|
November 6,
2001
|
|
By Michael Kanellos and
Rachel Konrad
November 5, 2001
C/Net |
Toshiba had big plans earlier this year for
the U.S. release of a laptop with an energy-efficient chip
from Transmeta, a product that would have represented a coup
for the upstart chipmaker in its rivalry with mighty Intel.
But in the latest setback for Transmeta, Toshiba quietly
pulled the plug on the notebook this summer, partly because of
delays surrounding Transmeta's latest chip, executives told
CNET News.com. |
|
By
Rick Merritt
November 5, 2001
EE Times |
The race to define a post-PCI
interconnect for communications systems will heat up Monday
(Nov. 5) as two fabless design companies announce competing
switch chips. API Networks Inc.'s four-port switch is designed
for HyperTransport, while Tundra Semiconductor's six-port
switch and supporting silicon target RapidIO.
Both companies hope to grab sockets in communications
systems that need a fast standard interconnect beyond PCI. And
both ultimately will face competition from 3GIO, a PCI
replacement backed by Intel Corp. but not expected to see
silicon until late 2003 or early 2004.
|
|
Truths...from the rumor mill |
|
By Mike Magee
November 5, 2001
The Inquirer |
INTEL HAS NEVER MADE A BIG song and dance about boxed products
for the server marketplace - this is because of old the big PC
makers found high margin and good business delivering big
machines with Xeon processors with heaps of cache. But the
times they are a changing, and in part this is due to Advanced
Micro Devices and Via, who wish to step into those lucrative
blue suede shoes that only La Intella was able to wear in days
of yore. |
|
By Tony Smith
November 5, 2001
The Register |
SIS is still in the running to offer a Rambus chipset for the
Pentium 4 despite the denials company executives issued last
week. Industry sources cited by EBN claim that the company
has offered to develop a chipset that would support Rambus'
quad-channel RDRAM - 4i Direct DRAM - after Intel decided that
it wasn't going to produce one of its own. |
|
By Mike Magee
November 5, 2001
The Inquirer |
THE INTRODUCTION OF THE 1.6GHz Athlon Palomino (1900+) today
shows how tough the price war is but appears to demonstrate
that Intel has lost the battle and tacitly accepts AMD's PR
rating. That's because the price of the AMD 1.6GHz (1900+)
is equivalent to the price of the 1.9GHz Intel Pentium 4, for
the time being at least. Intel took 27 per cent off the price
of its 1.9GHz (1900MHz) chip just a week ago, obviously
anticipating a swift AMD riposte. |
|
By Mike Magee
November 5, 2001
The Inquirer |
CHIP GIANT INTELLA appears to be so confident of its ability
to produce low power Pentium 4 mobile processors that it will
release boxed versions as early as the end of the first
quarter next year. That is the shape of the boxed mobile
roadmap Intel is presenting to its distributors, which has a
1.70GHz Pentium 4 using a .13 micron process appearing in the
"professional" category, with other flavours such as a
1.60GHz, 1.50GHz and a 1.40GHz mobile processor occupying
other mainstream categories in its spring lineup next year. |
|
By Mike Magee
November 5, 2001
The Inquirer |
IT ONLY SEEMS A MATTER OF weeks since we last saw an Intel
desktop roadmap and sure enough as the company accelerates its
Pentium 4 map we're off again. In fact, it is only a matter
of weeks. Intel has accelerated its plans to continue to
compete with AMD on both pricing and speed during the course
of 2001.
It now seems pretty certain that Intel will reserve its
"Northwood" .13 micron die shrink until Q1 of next year, with
the first processor out of its starting gate being a 2.2GHz
Pentium 4 complete with 512K of cache and the 400MHz front
side bus we all know. |
|
By Tony Smith
November 5, 2001
The Register |
Intel will begin to ship the integrated-graphics version of
its 845 chipset - codenamed Brookdale G - next February, ahead
of its official introduction in April and considerably in
advance of its previously timetabled launch in Q3 2002.
