi?pw.051302/bb1
--------------------------------------------------------------------------------
Contact:
AMD
Ron Dusek, 512/602-9514 (PR)
ron.dusek@amd.com
or
Tim Martin, 512/602-6047 (PR)
timothy.martin@amd.com
or
Toni Beckham, 408/749-3127 (IR)
SUN Microsystems seems to have plans for Nocona and Intel 64 Bit x86 technology:
"Schwartz noted that Solaris is now "up and running" on Intel's new Nocona processor for servers, which can run both 32-bit and 64-bit applications. "
Sun Microsystems will begin compensating its sales staff for Solaris deals that involve hardware from other manufacturers, as part of the company's efforts to fend off competition from low-cost servers running Linux.
Chief Operating Officer Jonathan Schwartz described the strategy in a posting to his blog Monday.
"So if a sales rep sells Solaris on Dell or IBM, or even HP (Xeon or Nocona), we pay them as if they sold the hardware," Schwartz wrote. "I'm not sure we could make the point more clearly that we're committed to making Solaris the volume leader on all systems."
The change comes at a time when Sun is trying to make its own servers more competitive with Linux, which continues to become more prevalent in corporate data centers. According to analysts, many of Sun's current financial woes stem from the fact the many corporate customers are opting to buy commodity servers that run Linux and use x86 processors from Intel or Advanced Micro Devices, instead of purchasing Sun's own Solaris on Sparc servers.
In response to the Linux competition, Sun has renewed its commitment to selling Solaris servers for the x86 platform. Last year, the company partnered with AMD to develop servers based on AMD's Operton 64-bit processor. Sun is also developing software called Janus for running Linux applications on Solaris.
Schwartz noted that Solaris is now "up and running" on Intel's new Nocona processor for servers, which can run both 32-bit and 64-bit applications.
Schwartz, formerly the head of Sun's software business, has sought to shake up the company's internal culture in an effort to bring it back to stable financial footing. The company, which currently makes the bulk of its revenue from its Sparc-based servers, has renewed its commitment to Solaris and other software product lines, such as its Java application server and open-source desktop system.
SUN Microsystems seems to have plans for Nocona and Intel 64 Bit x86 technology:
"Schwartz noted that Solaris is now "up and running" on Intel's new Nocona processor for servers, which can run both 32-bit and 64-bit applications. "
Sun Microsystems will begin compensating its sales staff for Solaris deals that involve hardware from other manufacturers, as part of the company's efforts to fend off competition from low-cost servers running Linux.
Chief Operating Officer Jonathan Schwartz described the strategy in a posting to his blog Monday.
"So if a sales rep sells Solaris on Dell or IBM, or even HP (Xeon or Nocona), we pay them as if they sold the hardware," Schwartz wrote. "I'm not sure we could make the point more clearly that we're committed to making Solaris the volume leader on all systems."
The change comes at a time when Sun is trying to make its own servers more competitive with Linux, which continues to become more prevalent in corporate data centers. According to analysts, many of Sun's current financial woes stem from the fact the many corporate customers are opting to buy commodity servers that run Linux and use x86 processors from Intel or Advanced Micro Devices, instead of purchasing Sun's own Solaris on Sparc servers.
In response to the Linux competition, Sun has renewed its commitment to selling Solaris servers for the x86 platform. Last year, the company partnered with AMD to develop servers based on AMD's Operton 64-bit processor. Sun is also developing software called Janus for running Linux applications on Solaris.
Schwartz noted that Solaris is now "up and running" on Intel's new Nocona processor for servers, which can run both 32-bit and 64-bit applications.
Schwartz, formerly the head of Sun's software business, has sought to shake up the company's internal culture in an effort to bring it back to stable financial footing. The company, which currently makes the bulk of its revenue from its Sparc-based servers, has renewed its commitment to Solaris and other software product lines, such as its Java application server and open-source desktop system.
AMD Demonstrates World's First x86 Dual-Core Processor
Tuesday August 31, 12:01 am ET
AMD's Newest Leadership Milestone Changes the Dynamics of the Industry
SUNNYVALE, Calif.--(BUSINESS WIRE)--Aug. 31, 2004--Today, AMD (NYSE:AMD - News) announces it is demonstrating the industry's first x86 dual-core processor. During demonstrations held at the company's Austin facilities, AMD is showing an HP ProLiant DL585 server powered by four dual-core AMD Opteron(TM) processors manufactured on 90nm silicon-on-insulator process technology.
ADVERTISEMENT
With a simple upgrade path to more efficient computing, based on AMD's existing system infrastructure and industry-standard architecture, enterprise customers can expect more efficient processing power without the penalties of increased power consumption and heat dissipation. The dual-core AMD Opteron processor for servers and workstations is expected to offer the best performance per watt in the market when AMD plans to make it available in mid-2005.
"This industry milestone changes the dynamics of the computing business," said Dirk Meyer, executive vice president, AMD Computation Products Group. "Once again, AMD is delivering a simplified approach to more efficient processing power, with products that will deliver multi-core 64-bit computing to our strong customer base."
This announcement follows a series of impressive firsts from AMD. As the first company to ship products that meet customer demands for high-performance, simultaneous x86-based 32- and 64-bit computing, AMD set in motion an industry-wide transition to pervasive 64-bit computing. Next, AMD was the first company to implement 64-bit computing and Enhanced Virus Protection (enabled by the Windows® Service Pack 2) in desktop and low-power mobile PC processors. AMD64 continues to lead the industry transition to pervasive 64-bit computing and AMD's processor roadmap continues to leverage the efficiency and benefits made possible by AMD64 technology.
"Dual-core technology provides an attractive path for increasing processor performance with little or no increase in power consumption or heat dissipation," observed Nathan Brookwood, principal analyst at Insight64. "AMD laid the groundwork for its dual-core processors years ago, when it gave its single-core AMD64 processors the on-chip plumbing they would need to support a second core at a later date. As AMD moves dual-core technology from theory to practice, it is reassuring to see that current investments in core logic and platform technology will remain relevant for years to come."
"AMD has recognized the importance of keeping a compatible system architecture while still meeting today's demanding computing needs," said Kevin Krewell, editor-in-chief, Microprocessor Report, Instat-MDR. "AMD's demonstration of a current platform operating on dual-core AMD Opteron processors based on AMD64 technology represents another industry first for AMD as the company continues to provide industry-leading innovations to the industry-standard architecture."
Industry Support
By collaborating with strong partners, AMD continues to lead technology transitions in the x86 industry, demonstrating technology that will enable the transition to dual-core products.
"Dual- and multi-core processor technologies on industry-standard servers will redefine scalability, performance and value for enterprise and SMB customers," said Paul Miller, vice president of marketing, HP Industry Standard Servers. "Having the industry's first dual-core x86 processors from AMD up and running on HP ProLiant servers demonstrates both HP's close relationship and collaboration with AMD and our ongoing commitment to rapidly deliver the best new technologies to our customers."
"As AMD's strategic partner, we are in a unique position to support AMD64 dual-core technology with our hardware and software products," said John Fowler, executive vice president of the Network Systems group at Sun Microsystems, Inc. "The Solaris OS combined with the AMD Opteron processor-based Sun servers and workstations will take full advantage of the AMD64 dual-core architecture, managing multi-threaded applications with superior efficiency and performance."
Innovation from an industry leader
Based on the existing 940-socket infrastructure, AMD expects the upcoming dual-core AMD Opteron processor to provide better performance on a majority of server/workstation workloads by combining two processing cores on a single die. The form factor, energy consumption and performance needs of today's computer designs demand new innovations. Dual-core processor technology will equip customers with more balanced performance based on industry-standard system architecture.
Dual-core processors are a natural extension of AMD64 technology with Direct Connect Architecture. In addition to being the first to help eliminate the bottlenecks inherent in x86 front-side bus architectures, AMD is now demonstrating the capabilities of being the first to directly connect two cores on the same die along with the memory controller, I/O and other processors - which will improve the overall system performance and efficiency.
Availability
AMD plans to introduce a full dual-core processor line-up for the one- to eight-socket server and workstation market in mid-2005 based on the existing 940-pin socket. Dual-core processors for the client market are expected to follow in the second half of 2005.
For more information on today's announcement, visit www.amd.com/dualcore2
About AMD64
AMD64 evolves the industry-standard 32-bit x86 architecture to support the demanding 64-bit environment that will enable future generations of computer functionality and productivity. AMD designed the AMD64 platform to allow end users to enjoy reliable, best-in-class performance on the 32-bit software they own today while preparing for a seamless transition to high-performance 64-bit applications. The AMD64 architecture is also designed to enhance the security of your computing environment by integrating Enhanced Virus Protection technology enabled by
advanced anti-virus features in Windows® XP Service Pack 2 and the upcoming Windows Server 2003 Service Pack 1.
Since the introduction of the AMD64 architecture in 2003, the AMD Opteron(TM) processor for servers and workstations and the AMD Athlon(TM) 64 processors for desktop and notebook computers have earned more than 55 awards for innovation and performance and the support of more than 2,000 OEMs, hardware and software developers, system builders, and distributors.
About AMD
AMD (NYSE:AMD - News) designs and produces innovative microprocessors, Flash memory devices and low-power processor solutions for the computer, communications and consumer electronics industries. AMD is dedicated to helping its customers deliver standards-based, customer-focused solutions for technology users, ranging from enterprises and governments to individual consumers. For more information, visit www.amd.com.
