SI
SI
discoversearch

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.
Strategies & Market Trends : Technical analysis for shorts & longs -- Ignore unavailable to you. Want to Upgrade?


To: j g cordes who wrote (17335)7/28/1998 3:18:00 PM
From: Johnny Canuck  Read Replies (1) | Respond to of 68172
 
Jim,

PMCS has 70 percent of the market for ATM chips for the
physical layer. This applies to LAN and WAN. ASND, NN
and pretty well anyone else who does ATM buys from them. They
also do ethernet chips. They hold quite a few ATM
related patents. VTSS is their biggest competitor.

They recently fell after beating estimates because
they are going to increase their R&D expenses over
the next few Q's. This will hurt margins.

It is worth looking at. They are based here in
Vancouver. They also have an office in California.
Unfortunately, I don't know anyone that works there.

Harry



To: j g cordes who wrote (17335)7/28/1998 8:29:00 PM
From: Johnny Canuck  Respond to of 68172
 
Jim,

It looks like ARMHY will have some jump in their stride.

********************************

Intel dives into new chip line
By Michael Kanellos
Staff Writer, CNET NEWS.COM
July 28, 1998, 2:00 p.m. PT

update After months of speculation, Intel is moving
full speed toward selling and developing processors
for consumer-oriented electronic devices based on
the StrongARM chip, an architecture that's
completely different from its long-established PC
chip standard.

QUOTE SNAPSHOT
July 28, 1998, 1:01 p.m. PT
Intel Corp. INTC
85.6250
-0.4375
-0.51%
> more from CNET
Investor
> Investor message
boards
Quotes delayed 20+ minutes

Even as it launches plans to more aggressively
market the current line of StrongARM
microprocessors, Intel is working on a succeeding
generation of chips that will be twice as powerful,
according to sources at the chipmaker.

The chipmaker's efforts to develop a StrongARM
line code-named SA2 strongly indicate the
company has big plans for new markets. Intel
acquired the StrongARM chip design from Digital
Equipment last year, but has yet to move on it.

"We're actively engaged with customers in the
design phase [of products]," said Mark Casey,
marketing director for StrongARM chips at Intel.

The faster SA2 chips will likely be released, or at
least announced, in 1999, with mass manufacturing
kicking off in 2000. The company already has two
separate design teams working on new versions of
StrongARM.

"The days of wishy-washiness are over," said Dean
McCarron. "They [Intel] are investing in multiple
design teams and multiple designs."

Intel's StrongARM moves will likely change the
competitive landscape for processors in the
"embedded" market, which comprises the low-cost
chips typically found in TV set-top computers,
intelligent devices such as Windows CE-based
handheld computers, cell phones, and other items.
Hitachi has been the dominant player with its SH
series of chips, while Silicon Graphics spin-off
MIPS is a strong contender.

Intel's Casey said the company's StrongARM chip
will be targeted at TV "set-top computers" that are
not tied to the Windows-Intel PC architecture. On
the other hand, "set-top PCs" would use a version
of the Pentium II processor and come with the
capability to use PC software, he said.

The StrongARM chips would be aimed at set-tops
priced below $400, while Pentium II chips would
go into PC-TV devices below the $1,000 mark, he
said.

Until now, Intel's x86 processors--the basis of the
Pentium and Pentium IIs chips found in up to 90
percent of all PCs--have been deemed too power
hungry and expensive for most consumer
electronics markets.

"We have fully embraced the StrongARM roadmap
and we are making investment to enhance our
design and marketing capabilities," said Ron Smith,
vice president and general manager of Intel's
computing enhancement group.

Although praised by both hardware designers and
analysts, StrongARM seemed to have disappeared
from the scene after Digital gave the chip's design
to Intel as part of a massive legal settlement last fall.
Its relative silence on the subject of StrongARM
was largely interpreted as a reluctance to promote
a chip it didn't originate. An Intel spokesperson
said that the company could not speak because it
was in a quiet period--a span of time when it is
prohibited from making public statements.

