Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Rambus (RMBS) - Eagle or Penguin -- Ignore unavailable to you. Want to Upgrade?

To: REH who wrote (15663)	2/16/1999 9:30:00 PM
From: MileHigh	Respond to of 93625

REH, I give up. I'll just pretend everything is great! <gg> =================== Pentium III Chips Could Hit 1 GHz This Year (02/16/99, 8:16 p.m. ET) By Andy Patrizio, TechWeb Intel officials are backtracking on reports that the company can demonstrate a gigahertz chip, saying it's still some time before Intel reaches that speed. Pierre Mirjolet, architecture marketing manager at Intel Europe, Middle East and Africa, said Intel would have major fabrication plants cranking out 0.18 micron CPUs by the end of this year at speeds ranging from 600 MHz to 800 MHz. The gigahertz chips would be only samples or prototypes late this year or early next, Mirjolet said. Intel has already started manufacturing chips with the 0.18 micron die, which will be mobile chips due in June, said Mirjolet. By early 2000, Intel expects to have three of its fabrication plants producing 0.18 micron chips, with two more in place shortly afterward. The current die size for Intel Pentium II CPUs is 0.25 microns. The advantage of the smaller die size is improved performance, cooler temperatures, lower power consumption, and cheaper manufacturing. Mirjolet's comments first appeared in The Register, a British IT publication. Intel spokesman Seth Walker said the projections were premature, and that Intel has not changed plans to deliver 1 GHz chips in late 2000 or early 2001. Intel could deliver chips at up to 800 MHz by the end of the year -- with a significant bump in the bus speed for motherboards, said Walker. Even as Intel pushes the performance of its P6 core technology further, one market that isn't seeing much in terms of benefits are those who use 3-D accelerators to improve the graphics of their games and make the game run faster. A benchmark report on Ars Technica found that a new 500-MHz Pentium III chip didn't offer much better performance than a Celeron 300A overclocked to 450 MHz. Overclocking is the process of tricking the CPU into running faster than it's supposed to. The conclusion of Scott Wasson, the webmaster of the website Ars-Technica, is that a Pentium III overclocked to 560 MHz was only 7.57 percent faster than the 300A at 450 MHz in 3-D benchmarks, even though the chip has a 24.4 percent increase in clock speed and costs more than 12 times as much as the 300A. "What it shows clearly is that with the current generation of 3-D boards, you can't squeeze much more performance out of the P6 core and those boards," said Wasson. Standard benchmarks, such as 2-D windows and CPU performance, were somewhat better but not enough to justify the price difference, he said. "The question we raise is who is Intel going to sell this to, because the performance improvements they claim for this chip are in 3-D gaming, and current generation of accelerators don't show much performance improvement," said Wasson. He said the next generation of 3-D graphics chips, including the Voodoo 3 and Riva TNT 2, would show better performance scalability. Intel wouldn't comment on the benchmarks. "We've heard this speculation every time we release a new generation, and every time it comes to market and people get to see it and kick the tires, the naysayers tend to quiet down," said Walker. Intel will formally introduce the Pentium III chip Wednesday.

To: REH who wrote (15663)	2/16/1999 9:36:00 PM
From: MileHigh	Read Replies (1) \| Respond to of 93625

IDF: Merced details, encryption top Intel Developers' Forum agenda By David Lammers EE Times (02/16/99, 5:07 p.m. EDT) PALM SPRINGS, Calif. — Intel Corp. will detail its plans for security and data encryption, next-generation interconnect and the Rambus memory architecture at the Intel Developers' Forum, which convenes here Feb. 23-25. Dan Russell, director of platform marketing, said Intel expects about 1,500 engineers to attend the 110 technical sessions and 18 labs over the three-day event. The general theme is how to design systems, based on the Pentium III processor, that are optimized for multimedia and the Internet. The IA-64 Merced architecture will be more fully described, and Intel executives will detail plans for embedded applications built around both StrongARM and Intel-architecture designs. Mobile technologies will be another theme. Software developers will get information about how to write drivers and applications that will handle the streaming single-instruction, multiple-data extensions to the Intel architecture, including new instructions and changes to the Pentium hardware that support video and other multimedia data types. Intel will describe new versions of the VTune tool used by software engineers, as well as the IPeak tool used by system developers. Russell said an interconnect lab will detail the tools and design techniques needed to tackle the 100-MHz and higher buses. "The transmission-line effects in CMOS GTL (Gunning transceiver logic) are not so much frequency dependent as a matter of the design techniques that are used to handle the new challenges at 100 and 133 MHz," said Russell. Peter MacWilliams, director of platform architecture at the Intel Architecture Labs, will discuss the Rambus memory technology at a Tuesday afternoon session. Justin Rattner, director of the server architecture lab at the IAL, will discuss the course Intel plans for next-generation I/O. And yet another area of contention-encryption and security issues-will be the subject of a full-scale tutorial by David Aucsmith, senior architect at the security technology lab.