Cautionary Statement
This release contains forward-looking statements, which are made pursuant to the safe harbor provisions of the U.S. Private Securities Litigation Reform Act of 1995. Forward-looking statements are generally preceded by words such as "plans," "expects," "believes," "anticipates" or "intends." Investors are cautioned that all forward-looking statements in this release involve risks and uncertainty that could cause actual results to differ materially from current expectations. Risks include the possibility that the company will not achieve its current product and technology introduction schedules; that adoption of AMD64 products, including upcoming dual-core processors, by tier-one OEMs and other customers will not occur as expected; that solutions providers will not timely provide the infrastructure, including operating systems and applications, to support the company's AMD64 technology; and that Intel Corporation's pricing, marketing programs, product bundling, new product introductions or other activities targeting the company's processor business will adversely affect the company's sales plans. We urge investors to review in detail the risks and uncertainties in the company's Securities and Exchange Commission filings, including but not limited to the Annual Report on Form 10-K for the year ended December 28, 2003, and the Quarterly Report on Form 10-Q for the quarter ended March 28, 2004.
AMD, the AMD Arrow logo, AMD Athlon, AMD Opteron and combinations thereof, are trademarks of Advanced Micro Devices, Inc. Windows is a registered trademark of Microsoft Corporation in the U.S. and/or other countries. Other names are for informational purposes only and may be trademarks of their respective owners.
--------------------------------------------------------------------------------
Contact:
Advanced Micro Devices, Inc, Sunnyvale
Jane Kovacs, 512-602-4929
jane.kovacs@amd.com
www.amd.com
AMD Demonstrates World's First x86 Dual-Core Processor
Tuesday August 31, 12:01 am ET
AMD's Newest Leadership Milestone Changes the Dynamics of the Industry
SUNNYVALE, Calif.--(BUSINESS WIRE)--Aug. 31, 2004--Today, AMD (NYSE:AMD - News) announces it is demonstrating the industry's first x86 dual-core processor. During demonstrations held at the company's Austin facilities, AMD is showing an HP ProLiant DL585 server powered by four dual-core AMD Opteron(TM) processors manufactured on 90nm silicon-on-insulator process technology.
ADVERTISEMENT
With a simple upgrade path to more efficient computing, based on AMD's existing system infrastructure and industry-standard architecture, enterprise customers can expect more efficient processing power without the penalties of increased power consumption and heat dissipation. The dual-core AMD Opteron processor for servers and workstations is expected to offer the best performance per watt in the market when AMD plans to make it available in mid-2005.
"This industry milestone changes the dynamics of the computing business," said Dirk Meyer, executive vice president, AMD Computation Products Group. "Once again, AMD is delivering a simplified approach to more efficient processing power, with products that will deliver multi-core 64-bit computing to our strong customer base."
This announcement follows a series of impressive firsts from AMD. As the first company to ship products that meet customer demands for high-performance, simultaneous x86-based 32- and 64-bit computing, AMD set in motion an industry-wide transition to pervasive 64-bit computing. Next, AMD was the first company to implement 64-bit computing and Enhanced Virus Protection (enabled by the Windows® Service Pack 2) in desktop and low-power mobile PC processors. AMD64 continues to lead the industry transition to pervasive 64-bit computing and AMD's processor roadmap continues to leverage the efficiency and benefits made possible by AMD64 technology.
"Dual-core technology provides an attractive path for increasing processor performance with little or no increase in power consumption or heat dissipation," observed Nathan Brookwood, principal analyst at Insight64. "AMD laid the groundwork for its dual-core processors years ago, when it gave its single-core AMD64 processors the on-chip plumbing they would need to support a second core at a later date. As AMD moves dual-core technology from theory to practice, it is reassuring to see that current investments in core logic and platform technology will remain relevant for years to come."
"AMD has recognized the importance of keeping a compatible system architecture while still meeting today's demanding computing needs," said Kevin Krewell, editor-in-chief, Microprocessor Report, Instat-MDR. "AMD's demonstration of a current platform operating on dual-core AMD Opteron processors based on AMD64 technology represents another industry first for AMD as the company continues to provide industry-leading innovations to the industry-standard architecture."
Industry Support
By collaborating with strong partners, AMD continues to lead technology transitions in the x86 industry, demonstrating technology that will enable the transition to dual-core products.
"Dual- and multi-core processor technologies on industry-standard servers will redefine scalability, performance and value for enterprise and SMB customers," said Paul Miller, vice president of marketing, HP Industry Standard Servers. "Having the industry's first dual-core x86 processors from AMD up and running on HP ProLiant servers demonstrates both HP's close relationship and collaboration with AMD and our ongoing commitment to rapidly deliver the best new technologies to our customers."
"As AMD's strategic partner, we are in a unique position to support AMD64 dual-core technology with our hardware and software products," said John Fowler, executive vice president of the Network Systems group at Sun Microsystems, Inc. "The Solaris OS combined with the AMD Opteron processor-based Sun servers and workstations will take full advantage of the AMD64 dual-core architecture, managing multi-threaded applications with superior efficiency and performance."
Innovation from an industry leader
Based on the existing 940-socket infrastructure, AMD expects the upcoming dual-core AMD Opteron processor to provide better performance on a majority of server/workstation workloads by combining two processing cores on a single die. The form factor, energy consumption and performance needs of today's computer designs demand new innovations. Dual-core processor technology will equip customers with more balanced performance based on industry-standard system architecture.
Dual-core processors are a natural extension of AMD64 technology with Direct Connect Architecture. In addition to being the first to help eliminate the bottlenecks inherent in x86 front-side bus architectures, AMD is now demonstrating the capabilities of being the first to directly connect two cores on the same die along with the memory controller, I/O and other processors - which will improve the overall system performance and efficiency.
Availability
AMD plans to introduce a full dual-core processor line-up for the one- to eight-socket server and workstation market in mid-2005 based on the existing 940-pin socket. Dual-core processors for the client market are expected to follow in the second half of 2005.
For more information on today's announcement, visit www.amd.com/dualcore2
About AMD64
AMD64 evolves the industry-standard 32-bit x86 architecture to support the demanding 64-bit environment that will enable future generations of computer functionality and productivity. AMD designed the AMD64 platform to allow end users to enjoy reliable, best-in-class performance on the 32-bit software they own today while preparing for a seamless transition to high-performance 64-bit applications. The AMD64 architecture is also designed to enhance the security of your computing environment by integrating Enhanced Virus Protection technology enabled by
advanced anti-virus features in Windows® XP Service Pack 2 and the upcoming Windows Server 2003 Service Pack 1.
Since the introduction of the AMD64 architecture in 2003, the AMD Opteron(TM) processor for servers and workstations and the AMD Athlon(TM) 64 processors for desktop and notebook computers have earned more than 55 awards for innovation and performance and the support of more than 2,000 OEMs, hardware and software developers, system builders, and distributors.
About AMD
AMD (NYSE:AMD - News) designs and produces innovative microprocessors, Flash memory devices and low-power processor solutions for the computer, communications and consumer electronics industries. AMD is dedicated to helping its customers deliver standards-based, customer-focused solutions for technology users, ranging from enterprises and governments to individual consumers. For more information, visit www.amd.com.
Cautionary Statement
This release contains forward-looking statements, which are made pursuant to the safe harbor provisions of the U.S. Private Securities Litigation Reform Act of 1995. Forward-looking statements are generally preceded by words such as "plans," "expects," "believes," "anticipates" or "intends." Investors are cautioned that all forward-looking statements in this release involve risks and uncertainty that could cause actual results to differ materially from current expectations. Risks include the possibility that the company will not achieve its current product and technology introduction schedules; that adoption of AMD64 products, including upcoming dual-core processors, by tier-one OEMs and other customers will not occur as expected; that solutions providers will not timely provide the infrastructure, including operating systems and applications, to support the company's AMD64 technology; and that Intel Corporation's pricing, marketing programs, product bundling, new product introductions or other activities targeting the company's processor business will adversely affect the company's sales plans. We urge investors to review in detail the risks and uncertainties in the company's Securities and Exchange Commission filings, including but not limited to the Annual Report on Form 10-K for the year ended December 28, 2003, and the Quarterly Report on Form 10-Q for the quarter ended March 28, 2004.
AMD, the AMD Arrow logo, AMD Athlon, AMD Opteron and combinations thereof, are trademarks of Advanced Micro Devices, Inc. Windows is a registered trademark of Microsoft Corporation in the U.S. and/or other countries. Other names are for informational purposes only and may be trademarks of their respective owners.
--------------------------------------------------------------------------------
Contact:
Advanced Micro Devices, Inc, Sunnyvale
Jane Kovacs, 512-602-4929
jane.kovacs@amd.com
www.amd.com
Looks like Intel is about ready to issue a press release on the ststus of their upcoming 65 nM silicon fab process.
Intel to throttle power by enhancing silicon
By Michael Kanellos
Staff Writer, CNET News.com
http://news.com.com/2100-1006-5327521.html
Story last modified August 29, 2004, 9:00 PM PDT
Intel will try to further cut power consumption in its next generation of chips by using improved strained silicon, along with transistors that block power to other circuits, and other added features.
The Santa Clara, Calif.-based PC maker will incorporate a number of changes in its 65-nanometer manufacturing process, which will begin in 2005, said Mark Bohr, an Intel senior fellow and director of processor architecture and integration. Experimental SRAM chips have already been produced on the process.
Depending on the goals of the chip designer, circuits made on the 65-nanometer process can boost performance or cut power consumption or do both, but Intel is putting a definite accent on energy conservation. (65-nanometer refers to the average size of features of a chip. A nanometer is a billionth of a meter.)
The enhanced version of strained silicon--a technology that first appeared in Intel's 90-nanometer chips--can increase performance by 30 percent compared with nonstrained chips or provide a 4x improvement in inadvertent electricity leakage.
"With strained silicon, Intel is at least a generation ahead," Bohr said. "The enhanced strained silicon can be used to enhance drive current or reduce leakage."
Competitors IBM and Advanced Micro Devices have also begun to include strained silicon in chips.
Shrinking transistors and transistor components is at the heart of Moore's Law, which postulates that the number of transistors on a chip doubles every two years. Smaller transistors generally are faster (electrons have to cover less ground) and can result in smaller, cheaper, better and more energy-efficient chips. Ten million transistors in the test SRAM chips could fit on the tip of a ballpoint pen.
