* DDJ Home

* Today's Headlines
* Past Headlines
* Microprocessor Articles
* Intel Secrets
* Intel Errata
* Undocumented Corner
* Processor Manuals
* Motherboard Manuals
* Links

Microprocessor Resources

Microprocessor
Headline News

Top Stories for July 9, 1999 (details below)
C/Net Intel's Merced chip may slip further
Electronic Buyers' News Via inks chip-set foundry deal with National, challenges Intel
C/Net How Via, National may skirt Intel restrictions
Electronic Buyers' News Intel identifies Intel 810 chipset bug
ZD Net News Man continues e-mail battle on horseback
Barred from sending e-mails to Intel employees, Ken Hamidi delivers his message on horseback.
The Register Files
The Register Optimizers not optimistic about Merced
The Register Intel's checkered background to Merced emerges
The Register Merced: the case for the defense
The Register System builders fear Intel's wrath as K7 launches
The Register Intel-NatSemi war looms over VIA, PC133
The Register Will Rambus fail?
The Register Intel cowboy talks to Intel cowboy

 

Microprocessor Headline News

Collected By Robert R. Collins

Week of July 5, 1999

Older News

July 9, 1999

Intel's Merced chip may slip further

By Stephen Shankland

July 8, 1999
C/Net

Prototypes of Merced still aren't out, and it looks like Intel's next-generation processor may slip until the third or fourth quarter of next year.

The latest lag, which could put off Merced's release by a few weeks or months, won't be as significant as the delay Intel imposed last year when it shifted the 64-bit chip's debut from the end of 1999 to mid-2000.

Still, the revised calendar may sap Intel of some psychological momentum and will likely solidify arguments that Merced, the first Intel server and workstation processor designed to compete directly against the most powerful Alpha and UltraSparc processors, will primarily serve as a test vehicle. Customers and computer companies may buy it, but could hold off on migrating to Intel's 64-bit chips until McKinley, Merced's successor, comes out in late 2001.

Via inks chip-set foundry deal with National, challenges Intel

By Mark Hachman

July 7, 1999
Electronic Buyers' News

Via Technologies Inc. today laid down a sharp challenge to Intel Corp. by signing a foundry agreement with National Semiconductor Corp. under which Via will continue to manufacture its disputed chip sets in Taiwan.

Intel has revoked Via's license to use its P6 bus technology and filed suit against Via in a San Jose federal court. Both sides now seem prepared to let the courts decide the issue, which could evolve into a legal dispute with no real impact on product shipments. 

According to a statement jointly issued by National and Via, "This alliance is for Via to continue to develop and market Via's chip set which offers the advantages of a 133-megahertz front-side bus, PC133 SDRAM, and AGP 4X compatibility. National will be manufacturing the chip set for sale and distribution by Via Technologies." Via chip sets bearing National's logo have already begun appearing in Taiwan, according to reports.

How Via, National may skirt Intel restrictions

By Michael Kanellos

July 7, 1999
C/Net

Via Technologies has discovered the secret ingredient that will allow the company to produce chipsets, and likely processors, that will compete directly against products from Intel: legal loopholes.

Via and National Semiconductor today announced that the two companies have signed an agreement under which National will manufacture chipsets designed by Via, including the controversial chipsets that will support faster 133-MHz memory and contain a 133-MHz system bus. Via sources said the chipsets will contain the logo of both companies.

Intel identifies Intel 810 chipset bug

By Mark Hachman

July 8, 1999
Electronic Buyers' News

Intel Corp. has identified a bug in its Intel 810 or "Whitney" chipset.

The bug, which involves an error with the real-time clock (RTC), occurs when "the RTC may inaccurately report that the clock is not busy when it is actually in the process of updating internal registers," according to a statement posted by an Intel support engineer on an Intel 810 hardware support forum on Intel's Web site. "This may result in invalid data being read in the date and time fields of the real-time clock, but the erratum does not modify the actual RTC values."

Man continues e-mail battle on horseback
Barred from sending e-mails to Intel employees, Ken Hamidi delivers his message on horseback.

By Reuters

July 7, 1999
ZD Net News

Barred by a California court last month from sending e-mail to employees at his former employer, an ex-Intel Corp. engineer showed up at Intel's headquarters on horseback Tuesday to deliver his latest communique.

Ken Hamidi, who worked for Intel (Nasdaq:INTC) for several years as a quality engineer before being fired, rode up to Intel's Robert Noyce Building on horseback in western attire and handed a floppy disk to an Intel representative.

"He indicated that he intended to appeal," said Intel spokesman Chuck Mulloy. "There is nothing new in his position." The legal battle, however, may determine how far the First Amendment's guarantee of free speech extends into cyberspace.

