Comms chips

Vitesse Drops Some Packets

Following in the footsteps of PMC-Sierra Inc. (Nasdaq: PMCS), Vitesse Semiconductor Corp. (Nasdaq: VTSS) appears to have abandoned development of packet-processing silicon (see PMC-Sierra Pulls Packet Silicon).

In developments that have come to light following last week's earnings call, Vitesse will suspend development of a future four-port Gigabit Ethernet network processor and the OC192 (10 Gbit/s) device planned for 2003 (see Vitesse Pumps Up its Net Processor ), according to reports from the Linley Group, a technical consultancy for network processors.

The 100-strong network processor division, which comprises the staff of startups Sitera and Xaqti, which Vitesse bought in 2000, will be cut down to 45 staff, who will be reassigned to other projects (see Vitesse to Buy Sitera for $750M). In aggregate, Vitesse paid about $1.3 billion for the startups. Despite the company's stated intention to continue marketing and supporting its existing IQ200 and IQ2200 products, this move will essentially see Vitesse bow out as a supplier of network processors, says one analyst.

"Although Gigabit Ethernet is the biggest segment of the network processor market and is growing rapidly, we doubt that many customers will be interested in the IQ2200 without a roadmap to higher performance," says Linley Gwennap, senior analyst at Linley.

As is the way with these things, the truth trickled out a few days after the company's earnings announcement. At the end of last week, Vitesse posted a massive loss of $887.2 million, or $4.48 loss per share, on revenues of $43 million (see Vitesse Reports Q3, Lays Off).

This huge loss includes a $695.5 million writeoff of goodwill, which under new regulations, must be recorded all at once, rather than being spread across several quarters. Pro forma net loss was $20.6 million.

Unsurprisingly, Vitesse announced restructuring plans at the same time, which include curtailing certain R&D projects, consolidating manufacturing facilities, and laying off 200 of its 1,160 staff. Two of the development areas that have been curtailed are network processors and traffic manager chips -- both chips specific to processing packet-based traffic.

"With these actions we expect our break-even point to be approximately $60 million per quarter," said Vitesse's CEO, Lou Tomasetta, in a prepared statement. "More importantly, we will focus our development efforts on products that give us the most immediate return on investment."

According to Linley's Linley Gwennap, the IQ2000 had fallen well behind low-end market leaders -- the IXP1200 from Intel Corp. (Nasdaq: INTC) and Applied Micro Circuits Corp. (AMCC)'s (Nasdaq: AMCC) nP family -- in terms of design wins. "Vitesse had been losing more NPU customers than it gained during the first half of this year," he notes.

Vitesse will also be abandoning any future development of its PaceMaker traffic manager chip, which it gained from its acquisition of Orologic in April 2000 for $450 million. Again, the company says it plans to continue marketing the chip, but without any R&D investment, the product will quickly fall behind competition from the likes of Azanda Network Devices (see Azanda Flips Its Chips).

All in all, Vitesse doesn't have much to show for the $1.3 billion it spent to acquire Sitera, Xaqti, and Orologic, Gwennap points out.

Vitesse did not return calls.

But these layoffs appear to be part of the ebb and flow in a competitive environment and don't indicate a problem with the network processor market in general, it seems. Other vendors have been pushing forward with new products.

Earlier this week, two companies announced new network processors: Agere Systems (NYSE: AGR) announced that it had shrunk its original four- or six-chip Payload Plus solution into a single device called INP5 (see Agere Supplies Pumps, Processors); Silicon Access Networks Inc. also announced that it had begun shipping its 20-Gbit/s network processor (see Silicon Access Tries 20-Gig).

— Pauline Rigby, Senior Editor, Light Reading
http://www.lightreading.com Want to know more? The big cheeses of the optical networking industry will be discussing this very topic at Opticon 2002, Light Reading’s annual conference, being held in San Jose, California, August 19-22. Check it out at Opticon 2002.

Register now and save $500 off the registration fee. Just use the VIP Code C2PT1LHT on your registration form, and deduct $500 from the published conference fee. It's that simple!

Page 1 / 2   >   >>
edgecore 12/4/2012 | 10:02:40 PM
re: Vitesse Drops Some Packets
Heard that IBM is reviewing whether to stay in this market or not with their NP4GS3 and NP2G

Anyone else hear the same thing?

edgecore 12/4/2012 | 10:02:40 PM
re: Vitesse Drops Some Packets
Heard that IBM is reviewing whether to stay in this market or not with their NP4GS3 and NP2G

Anyone else hear the same thing?

ethelred_the_unready 12/4/2012 | 10:02:39 PM
re: Vitesse Drops Some Packets If this is their 10Gbps chip, I heard in Nov/01
that they canceled it.


toonces_has_the_car 12/4/2012 | 10:02:37 PM
re: Vitesse Drops Some Packets IMHO, there are 3 companies with decent long-term prospects in the NPU market:


The remainder have their work cut out for them.
There may be a play in low-cost, not-quite-so-flexible routers for some of the others. Not really NPU, but rather HW-based forwarding.