StrongARM chips are based around a design by
Digital, which in turn is based around a design from
Advanced RISC Machines.

As time progressed, however, Intel apparently
found itself with a choice between getting in the
market now or waiting until later. Now won out.

"This is a segment where Intel has no offering. If
Intel does not play in that space, they leave a
massive amount of sales to someone else,"
commented Greg Blatnik, vice president at Zona
Research, who confirmed Hitachi and MIPS are
the field's principals. "The market is booming for
these chips."

"This [StrongARM] was a well-respected product.
They were handed something that was ultimately
everything you dreamed about for an embedded
processor," added McCarron. "The alternative was
to fall victim to the 'Not Invented Here' syndrome.
They made a business decision, which they are
good at."

Intel's StrongARM effort began with the
manufacture of two StrongARM chips, the SA 110
and the SA 1100, under the Digital name.
Manufacturing takes place at the Hudson,
Massachusetts, plant it acquired from Digital in its
$700 million settlement, according to Jim Turley, an
analyst with MicroDesign Resources.

The SA 110 and 1100 sell for between $27 and
$40 and run as fast as 233 MHz. Most Pentium-
and Pentium II-class chips are priced in the
hundreds.

Samples of a third chip, the SA 1500, were due in
the first half of this year, he pointed out. Intel has
yet to make a formal announcement of support for
the SA 1500, but the company has said it will
support the existing roadmap.

The next big stage will come with the SA2
generation. Although the company has been vague
on exact details, both Blatnik and McCarron said
that the chip will double the performance of current
StrongARM chips.

"It's a substantial alteration of the architecture to
boost performance," said McCarron.

Intel is an investor in CNET: The Computer
Network.



To: j g cordes who wrote (17335)8/4/1998 3:58:00 AM
From: Johnny Canuck  Respond to of 68172
 
>>What's with PMCS

Silicon Babylon: Dot.coms Are Dead.
But Long Live the Net.

By Cory Johnson
West Coast Bureau Chief
7/27/98 10:49 AM ET

SAN FRANCISCO -- Wall Street's cliche of the moment
seems to be "the sky-high valuation of Internet stocks." And
rooted in this (and every) cliche is a truth: Net stocks are
valued so richly that even true believers are hard-pressed to
make smart investments. Seasoned investors worry about
the S&P 500 trading at a less-than-conservative 1.8 times
revenue. So what's an investor think when Amazon.com
(AMZN:Nasdaq) -- a company whose long-term success is
far from a lead-pipe cinch -- is trading at 22 times revenue,
and Yahoo! (YHOO:Nasdaq) at 78 times revenue? Even
CNet (CNWK:Nasdaq), a company that reported -- gasp! -- a
profit this week, is trading at 22.3 times revenue.

No wonder Net stock investors are airsick. So despite big
earnings news, most Net stocks were flat for the week. And
traders and fund managers out here on the coast -- some of
the early cheerleaders for Net stocks -- are now looking at
the new Net stocks.

"Now you're singing my tune," said one West Coast trader
who asked not to be named. "All of my clients are looking
for some other way to play the Internet, because dot-coms
are just too scary. They're screaming short, so the 'other
Net' has been the sales pitch of the week."

Kevin Landis, of the $193 million FirstHand Technology
Value fund, is another fund manager taking a different angle.
"Amazon.com might succeed or fail, Lycos.com might clean
the clocks of Yahoo.com," says Landis. "These guys give
you a reason to use the Internet -- but good luck picking the
winners. But I'll tell you what: If there were a stock called
Internet Traffic Inc., everyone on the Street would be buying
it."

So Landis looks at companies that carry Internet traffic,
particularly advanced telecommunications companies like
Level 3 (LVLT:Nasdaq) and Qwest (QWST:Nasdaq), two
companies with more reasonable price-to-revenue ratios
(17.3 in the case of Qwest). "This is the plumbing of the
Net," says Landis. "People who own and resell bandwidth.
They're building out their networks to keep up with Internet
growth."