Following Moore's Law rigorously over the last 30 years, however, has resulted in extremely dense chips that are difficult to design, expensive to produce, tough to release on schedule and that exude as much heat, proportionally, as a rocket nozzle.
"As we scale dimensions, things get tougher," Bohr said.
Other alternations in the 65-nanometer generation will come in the oxide gate, a thin layer that helps control the flow of electrons between a transistor's source and the drain, a key element in controlling whether a computer sees a piece of data as a "1" or a "0."
In 65-nanometer chips, the length of gate oxide will get shorter, which will improve performance, but the gate oxide will retain its thickness. Typically, the oxide gets thinner in manufacturing transitions. By keeping the thickness the same, capacitance (a function of how much energy gets stored in a given material) goes down by 20 percent, which consequently reduces the potential for leakage.
Chips made on this process will also include sleep transistors that shut off power to other blocks of transistors.
Bohr couldn't quantify the power consumption these sleep transistors will curb, but potentially it could be quite a bit in both active power consumption and leakage.
"It is a noticeable gain in leakage reduction," he said.
The first chips made on the 65-nanometer process will come out in late 2005, Bohr said, first in Oregon and subsequently in Arizona and Ireland. He added that the transition from 90-nanometer to 65-nanometer probably won't be as taxing as the shift from 130-nanometer to 90-nanometer, which involved more changes in the underlying silicon.
Nonetheless, history shows that transitions in manufacturing take longer than expected. Designers often opt for performance over power savings.
"They are giving their circuit designers a lot of tools to control power, but it remains to be seen how they will use these skills," said Nathan Brookwood, an analyst with Insight 64.
Manufacturing transitions are also expected to start occurring at three-year, rather than two-year, intervals as time goes on. Many expect the 45-nanometer transition slated for 2007 to be particularly difficult, as chipmakers will likely have to change the materials in the transistor gate and gate oxide. By 2021 or so, the shrinking of transistor elements under Moore's Law is expected to come to an end.
Bohr added that Intel will adapt 248- and 193-nanometer lithography tools to produce 65-nanometer chips. Lithography tools "draw" transistors on wafers by shining light onto masks, maps of a chip's circuitry that cost millions of dollars to produce. The shadow cast by the masks (reduced greatly in size by lenses) in turn causes photoresistant chemicals on the wafer to react and expose metal.
New generations of lithography tools often take years to come to market and can cost $15 million or more, so adapting old tools for the new chips reduces risk exposure.
Intel will further engage in "dry" lithography for 65-nanometer manufacturing. Some manufacturers are currently exploring the possibility of immersion lithography, in which the wafers are submerged in water. The water helps focus the laser beam, which in turn lets engineers draw smaller circuits.
Extreme Ultraviolet, or EUV, lithography, which uses light with a much smaller wavelength, will start to replace 193-nanometer equipment toward the end of the decade.
One thing Intel's 65-nanometer chips won't contain is Silicon-on-Insulator, or SOI, an additional layer that advocates say cuts leakage power. In the early part of the decade, Intel experimented with adding what it called Ultra Thin SOI into its chips.
Now that notion has been scrapped, and Intel believes the power savings that Ultra Thin SOI would have brought will be accomplished through the tri-gate transistor, a future transistor that effectively triples the area inside a transistor for transporting electrons.
Intel to throttle power by enhancing silicon
By Michael Kanellos
Staff Writer, CNET News.com
http://news.com.com/2100-1006-5327521.html
Story last modified August 29, 2004, 9:00 PM PDT
Intel will try to further cut power consumption in its next generation of chips by using improved strained silicon, along with transistors that block power to other circuits, and other added features.
The Santa Clara, Calif.-based PC maker will incorporate a number of changes in its 65-nanometer manufacturing process, which will begin in 2005, said Mark Bohr, an Intel senior fellow and director of processor architecture and integration. Experimental SRAM chips have already been produced on the process.
Depending on the goals of the chip designer, circuits made on the 65-nanometer process can boost performance or cut power consumption or do both, but Intel is putting a definite accent on energy conservation. (65-nanometer refers to the average size of features of a chip. A nanometer is a billionth of a meter.)
The enhanced version of strained silicon--a technology that first appeared in Intel's 90-nanometer chips--can increase performance by 30 percent compared with nonstrained chips or provide a 4x improvement in inadvertent electricity leakage.
"With strained silicon, Intel is at least a generation ahead," Bohr said. "The enhanced strained silicon can be used to enhance drive current or reduce leakage."
Competitors IBM and Advanced Micro Devices have also begun to include strained silicon in chips.
Shrinking transistors and transistor components is at the heart of Moore's Law, which postulates that the number of transistors on a chip doubles every two years. Smaller transistors generally are faster (electrons have to cover less ground) and can result in smaller, cheaper, better and more energy-efficient chips. Ten million transistors in the test SRAM chips could fit on the tip of a ballpoint pen.
Following Moore's Law rigorously over the last 30 years, however, has resulted in extremely dense chips that are difficult to design, expensive to produce, tough to release on schedule and that exude as much heat, proportionally, as a rocket nozzle.
"As we scale dimensions, things get tougher," Bohr said.
Other alternations in the 65-nanometer generation will come in the oxide gate, a thin layer that helps control the flow of electrons between a transistor's source and the drain, a key element in controlling whether a computer sees a piece of data as a "1" or a "0."
In 65-nanometer chips, the length of gate oxide will get shorter, which will improve performance, but the gate oxide will retain its thickness. Typically, the oxide gets thinner in manufacturing transitions. By keeping the thickness the same, capacitance (a function of how much energy gets stored in a given material) goes down by 20 percent, which consequently reduces the potential for leakage.
Chips made on this process will also include sleep transistors that shut off power to other blocks of transistors.
Bohr couldn't quantify the power consumption these sleep transistors will curb, but potentially it could be quite a bit in both active power consumption and leakage.
"It is a noticeable gain in leakage reduction," he said.
The first chips made on the 65-nanometer process will come out in late 2005, Bohr said, first in Oregon and subsequently in Arizona and Ireland. He added that the transition from 90-nanometer to 65-nanometer probably won't be as taxing as the shift from 130-nanometer to 90-nanometer, which involved more changes in the underlying silicon.
Nonetheless, history shows that transitions in manufacturing take longer than expected. Designers often opt for performance over power savings.
"They are giving their circuit designers a lot of tools to control power, but it remains to be seen how they will use these skills," said Nathan Brookwood, an analyst with Insight 64.
Manufacturing transitions are also expected to start occurring at three-year, rather than two-year, intervals as time goes on. Many expect the 45-nanometer transition slated for 2007 to be particularly difficult, as chipmakers will likely have to change the materials in the transistor gate and gate oxide. By 2021 or so, the shrinking of transistor elements under Moore's Law is expected to come to an end.
Bohr added that Intel will adapt 248- and 193-nanometer lithography tools to produce 65-nanometer chips. Lithography tools "draw" transistors on wafers by shining light onto masks, maps of a chip's circuitry that cost millions of dollars to produce. The shadow cast by the masks (reduced greatly in size by lenses) in turn causes photoresistant chemicals on the wafer to react and expose metal.
New generations of lithography tools often take years to come to market and can cost $15 million or more, so adapting old tools for the new chips reduces risk exposure.
Intel will further engage in "dry" lithography for 65-nanometer manufacturing. Some manufacturers are currently exploring the possibility of immersion lithography, in which the wafers are submerged in water. The water helps focus the laser beam, which in turn lets engineers draw smaller circuits.
Extreme Ultraviolet, or EUV, lithography, which uses light with a much smaller wavelength, will start to replace 193-nanometer equipment toward the end of the decade.
One thing Intel's 65-nanometer chips won't contain is Silicon-on-Insulator, or SOI, an additional layer that advocates say cuts leakage power. In the early part of the decade, Intel experimented with adding what it called Ultra Thin SOI into its chips.
Now that notion has been scrapped, and Intel believes the power savings that Ultra Thin SOI would have brought will be accomplished through the tri-gate transistor, a future transistor that effectively triples the area inside a transistor for transporting electrons.
Itanium-Armed Start-up To Menace IBM's Precious Mainframe Monopoly
August 21, 2004, http://www.linuxworld.com/story/46071.htm
Summary
There's been no one to challenge IBM's unprecedented mainframe monopoly since Big Blue ran off the last of the PCMs years ago when it got light years ahead of them on the silicon front, forcing them to close up shop and slink away. Until Now. 1999 newcomer PSI is bent on making an industry-standard Itanium 2 system capable of running MVS, z/OS, S/390, Unix, Windows, and Linux simultaneously.
By Maureen O'Gara
There's been no one to challenge IBM's unprecedented mainframe monopoly since Big Blue ran off the last of the PCMs years ago when it got light years ahead of them on the silicon front, forcing them to close up shop and slink away.
At the same time IBM saved mainframes from being entombed in the La Brea tar pits along with other fossils, an exercise that has proven to be worth the effort. In its latest quarter, mainframes were credited with contributing 15% of IBM's profits, the tip of a huge multibillion-dollar business that has been IBM's to do with what it will.
Until now.
Enter Platform Solutions Inc (PSI), a company that was started in 1999 by Amdahl's old core engineers using an estimated half-a-billion-dollars worth of Amdahl IP and bent on making an industry-standard Itanium 2 system capable of running MVS, z/OS, S/390, Unix, Windows and Linux simultaneously.
PSI is playing for the half of the $5 billion-a-year mainframe hardware business that IBM barely cares about, all those mid-sized and middling large companies that aren't the Fortune 100.
IBM can't be bothered dirtying its own hands on anything less than a Fortune 100, according to PSI president and CEO Michael Maulick, himself an IBM veteran with over 20 years of service (ADSM is his). IBM passes their business on to stand-ins.