The Register Files

Optimizers not optimistic about Merced

By Mike Magee

July 7, 1999
The Register

A communication from a reader has pointed out that one big problem with Intel's up and coming Merced processor is optimising the compilers.

The reader, who preferred not to be named, said that optimising simple programs with small data sets is easy, but for real world applications, including server programs, it is almost impossible.

That is because the variation according to the type of data used is too extreme.

Intel's checkered background to Merced emerges

By Mike Magee

July 7, 1999
The Register

A reader of The Register has given us the background on Intel failures of the past.

He was responding to an earlier story today, where a chip architect was wondering how good the IA-64 Merced platform is.

The correspondent detailed a story about the past of Intel processors which also did not work too well.

He said: "Intel did well with the 8080/8085. They started to get wiped out by the Z80, and got the 8088/8086 out the door just in time. In parallel, they did OK for a while with the 8048 and 8051, but pulled out of that business really early, while other were making good money from 8051 derivatives.

Merced: the case for the defense

By Mike Magee

July 8, 1999
The Register

A reader has replied to a story we wrote yesterday that claimed Merced compilers were in bad shape.

Here are his comments: "I agree that Merced will only execute instructions in order. Load instructions just queue the request to memory. They do not wait until the result is available. If an instruction which requires that data executes before it is available then that instruction will stall and all the instructions behind it will stall.

"Now this sounds very bad but it is not in practice. Firstly all compilers will always put as much space as possible between a load and its use. Secondly in an Out-Of-Order(OOO) execution processor if an instruction stalls then other instructions can issue BUT this in practice is limited.

System builders fear Intel's wrath as K7 launches

By Linda Harrison

July 8, 1999
The Register

The AMD K7 chip, aka Athlon, looks set to enter the market with a bang this August, though many PC builders fear Intel's wrath should they choose to use it.

Among those launching Athlonian PCs are Compaq and Bolton-based reseller Dabs Direct. Dell, Viglen and Tiny Computers all told The Register they will stay 100 per cent loyal to Intel.

Others who are planning to adopt the K7 chip, but asked not to be named for fear of the great Satan’s reaction, expected Athlon to be a lower-cost alternative to the Pentium III.

Intel-NatSemi war looms over VIA, PC133

By John Lettice

July 8, 1999
The Register

VIA Technologies and NatSemi have thrown the gauntlet down to Intel with the announcement of a deal designed to allow VIA to produce PC133 chipsets. VIA has already agreed to buy NatSemi's Cyrix unit, in a move generally interpreted as an attempt to escape from Intel's licensing straightjacket, and as Intel has already gone on the record with the view that the deal won't give VIA a valid Intel license, it's now clearly war.

According to a VIA-NatSemi statement issued today, the two companies "strategic agreement" caters for NatSemi doing the foundry work for VIA chipset in Pentium II-type computers. "This alliance is for VIA to continue to develop and market VIA's chipset which offers the advantage of a 133MHz front-side bus, PC133 SDRAM, and AGP4X compatibility."

Will Rambus fail?

By Mike Magee

July 8, 1999
The Register

We at The Register have had an advance peek at some of the presentation material slated for delivery at the Platform 99 Technology Conference to be held in San Jose on July 21st and 22nd.

One of the analysts and a couple other presenters at the conference will be addressing critical DRAM related issues.

Analysts from MDR and other industry figures will thrash it out, we're given to understand.

Intel cowboy talks to Intel cowboy

By Mike Magee

July 7, 1999
The Register

Ken Hamidi, the scourge of Intel's email system and the organiser of Face Intel, took matters a step or four further on Monday when he rode a horse up to Chipzilla Central in Satan Clara to deliver mail by pony express.

Hamidi, who has campaigned against Intel over its employment practices for years, lost a case against the chip giant in spring, but is taking his case to the supreme court.

Although there is a cartoon of Hamidi riding up to the Intel OK Corrall on his web site, we haven't seen a pic of him astride his bucking bronco yet.

July 7, 1999

Firm Says Intel Rigged Panel On Standards

By Brenda Sandburg

July 6, 1999
The Recorder/Cal Law

Intel Corp. is in the antitrust hot seat once again.

Three months after negotiating an agreement with the Federal Trade Commission over antitrust charges, the Santa Clara-based chip maker is accused of wielding improper influence on a standards setting body to block a would-be competitor, MultiVideo Labs Inc.

But antitrust lawyers and scholars say the suit could be a tough one to prove. It hinges on the highly technical discussions among members of a standards setting body, and MultiVideo will have to show that Intel was able to sway members to reject MultiVideo's product as a standard.

Intel hits roadblocks on eight-way server push

By Stephen Shankland

July 2, 1999
C/Net

With the arrival of the eight-processor servers based on mass-produced Intel technology, computer makers and Intel have been stepping on each other's toes in a market that isn't yet that lucrative.