OK, but what about the huge expense of building those
networks as fast as the Net? Capital expenditures like that
can gobble up profits. No worries: That's when Landis digs
deeper.

"If you could listen in on the calls of a Qwest purchasing
agent," say Landis (making like Linda Tripp), "who would he
be talking to? It would be equipment makers: Nortel
(NT:NYSE), Ciena (CIEN:Nasdaq), Cisco (CSCO:Nasdaq),
Lucent (LU:NYSE) -- and nobody ever got fired for having
Cisco and Lucent in his portfolio." Indeed. Last week both
Cisco and Lucent saw outsized volume while dot-com
stocks cooled off. Adding fuel to that fire was a story in The
Wall Street Journal on Cisco and blowout earnings from
Lucent. These two companies alone brought in $1.4 billion in
profit on revenue of more than $37 billion in the last year.

OK, but nobody ever beat the Street by owning what the rest
of the Street owns. That's why Landis digs even deeper.

"I like to go one more layer down," says Landis. "Open up
the gear from Nortel, Ciena, Cisco and Lucent -- and what do
you see? You see chips from PMC Sierra (PMCS:Nasdaq),
Applied Micro Circuits (AMCC:Nasdaq), Vitesse
(VTSS:Nasdaq), Transwitch (TXCC:Nasdaq) -- now you're
singing my song."

Chips stocks, of course, have been crushed across the
board by the economic downturn in Asia (that's another Wall
Street cliche for bank runs, armed uprisings, developing
nation streets choked by smoke from burning tires...). But
there's a big difference between low-tech dynamic
random-access memory chips and the super high-end chips
from PMC Sierra. "Every time some pundit comes out and
says DRAM pricing is still really weak, all the chip stocks
fall," says Landis. "Traders on the Street don't know what's
in their portfolios, so they just sell. And that's when I buy."

Landis hasn't quite bet the farm on these chip companies,
but more than 10% of his fund is in these names, and PMC
Sierra is now his single largest holding. Landis says he's
tempted to buy more. "It's hard to double down on your
largest position," says Landis, "but I think these guys are as
close to bulletproof as a chip company can be."

Investors big and small will miss the boat if they try to avoid
Internet stocks altogether. "You've got to own them," Steve
Ross told TSC earlier this week. Ross manages a $4 billion
portfolio at Nicholas Applegate Capital Management and
his firm holds more than 28.8 million shares of PMC Sierra.

The more you believe that the Net is going to keep
exploding, says Landis, "the more those issues come to the
fore. You've just got to dig a little deeper than the average
investor."

c 1998 TheStreet.com, All Rights Reserved.

TOP | ABOUT US | CONTENTS | SUBSCRIBE | ADVERTISE | TRADE ONLINE |
FEEDBACK | SEARCH | HELP



To: j g cordes who wrote (17335)8/4/1998 4:02:00 AM
From: Johnny Canuck  Respond to of 68172
 
Networker Xylan Gets Key Contract With Nortel

Date: 8/4/98
Author: Michele Hostetler

Xylan Corp. is charging deeper into networking by winning a deal to make gear for Canadian
communications titan Northern Telecom Ltd., the companies plan to announce Tuesday.

The agreement is expected to help Xylan grab more sales to telecom firms and network service
providers, and aid in fending off its largest rival, Cisco Systems Inc., says Douglas Hill, Xylan's
co-founder and marketing vice president. The deal could net Xylan ''a couple of million dollars'' in
sales by the end of this quarter alone, Hill says.

Xylan could use another big partner, says Melinda LeBaron, analyst with Stamford, Conn.-based
Gartner Group Inc. Xylan already makes switches for IBM Corp. and Alcatel Alsthom.

''It will be difficult for Xylan to grow significantly for the next five years without another partner,''
she said.