Such is IBM's focus on the rarified high high-end and such is its fascination with 450-MIPS building blocks that it has stranded less prestigious accounts without an affordable migration path past the G4 and G5, Maulick says.
Things being what they are, he says, second-tier companies are being forced to pay $8,000-$10,000 per MIPS where their better-off betters are paying only $2,000.
Knowing that PSI is coming, IBM threw these second-tier accounts a bone in July when it introduced the dumbed-down z990 it dubbed the z890. It pulled a few chips out of the 990, lowered its clock speed, crippled its cache, reused its packaging and, knowing that PSI would start at 26 MIPS, said the 890 would do that.
Maulick claims it's not a solution and, at roughly the same price as the 990, certainly not a reasonable one.
PSI aims to address this disenfranchised market, believing that it's also elastic. It figures, for instance, that there are high-end Sun users that would love to go back to the safety and security of the MVS RAS and can't get there from where they are.
At this point, PSI is only thinking about carving off a piece of the mainframe iron market, not the $24 billion service and software business connected with it.
PSI is playing for big stakes - and ultimately a big IPO - remember that 60%-70% of the world's mission-critical data is still on the mainframe - and it has a few little friends helping it take on the IBM monopoly.
It just got an undisclosed amount of money in second-round financing - undisclosed because it doesn't want IBM to know what its resources are and try to drain them, but it must be a tidy pile because it's supposed to take PSI past profitability in, oh, 2006.
The money came from Goldman Sachs, which led Blueprint Ventures, InterWest Partners and InvestCorp, all of which must be delirious at the thought that PSI will have no competition.
As Maulick points out, it's not exactly a field where VCs are funding a hundred start-ups. The barrier to entry, he figures, is roughly $500 million, the value he puts on the tools, diagnostics and testing widgetry PSI got from Amdahl.
Before deciding to take the job with PSI, Maulick was instrumental in raising PSI's $10.2 million A round last September. Goldman wasn't part of that financing but Fujitsu, which owned Amdahl, and Intel were.
Until last year, PSI was kept alive by $500,000 in angel droppings, Maulick said.
PSI's charter was originally Fujitsu's Plan B for how to salvage Amdahl and the notion of using Itanium as Fujitsu's back door in the Alamo dates back to 1995 when the Itanium was still a gleam in Intel's eye. When Itanium was late, Fujitsu abandoned the field and would have deep six'd its IP if it hadn't been persuaded to spin it off. PSI subsequently looked at IBM's Power chip, but decided the "underrated" Itanium was better.
The guy who first saw PSI's commercial implications, according to Maulick, was George Hoyem, the long-time VC, now Blueprint Ventures' managing partner, who once upon a time was general manager of HP's Internet Commerce Division after he co-founded Visix Software. Hoyem put together PSI's first round.
Naturally, the mere mention of HP brings to mind its IBM-besting ambitions and the idea of its, oh, Itanium-based Superdome boxes being used as mainframe-aping systems against its hereditary rival. Maulick won't confirm the possibility, but PSI is developing its so-called Universal Server on HP hardware as well as Intel white boxes.
PSI says the second-round money will be used to expand its development efforts and to fund product introduction. The outfit was at the IBM Share user meeting in New York this week showing off its stuff to the elect in a suite under NDA.
Universal Server is currently in alpha. PSI intends to beta the system in Q4 and go GA in Q1.
PSI has wrapped core microcode and patented virtualization mojo into the Universal Server's architecture and Goldman is evidently heady with delight at the prospect of running z/OS on an open systems server. It speaks of the widgetry as "reinvigorating the PCM market in a 64-bit world."
Maulick describes it as "reincarnating Amdahl" without Amdahl. Amdahl used to have 24% of the mainframe market. If PSI gets just 10%, Maulick will count it highly successful.
PSI, which will sell multi-processor servers that range into the thousands of MIPS, thinks it's got a world-class alternative to IBM's z/Series and p/Series. It believes it can argue against the wisdom of investing in a single-purpose proprietary platform that, it says, limits the exploitation of open technologies and eats up IT resources trying to solve management issues.
In PSI's architecture, Windows, Linux and Unix will run natively on Itanium, and z/OS and OS/390 will run on top of what the company calls its PCMWare. All the operating systems will be controlled from a single management console.
Ironically, if IBM Software weren't so captive to IBM's hardware interests, IBM Software chief Steve Mills would probably be beating a path to PSI's door.
About the author
Maureen O'Gara is editor-in-chief of Maureen O'Gara's LinuxGram(TM) - published weekly by G2 Computer Intelligence Inc. and distributed by Linux Business Week.
Copyright © 1998-2004 LinuxWorld.com, an IDG Communications company.
Here's a little Itanium Wild Card that might shake up IBM more than just a little bit - and cause them to accelerate their Itanium system offerings.
Itanium-Armed Start-up To Menace IBM's Precious Mainframe Monopoly
August 21, 2004, http://www.linuxworld.com/story/46071.htm
Summary
There's been no one to challenge IBM's unprecedented mainframe monopoly since Big Blue ran off the last of the PCMs years ago when it got light years ahead of them on the silicon front, forcing them to close up shop and slink away. Until Now. 1999 newcomer PSI is bent on making an industry-standard Itanium 2 system capable of running MVS, z/OS, S/390, Unix, Windows, and Linux simultaneously.
By Maureen O'Gara
There's been no one to challenge IBM's unprecedented mainframe monopoly since Big Blue ran off the last of the PCMs years ago when it got light years ahead of them on the silicon front, forcing them to close up shop and slink away.
At the same time IBM saved mainframes from being entombed in the La Brea tar pits along with other fossils, an exercise that has proven to be worth the effort. In its latest quarter, mainframes were credited with contributing 15% of IBM's profits, the tip of a huge multibillion-dollar business that has been IBM's to do with what it will.
Until now.
Enter Platform Solutions Inc (PSI), a company that was started in 1999 by Amdahl's old core engineers using an estimated half-a-billion-dollars worth of Amdahl IP and bent on making an industry-standard Itanium 2 system capable of running MVS, z/OS, S/390, Unix, Windows and Linux simultaneously.
PSI is playing for the half of the $5 billion-a-year mainframe hardware business that IBM barely cares about, all those mid-sized and middling large companies that aren't the Fortune 100.
IBM can't be bothered dirtying its own hands on anything less than a Fortune 100, according to PSI president and CEO Michael Maulick, himself an IBM veteran with over 20 years of service (ADSM is his). IBM passes their business on to stand-ins.
Such is IBM's focus on the rarified high high-end and such is its fascination with 450-MIPS building blocks that it has stranded less prestigious accounts without an affordable migration path past the G4 and G5, Maulick says.
Things being what they are, he says, second-tier companies are being forced to pay $8,000-$10,000 per MIPS where their better-off betters are paying only $2,000.
Knowing that PSI is coming, IBM threw these second-tier accounts a bone in July when it introduced the dumbed-down z990 it dubbed the z890. It pulled a few chips out of the 990, lowered its clock speed, crippled its cache, reused its packaging and, knowing that PSI would start at 26 MIPS, said the 890 would do that.
Maulick claims it's not a solution and, at roughly the same price as the 990, certainly not a reasonable one.
PSI aims to address this disenfranchised market, believing that it's also elastic. It figures, for instance, that there are high-end Sun users that would love to go back to the safety and security of the MVS RAS and can't get there from where they are.
At this point, PSI is only thinking about carving off a piece of the mainframe iron market, not the $24 billion service and software business connected with it.
PSI is playing for big stakes - and ultimately a big IPO - remember that 60%-70% of the world's mission-critical data is still on the mainframe - and it has a few little friends helping it take on the IBM monopoly.
It just got an undisclosed amount of money in second-round financing - undisclosed because it doesn't want IBM to know what its resources are and try to drain them, but it must be a tidy pile because it's supposed to take PSI past profitability in, oh, 2006.
The money came from Goldman Sachs, which led Blueprint Ventures, InterWest Partners and InvestCorp, all of which must be delirious at the thought that PSI will have no competition.
As Maulick points out, it's not exactly a field where VCs are funding a hundred start-ups. The barrier to entry, he figures, is roughly $500 million, the value he puts on the tools, diagnostics and testing widgetry PSI got from Amdahl.
Before deciding to take the job with PSI, Maulick was instrumental in raising PSI's $10.2 million A round last September. Goldman wasn't part of that financing but Fujitsu, which owned Amdahl, and Intel were.
Until last year, PSI was kept alive by $500,000 in angel droppings, Maulick said.
PSI's charter was originally Fujitsu's Plan B for how to salvage Amdahl and the notion of using Itanium as Fujitsu's back door in the Alamo dates back to 1995 when the Itanium was still a gleam in Intel's eye. When Itanium was late, Fujitsu abandoned the field and would have deep six'd its IP if it hadn't been persuaded to spin it off. PSI subsequently looked at IBM's Power chip, but decided the "underrated" Itanium was better.
The guy who first saw PSI's commercial implications, according to Maulick, was George Hoyem, the long-time VC, now Blueprint Ventures' managing partner, who once upon a time was general manager of HP's Internet Commerce Division after he co-founded Visix Software. Hoyem put together PSI's first round.
Naturally, the mere mention of HP brings to mind its IBM-besting ambitions and the idea of its, oh, Itanium-based Superdome boxes being used as mainframe-aping systems against its hereditary rival. Maulick won't confirm the possibility, but PSI is developing its so-called Universal Server on HP hardware as well as Intel white boxes.
PSI says the second-round money will be used to expand its development efforts and to fund product introduction. The outfit was at the IBM Share user meeting in New York this week showing off its stuff to the elect in a suite under NDA.
Universal Server is currently in alpha. PSI intends to beta the system in Q4 and go GA in Q1.
PSI has wrapped core microcode and patented virtualization mojo into the Universal Server's architecture and Goldman is evidently heady with delight at the prospect of running z/OS on an open systems server. It speaks of the widgetry as "reinvigorating the PCM market in a 64-bit world."