As Intel carries its technology closer to the core of giant corporate networks, it's treading on computer design territory that manufacturers consider to be home turf. And Intel's one-size-fits-all approach doesn't sit well with some of those computer manufacturers, who are trying to make their eight-way servers different from their competitors' offerings, instead of indistinguishable boxes, as in the case of desktop computers.

AMD's Next Round With Intel
Amid K6 sales gloom, the super-fast Athlon chip gives it a fighting chance to make points in PC market

By Tom Quinlan

July 4, 1999
San Jose Mercury News

The clouds that seem to hover perpetually over 1 AMD Place in Sunnyvale have never seemed darker.

Last month Advanced Micro Devices Inc. announced that it expected to report an operating loss of more than $200 million for the quarter that ended in June -- largely because it was unable to sell about 40 percent of the microprocessor chips it made during that three-month span. A number of manufacturers had essentially replaced the AMD K6 line with Intel Corp.'s Celeron processors.

And even as the company starts shipping its super-fast Athlon processor, previously known as the K7, personal computers that use AMD's revolutionary design won't start showing up until August at the earliest. It's highly unlikely, AMD officials acknowledge, that the company will become profitable this year.

About Suffered Victories and Achieved Defeats

By Andreas Stiller

Volume 13, 1999
c't Magazine

Many enemies, many honors, this is Intel's motto for fighting at many fronts. Even though the enterprise is almighty it here and there still 'achieves' a strategic defeat.

This happened for example in the legal battle with Intergraph. The workstation manufacturer owns the patents for a processor named Clipper developed by Fairchild years ago, and Intel supposedly used Clipper patents in the Pentium and Pentium II. Even thought the company is depending on Intel Intergraph did not want to hand over the patents to the business partner without anything in exchange. So the chip baron angrily excluded the rebelling 'peasant' from the complete beta program and did not give him any prototypes or documentation.

National in no hurry to sell Maine fab; TSMC confirms interest

By Mark Hachman

July 1, 1999
Electronic Buyers' News

Unlike the sale of National Semiconductor Corp.'s Cyrix division to Via Technologies Inc. this week, the sale of a majority stake of National's fab in South Portland, Maine is under no fixed timetable, according to the company.

"We don't export to report anything for many weeks yet," said a spokesman for National, Santa Clara, Calif.

Among those rumored to be looking at the facility, Jim Kupec, president of United Microelectronics Corp. (UMC) Group USA in Sunnyvale, Calif., said his firm was not interested in purchasing National's fab.

Intel gadfly at it again
He will deliver message in person to Santa Clara office today

By Jonathan Rabinovitz

July 6, 1999
San Jose Mercury News

Seated in front of his home computer on a recent afternoon, Ken Hamidi needed only a few seconds to figure out the e-mail addresses of several top Intel Corp. executives.

Only 10 months earlier, from that same keyboard in his rented ranch house outside Sacramento, Hamidi went through the same step for some 30,000 Intel employees and sent each a message outlining how he believes the company abuses its workers.

That mass mailing placed Hamidi at the center of a legal battle that could determine how far the First Amendment's guarantee of free speech extends into cyberspace.

Analysts Upgrade Intel

By Warren S. Hersch

July 6, 1999
Computer Reseller News

Several Wall Street companies upgraded their recommendations on Intel shares on Tuesday.

Intel gained 3/8 to close at 63 13/16 after Merrill Lynch and Prudential Securities raised their recommendations on the stock. Intel was the most heavily traded stock on the Nasdaq, which was down 4.6 points to 2,736.5.

Trading among technology stocks was mixed following the July 4 weekend.

The Dow dipped 4.1 points to 11,135.1. The S&P 500, too, edged down 3.1 points to 1,388.1.

The Register Files

Rise processor has FP problem

By Mike Magee

July 6, 1999
The Register

Reliable sources said today that there is a floating point problem with the Rise 266MHz processor. 

According to the sources, the glitch is similar to the FDIV bug which plagued Intel's Pentium chip four years back. 

With some calculations, the source said, the processor will throw up erroneous results.

Doubts emerge over Intel chip tracking

By Mike Magee

July 5, 1999
The Register

The massive haul of Intel Pentiums that the Chinese (HK) government caught last week should be trackable by Intel itself, it has emerged.

Earlier this year, Intel admitted that processor serial numbers (PSNs) were embedded in Pentium IIs and Celerons as well.

And earlier today, facts and figures emerged from Hong Kong about the level of Intel counterfeiting going on. Making it important.

Europe hit by re-marked Intel chips

By Mike Magee

July 5, 1999
The Register

A report on NEWS.COM over the weekend suggests that the problem of re-marked CPUs in Europe this year was greater than earlier anticipated.

The news wire reported that the local authorities seized £500,000 of re-marked Intel microprocessors last week.