BobbyMax 12/4/2012 | 10:02:35 PM
re: Vitesse Drops Some Packets Vitesse has lost every penny of $1.3B Billion by the acquisitions of Sitera and Xaqti ( plus one more company). These acquisitions were questionable to start with. It was known that the company would not gain anything by acquiring Xaqti, but it did not do proper due dilligence on Xaqti. Even after the it became clear that Xaqti was not a strong acquisition it kept the company fully funded for a period beyond 12 months or so.

The company is not accepting its mistakes and ibcompetence. The upper management keeps drawing their salaries and benefits whereas the shareholders have been wiped out.
lightbulb0 12/4/2012 | 10:02:35 PM
re: Vitesse Drops Some Packets NPU will not find places in any high end boxes while the low-end is much too crowded.
jamesbond 12/4/2012 | 10:02:34 PM
re: Vitesse Drops Some Packets
I hope Ciscos, PMC-Sierras, Vitesses, Intels of
the world learn a lesson from this. It seems
like the execs of the companies that get
acquired always do well financially, no matter
the state of the product they were making.

The effective "price" paid should be related to
the sale of the product(s) acquired. Obviously
that means acquirer pays the price over some
years (maybe 4 yrs).

This will do two things -
1. Keep people from getting rich overnight and
losing any motivation/incentive etc.

2. Only serious people (i.e who are in it for
long run and believe in it) will start the

why do you guys think this is not done?
wilecoyote 12/4/2012 | 10:02:33 PM
re: Vitesse Drops Some Packets Good point but be careful what you ask for JB...

When a large company acquires a startup, there are a dozen variables in pricing the deal but a lot of the time it really comes down to "how badly do we think we need this product in our portfolio?" If CSCO had told Cerent: "you have no sales yet, we're going to pay a paltry sum" the Cerent VCs and executives would have said "then we'll go it alone," or "we will sell for a paltry sum to someone else just to spite you." Next thing you know, Nortel is shipping a billion in 15454s a year later and CSCO is still nowhere in metro when the market is still hot, Pirelli looks like that much bigger a dog, and Chambers looks like a chump. Yes, image plays a part...

Startups generate some buzz, maybe get a handful of design wins which don't mean much until they translate into real revenues over a year later, so you can't really value startups according to revenues or you may lose out on the deal. Now, what I'm saying assumes the company really has something that they can productize (rare), that the public company really needs. Big assumptions that almost never really pan out for, as you say, anyone but the CEO, VCs and Executive staff.
snowbound 12/4/2012 | 10:02:33 PM
re: Vitesse Drops Some Packets It is not uncommon for acquisitions to have performance clauses built in and I suspect this will become more the rule going forward.

During the bubble when a lot of these "questionnable" acquisitions occured, it was a bit of a seller's market and terms ended up being quite generous. Many acquisition deals were fueled out of fear of not being viable unless you were a one stop shop and needing to have a story in the hot-technology du jour. Not a lot of good decisions are based in fear ... I I think the economy is self correcting this phenomenon for the most part ...

Easy to criticize these acquisitions now in hindsight - but imagine the crap these companies would have put up with from the analysts / investors if they hadn't put some sort of NP strategy in place back then. And of course at the time, growth by acquisition was considered an excellent strategy.

Execs of companies that get acquired do well because in general they take significant personal risk to get the start up going. How many line workers take out a mortgage to finance their company ? Of course there is some abuse. And we all see the big payouts on the successful ones and are a little jealous and are pissed that our RRSP (canada) or 401k (US) are way down but think of the 90% of start ups that fail and you realize that the reward has to be huge to justify the personal risk.
jamesbond 12/4/2012 | 10:02:32 PM
re: Vitesse Drops Some Packets Execs of companies that get acquired do well because in general they take significant personal risk to get the start up going. How many line workers take out a mortgage to finance their company ?


You are kidding right? How many networking
startups in the 90s that got acquired for
big bucks were started by CEO's personal

if you think about it, it was my and your
money that enabled the startups. Where is that
money now? it is in those CEO's and exec;s
Page 1 / 2   >   >>
Sign In