Maulick describes it as "reincarnating Amdahl" without Amdahl. Amdahl used to have 24% of the mainframe market. If PSI gets just 10%, Maulick will count it highly successful.
PSI, which will sell multi-processor servers that range into the thousands of MIPS, thinks it's got a world-class alternative to IBM's z/Series and p/Series. It believes it can argue against the wisdom of investing in a single-purpose proprietary platform that, it says, limits the exploitation of open technologies and eats up IT resources trying to solve management issues.
In PSI's architecture, Windows, Linux and Unix will run natively on Itanium, and z/OS and OS/390 will run on top of what the company calls its PCMWare. All the operating systems will be controlled from a single management console.
Ironically, if IBM Software weren't so captive to IBM's hardware interests, IBM Software chief Steve Mills would probably be beating a path to PSI's door.
About the author
Maureen O'Gara is editor-in-chief of Maureen O'Gara's LinuxGram(TM) - published weekly by G2 Computer Intelligence Inc. and distributed by Linux Business Week.
Copyright © 1998-2004 LinuxWorld.com, an IDG Communications company.
"One of the writers of the so called "test" worked for Intel Randall C. Kennedy"
Attack mode - ON - in full force.
pgerassi - Reporting For Duty !
how sweet and predictable you FBs are.
"Intel's Xeon-based workstations are much faster than workstations based on AMD's Opteron when it comes to heavy multitasking ...
the Opteron system took an average of 15 percent longer to complete the tasks than the Xeon. In some cases, most notably client/server workflow against a MAPI message store, the Opteron took over 30 percent longer. "
Let's see if the AMD FB Vigilantes go into ATTACKMODE now - to force the reviewer to drop any benchmarks that are favorable to Intel products.
http://www.infoworld.com/article/04/08/13/33TCworkstation_1.html
Workstation showdown: Xeon vs. Opteron
Intel's Xeon-based workstations are much faster than workstations based on AMD's Opteron when it comes to heavy multitasking
By Wayne Rash , Randall C. Kennedy
August 13, 2004
In business, dual-processor workstations are the trucks of desktop computing. Pickups aren't very glamorous, but when you need to move lots of stuff from one point to another efficiently, glamour isn't the point. You need a truck.
Sure, you can get computers that are snazzier, have faster clock rates, and sport cool accessories, but as fast as they may go, they can only do one thing well at a time. These dual-processor workstations are designed to perform many processes at the same time and to work quickly and efficiently.
For this reason, we were intrigued when IBM offered up its new dual-Opteron IntelliStation A Pro workstation for a test. The hype on the street had been singing the Opteron's praises for some time. According to its proponents, the Opteron was the greatest thing since night baseball; curmudgeons that we are (though we do like night baseball), we weren't so sure.
So we decided to pit the Opteron against its primary rival, Intel's tried and true Xeon. We asked MPC to send over the workstation version of its NetFrame 600. This platform, designed initially as a server, can be ordered in a workstation version with a high-end video card and an attractive beige case.
The NF 600 sports two 3.2GHz Xeons, a gig of memory, and some very fast SATA hard disks. This matched nicely with IBM's SATA drives, although the IBM A Pro had 4GB of memory.
Then, just to push the truck comparison one step further, we purchased a refurbished Hewlett-Packard xw8000 workstation with a pair of 3.06GHz Xeon processors, an ATA drive, and a gig of memory. After all, sometimes your best deal is to pick a used truck at the dealer if you need the capability and want to save a ton of money.
Load 'Em Up
Once we had the workstations in-house, we ran a series of tests simulating a real-world work environment and then testing the machines for their absolute best multitasking performance.
In the first series, we loaded up each machine with the full suite of commercial applications you'd expect to see used in the financial services or the content-production industries. This test included the full install of Microsoft Office 2003, Adobe Photoshop, and Premier Pro 1.5. We also installed Symantec's Norton AntiVirus 2004. We used the workstations to move through a defined set of tasks in manipulating still images and then in producing a movie starting with an original digital video. We timed processing wherever possible (these machines are so fast that Photoshop works with no discernable delay).
In the second test, we loaded CSA Research's OfficeBench test tool. Someone actually using these workstations would never have a completely clean environment, so we ran a test series with everything installed and Norton AntiVirus running. (We also ran OfficeBench on the clean systems.)
Next, we loaded up another CSA Research tool, Clarity Studio. With Clarity Studio, we simulated multiple, concurrent workloads running in parallel -- the kind of complex, data-intensive multitasking that's becoming commonplace in emerging workstation markets (see "How we put the workstations under pressure").
What we found was eye-opening. The Opteron machine outperformed the Xeons when lightly loaded with minimal multitasking, but once the real work started, the Opteron stopped. It was effectively shut down by the same multitasking load that the two Xeons performed with ease. In the clean environment, it still performed at less than half the speed of the older and allegedly less-capable Xeons.
Initially we suspected that part of the reason for the A Pro's surprising performance under heavy loads might have been due to the fact that the unit we reviewed was a late preproduction model. We brought in and tested a new production A Pro from IBM, which did come through with improved performance numbers.
However, to be fair, we updated the BIOS in the NF 600 and installed the latest drivers. It also showed an improvement in speed that almost exactly equaled the IBM's improvement, so there was no difference in the relative performance of these two computers.
IBM A Pro
With its 4GB of memory and Nvidia 1100 graphics card, the IBM A Pro should have been the ultimate workstation in this review. The vast memory resources make it a natural for the classic workstation environment where lots of applications need to run together in real time.
The A Pro has the usual IBM workstation case that opens easily to reveal a well-designed, accessible interior with a selection of PCI-X slots for easy expansion. As has become the norm in high-end desktops and workstations, the A Pro includes USB, IEEE 1394, and sound card connectors on the front panel and on the rear of the box. Video capabilities include support for dual monitors using DVI-I connectors.
Although larger (mostly in depth) than usual for a desktop computer, the A Pro would still fit on most desktops. It also slides nicely alongside a desk, with the optical drives and the I/O ports conveniently at hand. Clearly, a great deal of thought has gone into the ergonomic design of the A Pro.
Initial testing kept our hopes up that the A Pro would fulfill its initial promise. We used Adobe Premier Pro to create a movie from an original digital source, and the encoding process was very fast indeed. Operations using Photoshop showed no discernable delay, regardless of the filter applied or the operations attempted. We also ran Futuremark's PCMark04 benchmark, which is mostly single-tasking, and the A Pro ran slightly faster than either of the Xeon machines.
(Page 2 of 2)
By Wayne Rash , Randall C. Kennedy
August 13, 2004
When we moved on to the multitasking tests using OfficeBench and Clarity Studio, however, the Opteron showed its limitations. By the time we got to the tests that used heavy multitasking, the A Pro was running a lot slower. In its best case, the Opteron ran about two-thirds as fast as the least capable Xeon, the HP xw8000. Though it handles single-task processing very well, if multitasking is in your future, the A Pro is not the right choice.
MPC NetFrame 600
The NF 600 workstation is really MPC's dual-Xeon server with better video and a nice case. It uses an Intel server motherboard and retains all of the reliability features you'd expect in a departmental server. This includes hot-swap power supplies and fans, as well as front-accessible disk drives. It also means, however, that some workstation features (notably the IEEE 1394 port) are missing.
The NF 600's case also reminds you that it was born a server -- the cover slides off to the rear, just like its rack-mount siblings. Inside, the hot-swap, redundant, fan array, foam-mounted for sound isolation, reinforces the NF 600's server platform roots.
The PCI-X expansion slots are protected by clear plastic air baffles, and everything is easily accessible for quick, convenient replacement, even while rack-mounted. The redundancy and reliability features, as well as the heavy-duty construction, make this workstation much larger and heavier than the others in this test. You wouldn't want the NF 600 on your desktop.
However, the server-based MPC showed its mettle in our tests. The NF 600 was nearly as fast as the IBM A Pro even in single tasking, and it blew the IBM away in multi-tasking, despite the IBM A Pro having four times the memory.
In our real-world testing with multiple applications and tasks, the MPC was many times faster than the Opteron-based IBM. Even in our clean multitasking tests, the MPC was 30 percent faster than the IBM. It simply wasn't a contest. As a result, the NF 600 turned out to be the fastest workstation that InfoWorld has tested to date.
HP Workstation xw8000
We kind of threw in a ringer here. To get an idea of what a business could buy that provided both excellent performance and reasonable cost, we called Sabrina Bozant, who sells HP's refurbished workstations, and asked for a 3.06GHz Xeon workstation. She sold us one for $1,395 and added an additional processor for $549.
Installing the second processor was a simple task because these earlier generation Xeons had a smaller 512KB cache. Like the MPC, it has 1GB of memory and an Nvidia video card, if a lower-end one.
By choosing a refurbished high-end workstation, we paid about one third of the cost of the MPC and a quarter of the price of the IBM (a new HP xw8000 would have cost about twice as much as our refurbished model).
The xw8000 is beautifully designed for its intended use as an office workstation. It's small enough that you could use it on the desktop, but it will work just fine alongside a desk. The top-mounted optical drives are easy to reach, and the midmounted power button is also convenient. Unfortunately, USB, FireWire, and sound connectors are inconveniently located at the bottom of the front panel (there's also a full set on the rear).
The case opens easily with a simple latch to reveal a very clean design with minimal cable intrusion. You will need to remove a holder for the video card -- but not the card itself -- to install the second processor and perform other work inside the case, but this is a minor inconvenience.
For the most part, the xw8000 performed like the MPC NF 600, other than being slightly slower, which we expected from the slightly slower processor. Single-tasking tests showed that the xw8000 was slightly slower than the Opteron.
But also like the MPC, the xw8000 was much faster than the Opteron when multitasking demands grew heavier, and the xw8000 breezed through the most demanding tests that brought the IBM to its knees. A current model of the xw8000 with a larger cache would likely do even better, as would HP's just-released Workstation xw8200, which we'll review in the near future.