Crooks have re-marked Pentium IIs to look like Pentium IIIs, the report suggested. It added that over 3,000 CPUs seized were intended for export to Europe.

Intel pulls plugs on plans for Kahneeta PC

By John Lettice

July 5, 1999
The Register

Want to know about Intel's Kahneeta reference platform? Well you're too late, Intel seems to have pulled the plugs on it well in advance of most of the world having heard of it.

If you look here, you'll see that Kahneeta is no more, "due to limited supply," as Intel enigmatically reveals. Other references to what it is/was are scarce, but we've managed to track its appearance in a Microsoft OnNow for Windows NT 5.0 (alias Win2k) late last year.

Kahneeta was one of the key desktop platforms Microsoft was aiming NT 5.0 OnNow features at, along with "Intel Atlanta based motherboards," and Compaq and Dell machines. And just to show you how anally-retentive old Chipzilla is, a search of intel.com for Atlanta and motherboard throws up a page in the support section that says: "The phrase Atlanta is not an Intel product name. The correct product name is Intel AL440LX motherboard." So there.

Curious synergy exists between Intel and AMD Athlon Web sites

By Mike Magee

July 4, 1999
The Register

We regularly receive notifications here of new Intel domains that come into being. 

Imagine our surprise, then, when we heard of a new one called Intel-Store, seemingly owned by the Intel Corporation, which looks startlingly like another site called AMD Athlon we noticed two weeks ago.

Indeed, the resemblance is so startling that alarum bells ought to be ringing somewhere. Both are Network Solutions registrations.

Cyrix-Via shrinks socket seven die on M2-400

By Mike Magee

July 4, 1999
The Register

Our friend Daiki has alerted us to the fact that Cyrix has released the M2-400 processor in Japan.

But, he points out, although it seems to have a smaller die size than the previous M2, it has 321 pins rather than 296 pins.

The M2-400 could use a .18 micron rather than a .25 micron process, but, as Daiki points out, we don't know because Cyrix-Via hasn't told us...

Cyrix has real chance now...

By Mike Magee

July 6, 1999
The Register

An industry analyst said today that he expects good things from Cyrix, provided it can keep to its existing roadmap.

Joe D'Elia, senior microprocessor analyst at Dataquest Europe, said that he thought the combination of chipset company Via and Cyrix would produce good technology.

He said that Cyrix had somewhat languished under the National Semiconductor stewardship, but that he thought that would change.

AMD tilts to PC133 - 1Ghz Athlon, SMP by 2000

By John Lettice

July 6, 1999
The Register

AMD may be poised to jump onto the PC133 bandwagon, company exec Drew Prairie seems to hint in an interview last week. Speaking to CPUReview editor William Henning Prairie commented: "DDR SDRAM seems to make more sense than RAMBUS."

That puts flesh on our earlier scoop: K7 to get PC-133 support from Via

Although Prairie made no firm commitment on AMD's behalf, and Henning stresses that this is his personal view rather than a company policy, it does suggest that the company may go with support for the Via-Cyrix camp instead.

PC133 delivers 'awesome stability'

By Drew Cullen

July 4, 1999
The Register

BX Boards has turned its screwdrivers on PC133 memory.

With its usual thoroughness, the British-based hardware review site puts modules from Samsung, Enhanced Memory Systems, Crucial and Corsair through their paces.

And the conclusions? Don’t throw away PC100, just for the sake of it. But if you’re in the market for new memory, buy PC133. This offers future proofing and "awesome stability" Finally, reviewer Andy Drake advises: "shop around and go for whatever is cheapest!"

Intel turns Pentium III/600 price screw on AMD K7

By Mike Magee

July 5, 1999
The Register

c't magazine in Germany is reporting that Intel will ratchet its price screw on AMD when it releases its 600MHz Pentium III part later this month.

Last week, we reported that Intel will also cut prices on its mainstream Pentium III/500 parts to compete aggressively with AMD's soon-to-be-released K7 Athlon chips.

Intel had hoped to release its 600MHz Pentium III with 133MHz front side bus (FSB) in September, but problems with the i820 chipset caused that idea to burn to flames. Instead, a few weeks back, it announced it would have a .25 micron 600MHz part out in July.

AMD stands firm on Athlon K7 pricing

By Mike Magee

July 5, 1999
The Register

A senior executive at AMD Europe said today that speculation Intel is cutting its prices would not alter its position its Athlon K7 pricing.

Meanwhile, an industry observer said he doubted whether Intel would take such steps pro-actively and wondered whether the company was managing to build more Pentium IIIs faster.

Robert Stead, European marketing director at AMD, said: "Our position is we will offer a good product at a fair price. A fair price is determined by infrastructure costs and the competitive position."

Advertisement
Copyright © 2009 Dr. Dobb's Journal