According to a report in DigiTimes, the chip giant has told
mobo makers that it has had to bring the schedule so far
forward to allow it to compete more effectively with rival
Pentium 4 chipsets from SIS, Acer Labs and, yes, VIA too. |
|
November 5,
2001
|
|
By
Jack Robertson
November 2, 2001
EBN |
Taiwan's Silicon Integrated Systems Inc. (SiS) is
preparing to launch a chipset next year to support Rambus
Inc.'s 4i four-bank Direct Rambus DRAM, following a decision
by Intel Corp. to table a similarly-positioned chipset,
code-named Tulloch, according to industry sources. The 4i
RDRAM provides similar performance to mainstream 32-bank
Rambus DRAM but at a lower cost, because it can be
accommodated on a smaller die given that designers need to
incorporate less logic ciruitry to control the four memory
banks. |
|
By Mike Magee
November 2, 2001
The Inquirer |
TAIWANESE
WIRE Digitimes reports today that SiS will not
introduce a chipset supporting Rambus memory types. That
follows reports on Cnet that the company was
contemplating such a move.
But, the wire claims, SiS will go to DDR 400 during next
year, although there is no specific date for such a release. |
|
By
Jack Robertson
November 2, 2001
EBN |
At the very time Intel Corp. is readying support for
double-data-rate SDRAM for its PC processors, the company has
bypassed SDRAM in favor of Direct Rambus DRAM for its
next-generation network processor. Though DDR and RDRAM will
compete for sockets in Intel's mainstream PC platform, the
Rambus architecture is proving particularly suited for the
network processor the company plans to unveil in the first
half of 2002. |
|
November 1, 2001
Semiconductor Business News |
Taiwan's Via Technologies Inc.
today announced a new version of its Pentium 4-compliant chip
set line, this time with integrated graphics and modular
features. The chip set, called the Via ProSavageDDR P4M266,
supports PCs based on Intel Corp.'s Pentium 4 processor line.
The device supports DDR (double-data-rate) SDRAM memory, AGP
8X, and 2D/3D graphics, based on S3 Graphics' ProSavage8 core.
It also handles Ethernet-based protocols and Ultra ATA-133. |
|
By
Stephen Shankland
November 1, 2001
C/Net |
Intel will
release new chips at the Comdex trade show, its first
low-power designs for super-thin servers squeezed into
cabinets by the dozens, a source familiar with the plan said.
The new Pentium III model is a
gussied-up chip taken from the company's product line for
portable computers, which share many of the same constraints
as "ultradense"
servers. These systems can't consume as much power or give off
as much heat as ordinary CPUs because overheating causes
processing errors. |
|
Truths...from the rumor mill |
|
By an INQUIRER staffer
November 2, 2001
The Inquirer |
BECOMING
A MEMBER of Intel's system builder programmes means you have
to attend two training sessions a year. Here are the
"confessions" of a system integrator who has just been through
La Intella's mill. His identity is being kept secret on the
grounds of no name, no pack drill.
"I WENT to an Intel "training session" about a month ago. |
|
By Mike Magee
November 2, 2001
The Inquirer |
SOURCES
SAID Sun Microsystems, in its bid to cut costs, is
contemplating slashing its efforts to support the X86 platform
in any way. The sources said Sun wants to stop shipping
software on X86 platforms and Solaris 9 for IA-32 will remain
a pipe dream. And IA-64 is even more of a pipe dream.
Further, the sources said, other X86 products such as
Solaris 8 and earlier revs, will not get patched or updated. |
|
By Mike Magee
November 1, 2001
The Inquirer |
YESTERDAY
WE REPORTED that there appears to be a shortage of Itanium
733MHz and 800MHz processors and we're still attempting to
discover exactly what the problem is.
But sources tell us - and so far we certainly cannot
confirm them - that there could be a serious problem with the
die itself, an erratum, which is currently being investigated
by both Intel and a number of third parties. |
|
By Mike Magee
November 4, 2001
The Inquirer |
THANKS TO
THE LINUX COMMUNITY, hyperthreading will be available for the
Pentium 4, as the patch
here quite clearly shows. But hang on a cotton-dogging
minute. Jacksonville, SMT or whatever Intel's calling it right
now, isn't going to be implemented for the Pentium 4
(Willamette) core, is it?
At the Developer Forum last, Intel's Paul Otellini appeared
to be saying this kind of technology will be implemented in
Xe on
servers next year. |