Xeon 1, Opteron 0
After all our tests, we found that the most demanding jobs ran best on the dual-Xeon processor with its ability to run hyperthreading. The dual-Opteron, although faster in less demanding environments, simply wasn't a match when the going got tough.
That doesn't mean it doesn't have a place in the enterprise, though; an Opteron-based system would be a good choice for tasks such as CAD, which is basically a single-task, high-performance-requiring process.
Xeon's speed is good news for financial services companies such as Morgan Stanley, Goldman Sachs, and Credit Suisse First Boston, which have long used workstations to deliver the massive computing power required to drive their trading operations (a single active trader can easily bury a top-of-the-line PC). In an environment where time literally is money, the improvements coming down the pike for the Xeon platform should be welcome news for those firms with heavy investments in Intel-based workstations. It also means that AMD will have to do some serious tuning before Opteron poses a significant threat to Intel in the high-end workstation market.
<< Previous Page / 1 / 2
HP Workstation xw8000
Hewlett-Packard, hp.com
Excellent 8.9
criteria
score
weight
Performance
9
30%
Scalability
9
25%
Manageability
9
20%
Ease of use
9
15%
Value
8
10%
Cost:
$1,944 (refurbished)
Platforms:
Windows 2000 or XP Professional, Red Hat Linux
Bottom Line:
HP’s xw8000 is a well-designed, very capable workstation that easily handles the heaviest workloads. It is available with single or dual Xeon processors running as fast as 3.06GHz and is designed for easy access and tool-free maintenance. It sets the standard for dual-processor workstations and bested Opteron in heavy multitasking tests.
About our Reviews and Scoring Methodology
IBM IntelliStation A Pro with AMD Opteron Processor
IBM, ibm.com
Very Good 7.7
criteria
score
weight
Performance
6
30%
Scalability
8
25%
Manageability
9
20%
Ease of use
9
15%
Value
7
10%
Cost:
$8,345 as tested
Platforms:
Windows 2000 and XP Professional, Red Hat Enterprise Linux
Bottom Line:
This is a very fast machine -- as long as you’re only trying to do a few tasks at a time. It’s well-designed, and access is easy and convenient. However, if you load it down with heavy processing, the 2.4GHz Opterons show their limitations and the A Pro starts to crawl.
About our Reviews and Scoring Methodology
MPC NetFrame 600 (Workstation version)
MPC, buympc.com
Very Good 8.5
criteria
score
weight
Performance
9
30%
Scalability
9
25%
Manageability
8
20%
Ease of use
8
15%
Value
7
10%
Cost:
$6,055 as tested
Platforms:
Windows XP Professional
Bottom Line:
The NF 600 is big, and it has the hot-swap redundant features you’d expect from a server. There’s plenty of room for storage and it includes a wide variety of high-end video cards and an office-friendly case. Unfortunately, its server roots also showed in its lack of an IEEE 1394 port. Its 3.2GHz Xeons helped make this the performance king.
About our Reviews and Scoring Methodology
http://www.infoworld.com/article/04/08/13/33TCworkstation-sb_1.html
How we put the workstations under pressure
Loading up the HP, IBM, and MPC workstations involved lots of parallel workloads
By Randall C. Kennedy
August 13, 2004
Enter the workstation: designed for concurrent multiprocessing, workstations are rugged and reliable, with multiple, symmetric CPUs and gobs of memory to power through even the toughest workloads. You need to really load these machines down before their relative merits begin to surface, and that means generating concurrent workloads that exercise a variety of OS and application subsystems.
For this review, we did just that. I utilized one of my favorite test tools, Clarity Studio from CSA Research. Using a combination of parallel workloads -- client/server database (specifically, ActiveX Data Objects), workflow (MAPI), Windows Media playback, and Windows Media encoding -- I generated a hailstorm of CPU and memory activity.
I then scaled these workloads on each system, increasing the number of concurrent tasks as well as their complexity, all the while tracking the systems' performance and health through various internal and external metrics counters.
The net result? Despite a great deal of hype, AMD's 2.2GHz Opteron 248 CPU -- as embodied in the IBM IntelliStation A Pro workstation -- doesn't fare well under heavy workloads. When compared head-to-head with last year's Intel Xeon platform, a 3.2GHz/533MHz Front Side Bus model represented here by the MPC NetFrame 600, the Opteron fades as the workload level increases.
In fact, across the range of tests, the Opteron system took an average of 15 percent longer to complete the tasks than the Xeon. In some cases, most notably client/server workflow against a MAPI message store, the Opteron took over 30 percent longer.
An examination of OS metrics data collected by Clarity Studio showed that the Opteron was definitely struggling to juggle all those threads. One metric in particular shed additional light on the results. The Peak CPU Saturation Index, which is calculated from a sampling of the Processor Queue Length counter as exposed by the Windows Performance Data Helper libraries, showed that, on average, the Opteron system had 16 percent more waiting threads in its queue -- a clear indication that the system was in fact CPU-bound and running out of processor bandwidth.
My interpretation: Hyper-threading support on the Xeon allowed it to continue to scale thanks to its ability to execute more than one instruction at a time. Once again, Intel's simultaneous multitasking technology -- where underutilized pipeline resources are shared to create a second, virtual processor image -- is looking like an ace in the hole for the company's workstation strategy.
The story gets worse for AMD when you factor in the newest Xeon processors from Intel. Preliminary results from two systems based on the new 800MHz FSB Xeon show the aforementioned average performance gap widening to nearly 50 percent (the MAPI workload, in particular, is now running 115 percent faster than Opteron), with CPU Saturation now 30 percent higher for Opteron when compared to the next-generation Xeon CPU (watch for our expanded coverage in an upcoming issue).
"Intel's Xeon-based workstations are much faster than workstations based on AMD's Opteron when it comes to heavy multitasking ...
the Opteron system took an average of 15 percent longer to complete the tasks than the Xeon. In some cases, most notably client/server workflow against a MAPI message store, the Opteron took over 30 percent longer. "
Time for you, dougie, to organize your AMDFanboy Vigilantes and attack the Infoworld Authors - and get them to drop any benchmaks such as multitasking - that are dominated by Intel's CPUs.
http://www.infoworld.com/article/04/08/13/33TCworkstation_1.html
Workstation showdown: Xeon vs. Opteron
Intel's Xeon-based workstations are much faster than workstations based on AMD's Opteron when it comes to heavy multitasking
By Wayne Rash , Randall C. Kennedy
August 13, 2004
In business, dual-processor workstations are the trucks of desktop computing. Pickups aren't very glamorous, but when you need to move lots of stuff from one point to another efficiently, glamour isn't the point. You need a truck.
Sure, you can get computers that are snazzier, have faster clock rates, and sport cool accessories, but as fast as they may go, they can only do one thing well at a time. These dual-processor workstations are designed to perform many processes at the same time and to work quickly and efficiently.
For this reason, we were intrigued when IBM offered up its new dual-Opteron IntelliStation A Pro workstation for a test. The hype on the street had been singing the Opteron's praises for some time. According to its proponents, the Opteron was the greatest thing since night baseball; curmudgeons that we are (though we do like night baseball), we weren't so sure.
So we decided to pit the Opteron against its primary rival, Intel's tried and true Xeon. We asked MPC to send over the workstation version of its NetFrame 600. This platform, designed initially as a server, can be ordered in a workstation version with a high-end video card and an attractive beige case.
The NF 600 sports two 3.2GHz Xeons, a gig of memory, and some very fast SATA hard disks. This matched nicely with IBM's SATA drives, although the IBM A Pro had 4GB of memory.
Then, just to push the truck comparison one step further, we purchased a refurbished Hewlett-Packard xw8000 workstation with a pair of 3.06GHz Xeon processors, an ATA drive, and a gig of memory. After all, sometimes your best deal is to pick a used truck at the dealer if you need the capability and want to save a ton of money.
Load 'Em Up
Once we had the workstations in-house, we ran a series of tests simulating a real-world work environment and then testing the machines for their absolute best multitasking performance.
In the first series, we loaded up each machine with the full suite of commercial applications you'd expect to see used in the financial services or the content-production industries. This test included the full install of Microsoft Office 2003, Adobe Photoshop, and Premier Pro 1.5. We also installed Symantec's Norton AntiVirus 2004. We used the workstations to move through a defined set of tasks in manipulating still images and then in producing a movie starting with an original digital video. We timed processing wherever possible (these machines are so fast that Photoshop works with no discernable delay).
In the second test, we loaded CSA Research's OfficeBench test tool. Someone actually using these workstations would never have a completely clean environment, so we ran a test series with everything installed and Norton AntiVirus running. (We also ran OfficeBench on the clean systems.)
Next, we loaded up another CSA Research tool, Clarity Studio. With Clarity Studio, we simulated multiple, concurrent workloads running in parallel -- the kind of complex, data-intensive multitasking that's becoming commonplace in emerging workstation markets (see "How we put the workstations under pressure").
What we found was eye-opening. The Opteron machine outperformed the Xeons when lightly loaded with minimal multitasking, but once the real work started, the Opteron stopped. It was effectively shut down by the same multitasking load that the two Xeons performed with ease. In the clean environment, it still performed at less than half the speed of the older and allegedly less-capable Xeons.
Initially we suspected that part of the reason for the A Pro's surprising performance under heavy loads might have been due to the fact that the unit we reviewed was a late preproduction model. We brought in and tested a new production A Pro from IBM, which did come through with improved performance numbers.
However, to be fair, we updated the BIOS in the NF 600 and installed the latest drivers. It also showed an improvement in speed that almost exactly equaled the IBM's improvement, so there was no difference in the relative performance of these two computers.
IBM A Pro
With its 4GB of memory and Nvidia 1100 graphics card, the IBM A Pro should have been the ultimate workstation in this review. The vast memory resources make it a natural for the classic workstation environment where lots of applications need to run together in real time.