Calabasas, Calif.-based Xylan is battling with Cisco over the fast-growing networking switch
market. Switches speed up networks by adding an extra lane on which information can travel.
Xylan's flagship product is called the OmniSwitch.

The company must move fast this year because the industry is in transition. Voice experts Lucent
Technologies Inc. and Nortel want a piece of data networks. As a midsize company, Xylan is in
danger of being gobbled up by bigger players or getting outmaneuvered by smaller ones.

''We're getting ready for the battle of the behemoths,'' LeBaron said.

Xylan won't be one of those giants duking it out, but it is getting bigger. The company's '97 sales
jumped to $210 million from $29 million in '95. This year, the company could post nearly $350
million
in revenue, according to Van Kasper & Co.

The Xylan agreement might come as a surprise, because Nortel plans to complete its purchase
of
Bay Networks Inc. in September, a stock deal worth roughly $7 billion. Xylan has a different
piece
of the puzzle, though. Its equipment is stronger in carrier requirements, says the company's Hill.

''Bay simply doesn't have specific areas of technology that, in this case, we do,'' Hill said.

Xylan's switches will be used with Nortel's version of digital subscriber line, or DSL, technology.
Phone carriers are the prime customers. Nortel has received $1 billion in orders for one-megabit
modem service, Hill says.

Nortel will be a key piece of Xylan's strategy to gain more business from telecom firms, Hill says.
Telecom sales could reach 20% of revenue in the next year, he says.

''Nortel is a tremendous way for us to get our products out there,'' Hill said. ''It gives us a lot of
credibility. It's a strong stamp of approval.''

Xylan's fortunes also will grow along with the popularity of one-megabit modems or DSL, he
says.

Nortel will give Xylan more stability. Xylan won't be so reliant on the two core partners it has.
Wall
Street has been worried about Xylan's vulnerability on that front. Xylan shares dropped to $24.69
on
July 31 from $31.06 on June 20.

During a call with financial analysts on July 20, Xylan executives said IBM had six to seven
weeks
of inventory. Xylan is holding back products to reach the inventory goal of three to four weeks,
Hill
says.

Unlike its larger competitors, Xylan focuses on just the switch market. This approach has made
the
company a strong force in tech, says Scott Heritage, analyst with Warburg Dillon Read in New
York.

''They can beat Bay, 3Com and Cabletron,'' Heritage said. ''Xylan can do that because they have
better products than those other companies.''

But Xylan will have a hard time prying customers away from Cisco. Still, the quality of Xylan's
products gives it an edge, analysts say. They add Xylan is able to react more quickly to customer
wants than slower-moving larger players.

But Xylan's strength also is its weakness. In the topsy-turvy networking world, customers want to
know their equipment maker is going to be around in a few years.

''They don't have the size of Cisco, Bay, 3Com and Cabletron,'' Heritage said. ''They don't have
the
reach necessarily that those other companies have.''

But Xylan is poised for a breakout, says Marc Woodward, a Van Kaspar analyst.

''Xylan is starting to get out of the start-up stage,'' he said. ''I think they're starting to gain critical
mass.''

Xylan's gigabit technology for company networks, called X-Frame, is expected to give its product
line a boost once it hits the market this month, analysts say. Other networking companies
already
have gigabit products on the market.

''It's critical that they get this out,'' Gartner's LeBaron said. ''I don't think it's too late.''

Xylan's expertise makes it an acquisition target, Woodward adds. Hill wouldn't comment on
acquisition rumors.

Lucent, Nortel, Alcatel and IBM would be logical choices, Woodward says.

''There's a new rumor every month on who's going to buy Xylan and for how much,'' Woodward
said. ''But I don't believe they're actively trying to sell the company.''

(C) Copyright 1998 Investors Business Daily, Inc.
Metadata: XYLN NT CSCO GART IBM ALA LU BAY I/3574 I/4890 I/7392 I/3573 E/IBD E/SN1
E/TECH