The A Pro has the usual IBM workstation case that opens easily to reveal a well-designed, accessible interior with a selection of PCI-X slots for easy expansion. As has become the norm in high-end desktops and workstations, the A Pro includes USB, IEEE 1394, and sound card connectors on the front panel and on the rear of the box. Video capabilities include support for dual monitors using DVI-I connectors.
Although larger (mostly in depth) than usual for a desktop computer, the A Pro would still fit on most desktops. It also slides nicely alongside a desk, with the optical drives and the I/O ports conveniently at hand. Clearly, a great deal of thought has gone into the ergonomic design of the A Pro.
Initial testing kept our hopes up that the A Pro would fulfill its initial promise. We used Adobe Premier Pro to create a movie from an original digital source, and the encoding process was very fast indeed. Operations using Photoshop showed no discernable delay, regardless of the filter applied or the operations attempted. We also ran Futuremark's PCMark04 benchmark, which is mostly single-tasking, and the A Pro ran slightly faster than either of the Xeon machines.
(Page 2 of 2)
By Wayne Rash , Randall C. Kennedy
August 13, 2004
When we moved on to the multitasking tests using OfficeBench and Clarity Studio, however, the Opteron showed its limitations. By the time we got to the tests that used heavy multitasking, the A Pro was running a lot slower. In its best case, the Opteron ran about two-thirds as fast as the least capable Xeon, the HP xw8000. Though it handles single-task processing very well, if multitasking is in your future, the A Pro is not the right choice.
MPC NetFrame 600
The NF 600 workstation is really MPC's dual-Xeon server with better video and a nice case. It uses an Intel server motherboard and retains all of the reliability features you'd expect in a departmental server. This includes hot-swap power supplies and fans, as well as front-accessible disk drives. It also means, however, that some workstation features (notably the IEEE 1394 port) are missing.
The NF 600's case also reminds you that it was born a server -- the cover slides off to the rear, just like its rack-mount siblings. Inside, the hot-swap, redundant, fan array, foam-mounted for sound isolation, reinforces the NF 600's server platform roots.
The PCI-X expansion slots are protected by clear plastic air baffles, and everything is easily accessible for quick, convenient replacement, even while rack-mounted. The redundancy and reliability features, as well as the heavy-duty construction, make this workstation much larger and heavier than the others in this test. You wouldn't want the NF 600 on your desktop.
However, the server-based MPC showed its mettle in our tests. The NF 600 was nearly as fast as the IBM A Pro even in single tasking, and it blew the IBM away in multi-tasking, despite the IBM A Pro having four times the memory.
In our real-world testing with multiple applications and tasks, the MPC was many times faster than the Opteron-based IBM. Even in our clean multitasking tests, the MPC was 30 percent faster than the IBM. It simply wasn't a contest. As a result, the NF 600 turned out to be the fastest workstation that InfoWorld has tested to date.
HP Workstation xw8000
We kind of threw in a ringer here. To get an idea of what a business could buy that provided both excellent performance and reasonable cost, we called Sabrina Bozant, who sells HP's refurbished workstations, and asked for a 3.06GHz Xeon workstation. She sold us one for $1,395 and added an additional processor for $549.
Installing the second processor was a simple task because these earlier generation Xeons had a smaller 512KB cache. Like the MPC, it has 1GB of memory and an Nvidia video card, if a lower-end one.
By choosing a refurbished high-end workstation, we paid about one third of the cost of the MPC and a quarter of the price of the IBM (a new HP xw8000 would have cost about twice as much as our refurbished model).
The xw8000 is beautifully designed for its intended use as an office workstation. It's small enough that you could use it on the desktop, but it will work just fine alongside a desk. The top-mounted optical drives are easy to reach, and the midmounted power button is also convenient. Unfortunately, USB, FireWire, and sound connectors are inconveniently located at the bottom of the front panel (there's also a full set on the rear).
The case opens easily with a simple latch to reveal a very clean design with minimal cable intrusion. You will need to remove a holder for the video card -- but not the card itself -- to install the second processor and perform other work inside the case, but this is a minor inconvenience.
For the most part, the xw8000 performed like the MPC NF 600, other than being slightly slower, which we expected from the slightly slower processor. Single-tasking tests showed that the xw8000 was slightly slower than the Opteron.
But also like the MPC, the xw8000 was much faster than the Opteron when multitasking demands grew heavier, and the xw8000 breezed through the most demanding tests that brought the IBM to its knees. A current model of the xw8000 with a larger cache would likely do even better, as would HP's just-released Workstation xw8200, which we'll review in the near future.
Xeon 1, Opteron 0
After all our tests, we found that the most demanding jobs ran best on the dual-Xeon processor with its ability to run hyperthreading. The dual-Opteron, although faster in less demanding environments, simply wasn't a match when the going got tough.
That doesn't mean it doesn't have a place in the enterprise, though; an Opteron-based system would be a good choice for tasks such as CAD, which is basically a single-task, high-performance-requiring process.
Xeon's speed is good news for financial services companies such as Morgan Stanley, Goldman Sachs, and Credit Suisse First Boston, which have long used workstations to deliver the massive computing power required to drive their trading operations (a single active trader can easily bury a top-of-the-line PC). In an environment where time literally is money, the improvements coming down the pike for the Xeon platform should be welcome news for those firms with heavy investments in Intel-based workstations. It also means that AMD will have to do some serious tuning before Opteron poses a significant threat to Intel in the high-end workstation market.
<< Previous Page / 1 / 2
HP Workstation xw8000
Hewlett-Packard, hp.com
Excellent 8.9
criteria
score
weight
Performance
9
30%
Scalability
9
25%
Manageability
9
20%
Ease of use
9
15%
Value
8
10%
Cost:
$1,944 (refurbished)
Platforms:
Windows 2000 or XP Professional, Red Hat Linux
Bottom Line:
HP’s xw8000 is a well-designed, very capable workstation that easily handles the heaviest workloads. It is available with single or dual Xeon processors running as fast as 3.06GHz and is designed for easy access and tool-free maintenance. It sets the standard for dual-processor workstations and bested Opteron in heavy multitasking tests.
About our Reviews and Scoring Methodology
IBM IntelliStation A Pro with AMD Opteron Processor
IBM, ibm.com
Very Good 7.7
criteria
score
weight
Performance
6
30%
Scalability
8
25%
Manageability
9
20%
Ease of use
9
15%
Value
7
10%
Cost:
$8,345 as tested
Platforms:
Windows 2000 and XP Professional, Red Hat Enterprise Linux
Bottom Line:
This is a very fast machine -- as long as you’re only trying to do a few tasks at a time. It’s well-designed, and access is easy and convenient. However, if you load it down with heavy processing, the 2.4GHz Opterons show their limitations and the A Pro starts to crawl.
About our Reviews and Scoring Methodology
MPC NetFrame 600 (Workstation version)
MPC, buympc.com
Very Good 8.5
criteria
score
weight
Performance
9
30%
Scalability
9
25%
Manageability
8
20%
Ease of use
8
15%
Value
7
10%
Cost:
$6,055 as tested
Platforms:
Windows XP Professional
Bottom Line:
The NF 600 is big, and it has the hot-swap redundant features you’d expect from a server. There’s plenty of room for storage and it includes a wide variety of high-end video cards and an office-friendly case. Unfortunately, its server roots also showed in its lack of an IEEE 1394 port. Its 3.2GHz Xeons helped make this the performance king.
About our Reviews and Scoring Methodology
http://www.infoworld.com/article/04/08/13/33TCworkstation-sb_1.html
How we put the workstations under pressure
Loading up the HP, IBM, and MPC workstations involved lots of parallel workloads
By Randall C. Kennedy
August 13, 2004
Enter the workstation: designed for concurrent multiprocessing, workstations are rugged and reliable, with multiple, symmetric CPUs and gobs of memory to power through even the toughest workloads. You need to really load these machines down before their relative merits begin to surface, and that means generating concurrent workloads that exercise a variety of OS and application subsystems.
For this review, we did just that. I utilized one of my favorite test tools, Clarity Studio from CSA Research. Using a combination of parallel workloads -- client/server database (specifically, ActiveX Data Objects), workflow (MAPI), Windows Media playback, and Windows Media encoding -- I generated a hailstorm of CPU and memory activity.
I then scaled these workloads on each system, increasing the number of concurrent tasks as well as their complexity, all the while tracking the systems' performance and health through various internal and external metrics counters.
The net result? Despite a great deal of hype, AMD's 2.2GHz Opteron 248 CPU -- as embodied in the IBM IntelliStation A Pro workstation -- doesn't fare well under heavy workloads. When compared head-to-head with last year's Intel Xeon platform, a 3.2GHz/533MHz Front Side Bus model represented here by the MPC NetFrame 600, the Opteron fades as the workload level increases.
In fact, across the range of tests, the Opteron system took an average of 15 percent longer to complete the tasks than the Xeon. In some cases, most notably client/server workflow against a MAPI message store, the Opteron took over 30 percent longer.
An examination of OS metrics data collected by Clarity Studio showed that the Opteron was definitely struggling to juggle all those threads. One metric in particular shed additional light on the results. The Peak CPU Saturation Index, which is calculated from a sampling of the Processor Queue Length counter as exposed by the Windows Performance Data Helper libraries, showed that, on average, the Opteron system had 16 percent more waiting threads in its queue -- a clear indication that the system was in fact CPU-bound and running out of processor bandwidth.
My interpretation: Hyper-threading support on the Xeon allowed it to continue to scale thanks to its ability to execute more than one instruction at a time. Once again, Intel's simultaneous multitasking technology -- where underutilized pipeline resources are shared to create a second, virtual processor image -- is looking like an ace in the hole for the company's workstation strategy.
The story gets worse for AMD when you factor in the newest Xeon processors from Intel. Preliminary results from two systems based on the new 800MHz FSB Xeon show the aforementioned average performance gap widening to nearly 50 percent (the MAPI workload, in particular, is now running 115 percent faster than Opteron), with CPU Saturation now 30 percent higher for Opteron when compared to the next-generation Xeon CPU (watch for our expanded coverage in an upcoming issue).
This will gove the AMD FBs some indigestion:
Anandtech is reporting some rather striking benchmark results comparing the Athlon 64 3500 to Intel's Xeon 3.6 MHz with 64 bit extensions.
64 Bit SUSE Linux was used along with an array of Linux benchmarks.
http://www.anandtech.com/linux/showdoc.aspx?i=2158
Linux and EM64T; Intel's 64-bit Suggestion
In summary:
"Without a doubt, the 3.6GHz Xeon trounces over the Athlon 64 in math-intensive benchmarks. Intel came ahead in every severe benchmark that we could throw at it,"
Can 't wait for the AMD FB spinmeisters to twist this one into a pretzel.
Maybe Dell, IBM, HP, etc. had a good reason to launch all those Xeon64 servers and workstations last week.
Anandtech is reporting some rather striking benchmark results comparing the Athlon 64 3500 to Intel's Xeon 3.6 MHz with 64 bit extensions.
64 Bit SUSE Linux was used along with an array of Linux benchmarks.
http://www.anandtech.com/linux/showdoc.aspx?i=2158
Linux and EM64T; Intel's 64-bit Suggestion
In summary:
"Without a doubt, the 3.6GHz Xeon trounces over the Athlon 64 in math-intensive benchmarks. Intel came ahead in every severe benchmark that we could throw at it,"
Can 't wait for the AMD FB spinmeisters to twist this one into a pretzel.
Maybe Dell, IBM, HP, etc. had a good reason to launch all those Xeon64 servers and workstations last week.
"high performance, boutique stuff doesn't usually pay the bills."
Sounds like Nvidia was expecting it to paytheir bills this past quarter - and that they were banking on Intel platforms to help pay those bills.
Oh well, Nvidia has all those AMD-based chip sets to pull them into financial prosperity ( yeah, right !)
"Because it is already shipping? So how does that make it late?"
That article you sited confirms that Odessa did NOT ship in H1 2004 the orginal ship date - depsite its claim that it will ship early - and that it has slipped at least into H2 2004 - = and EVEN WORSE - it is no longer a 90 nanometer device - but had to be sucked back to 130 nanometers - another AMD slip up.
You just confirmed everything I said - thank you.
And don't forget to reprimand dougie for not bashing AMD about AMD's myriad of product slips.
"AMD sneaks out 90nm core in 130nm chip
By Tony Smith
Published Tuesday 11th May 2004 08:57 GMT
AMD's 'Odessa' chip, originally planned to be the company's first 90nm mobile part, has come to market early, it has emerged.
But AMD watchers hoping for an early release of the chip maker's next-generation fabrication technology will be disappointed: Odessa is now a 130nm part.
AMD's public roadmap currently calls for Odessa to appear sometime during H2 2004.
"it's all about volume..."
The volume of the fanboys' vocal denials?
"And Odessa has been pulled in 2 quarters"
How can Odessa have been pulled in two quarters when it is already at least one quarter late?
"you don't know that they haven't produced those parts. Just that they haven't announced them. Of course, facts aren't your strong point, are they?"
Facts are obviously not your strong point either - are they? You just throw up a straw man assumption - with absolutely no facts to back it up yourself - right?
Maybe you are suggesting that once again, AMD is producing volumes of brand new products and stockpiling them in an already-over-stuffed-warehouse-filled-with-old-stockpiled-CPUs.
Hm.....the old AMD stockpiling gambit...always a nice ploy to explain away unavailable devices from AMD - n'est pas?
"Intel has had a lot of parts "scheduled" for certain timeframes, lately."
AMD had Athens, Odessa and San Diego - all 90 nanometer devices - scheduled for the first half of 2004 - which concluded over a month ago.
However, AMD has produced none of these parts - and obvioulsly has slipped their scheduled.
Why aren't you burning up this thread and the others that you post on, doug, with your slams against AMD's execution?
"Nvidia...they also blamed the grantsdale recall for stalling their higher end stuff as well."
Rather an interesting detail of Nvidia's business - which must bring fear into the AMD fan boys' hearts.
What Nvidia is saying is that their high end graphics boards are heading into Intel-based PCs - and not the ga-ga Opteron PCs that the fan boys would have you believe.
Will the Sempron brand be as Durable as the Duron brand?
AMD will just have to be satisfied with having the fastest 32 bit desktop under Windows
Can't be that fast - cuz nobody seems to be buying them.
Or else, AMD can't make them in any volume.
AMD's pig in a poke Athlon 64/Opteron - no 64 bit OS for another half year - or possibly another year - courtesy of Microsoft.
Ah yes - AMD - the promise of 64 bit addressing - is still only a promise - one that keeps getting broken.
Mr. Sanders should have given better testimony at the US trial against Microsoft.
I guess the AMD FBs will reconvert to the Linux Faith.
Microsoft delays Windows Server update, 64-bit OS
By Ina Fried
Staff Writer, CNET News.com
http://news.com.com/2100-1016-5285913.html
Story last modified July 27, 2004, 2:32 PM PDT
Microsoft said Tuesday that it was delaying an update to its Windows Server software as well as the first version of Windows to support 64-bit x86 chips such as Advanced Micro Device's Athlon 64.
The software maker said Windows Server 2003 Service Pack 1, Windows Server 2003 for 64-bit Extended Systems and Windows XP 64-bit Edition for 64-bit Extended Systems will all be pushed back until the first half of next year. All three products, which are being developed together, were slated to ship by the end of this year.
The move is a blow to AMD, which has had its 64-bit desktop and server chips on the market for some time and has been waiting for a version of Windows that can take advantage of their capabilities. Intel has also said it will have chips that support the 64-bit extensions to the existing Pentium architecture, known as x86. There is already a version of Windows that supports Intel's 64-bit Itanium architecture, which uses an entirely separate instruction set than x86 chips.
"As is the case with all Microsoft product schedules, the development cycle is driven by quality with a focus on the needs of our customers, rather than an arbitrary date," Microsoft said in a statement.
The move is the latest in a series of delays for the 64-bit version of Windows XP.
AMD announced its first 64-bit chips in September 2003, at which time Microsoft had planned to release, by the first half of this year, a version of Windows that would support the chips.
Microsoft has also had to push back other versions of Windows. The long-awaited security update to Windows XP--Service Pack 2--has taken longer than expected and is now set to ship in August. Microsoft has also said that the next major release of Windows, code-named Longhorn, won't ship until 2006.
AMD's pig in a poke Athlon 64/Opteron - no 64 bit OS for another half year - or possibly another year - courtesy of Microsoft.
Ah yes - AMD - the promise of 64 bit addressing - is still only a promise - one that keeps getting broken.
Mr. Sanders should have given better testimony at the US trial against Microsoft.
I guess the AMD FBs will reconvert to the Linux Faith.
Microsoft delays Windows Server update, 64-bit OS
By Ina Fried
Staff Writer, CNET News.com
http://news.com.com/2100-1016-5285913.html
Story last modified July 27, 2004, 2:32 PM PDT
Microsoft said Tuesday that it was delaying an update to its Windows Server software as well as the first version of Windows to support 64-bit x86 chips such as Advanced Micro Device's Athlon 64.
The software maker said Windows Server 2003 Service Pack 1, Windows Server 2003 for 64-bit Extended Systems and Windows XP 64-bit Edition for 64-bit Extended Systems will all be pushed back until the first half of next year. All three products, which are being developed together, were slated to ship by the end of this year.
The move is a blow to AMD, which has had its 64-bit desktop and server chips on the market for some time and has been waiting for a version of Windows that can take advantage of their capabilities. Intel has also said it will have chips that support the 64-bit extensions to the existing Pentium architecture, known as x86. There is already a version of Windows that supports Intel's 64-bit Itanium architecture, which uses an entirely separate instruction set than x86 chips.
"As is the case with all Microsoft product schedules, the development cycle is driven by quality with a focus on the needs of our customers, rather than an arbitrary date," Microsoft said in a statement.
The move is the latest in a series of delays for the 64-bit version of Windows XP.
AMD announced its first 64-bit chips in September 2003, at which time Microsoft had planned to release, by the first half of this year, a version of Windows that would support the chips.
Microsoft has also had to push back other versions of Windows. The long-awaited security update to Windows XP--Service Pack 2--has taken longer than expected and is now set to ship in August. Microsoft has also said that the next major release of Windows, code-named Longhorn, won't ship until 2006.
HP sells three Itanium-based Superdome computers to First American Title Insurance, running Windows Server 2003 Datacenter Edition.
HP's Superdomes land real-estate deal
By Stephen Shankland
Staff Writer, CNET News.com
http://news.com.com/2110-1010-5285601.html
HP has sold three Itanium-based Superdome servers to First American Title Insurance for processing title and escrow settlement transactions, the Palo Alto, Calif.-based company said on Tuesday. Terms of the deal were not disclosed. The systems replace two Unisys servers, HP said. The sale marks the first time the title insurance company has bought models based on Intel's Itanium chips.
HP's first Superdome models used its own PA-RISC processor, which meant users could only run its HP-UX version of Unix on the machines. The Itanium-based models also tap Microsoft's Windows and the open-source Linux operating systems. First American Title's machines use Windows Server 2003 Datacenter Edition, HP said. The Santa Ana, Calif.-based company, which provides services related to real estate and business, already has HP's ProLiant servers, a product line that uses Intel's Xeon and Advanced Micro Devices' Opteron processors.
Itanium based Superdomes to
AMD has reached an important milestone.
I wonder why you and the AMD FBs haven't noticed?
AMD just CROSSED UNDER Intel by 50% !
AMD - $11.40
Intel - $23.11
Don't the FBs usually go ballistic when this happens?