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.
Technology Stocks : Lam Research (LRCX, NASDAQ): To the Insiders -- Ignore unavailable to you. Want to Upgrade?


To: etchmeister who wrote (5670)3/4/2008 11:59:57 PM
From: etchmeister  Read Replies (1) | Respond to of 5867
 
Micron and Nanya signed MOU on sub 50nm DRAM technology development; Keep competition in UFD makers spurs vendors to introduce low-price high-density products to ease inventory

Published Mar.4, 2008

Micron and Nanya signed MOU on sub 50nm DRAM technology development

DRAM technology, which used to be clearly classified into stack and trench technology, is about to see fundamental change amid leading advocators’ technology development and cross partnership.

Micron and Nanya announced on March 3 that the companies have signed a memorandum of understanding (MOU) to explore potential technology sharing, joint technology development and development of a new joint venture. Both parties will jointly focus on the development of sub-50nm technologies.

Prior to Micron and Nanya’s announcement, Qimonda announced that technology breakthrough with DRAM roadmap to 30nm and feature cell sizes of 4F2, with production of 1Gb DDR2 slated for the second half of 2008. The company will transit to 58nm production in 2008 and advance to 46nm in 2H09. While the announcement raised eyebrows among marketers, some hinted that the coincidence of the two announcements’ timing implies that Qimonda may foretell some hidden message.

By referring to sales from the mentioned companies, DRAMeXchange believes that market share of respective stack and trench will see structural changes alongside with the strengthen ties among involved parties.

As of 2006, the trench camp records a market share of 23%. But as DRAM makers competed over capacity in 2007, the share has decline to 18%. In light of the emergence of new partnership mode, as well as competition from the stack camp, market share of the trench camp may be eroded further.

Strong NT dollar mutes DRAM transaction

DRAM spot prices maintained flat in the week Feb 29-Mar 3 with minimal transaction amid strong NT dollar and seasonality downturn.

After seeing price of DDR2 512Mb eTT hit the record-high US$1.17 prior to the Lunar New Year in early Feb, price headed south afterwards with price once dropped by almost 36% to US$0.86 during the holidays. Price trend was notionally stable last week with limited fluctuation in both eTT and branded chips. Only DDR2 512Mb eTT chip posted a 1.1% drop and DDR2 667MHz 1Gb posted a 1% gain last week.

Marketers attributed the strong NT dollar appreciation against US dollar as the primarily reason for such a price stability. As NT dollar now appreciates to one US dollar to 31.1 NT dollar, up from the previous 32.4 NT dollar, DRAM makers are exposing to aggravated exchange loss pressure with considerable DRAM inventory piled up. Coupling with the slow season effect, transaction thus was discouraged further.

Keep competition in UFD makers spurs vendors to introduce low-price high-density products to ease inventory

Seasonality downturn has prompted more USB Flash drive (UFD) makers to introduce high-density products with attractive prices. Prior to the Lunar New Year in early February, some second- and third-tier branded UFD vendors introduced UFD in 8GB memory density at US$30 only. When the holidays arrived, more vendors joined the league with a few first-tier branded vendors also introducing 8GB UFB at similar price levels. Not only did the existence of these low-price high-volume UFD highlight the keen competition in the NAND Flash application market, it also indicated UFD makers’ attempts in easing the inventory and boost market share.

The current mainstream memory density of UFD is 2GB, with most priced in the range of US$15. The emergence of 8GB UFDs priced at only US$30 highlights vendors’ attempt to spur sales via boosting memory density with limited price adjustment. In the UFD and memory card market, when a product that supports a certain specification is priced below US$30, rapid penetration follows as many consumers will use the price mark as a benchmark. Once retail price of these applications break this benchmark, noticeable growth in consumption incentive will be seen.

UFD with higher density penetration is also driven by the mass adoption of USB interface at a broad range of multimedia devices. In addition to PC users, many digital TV (DTV), set-top-box (STB), DVD player, mini hi-fi system (home and car), digital photo frame, support USB interface for data transmission. The extended presence of UFD beyond PC thus encourages more vendors to introduce 16GB or 32GB UFD to channels. From the supply point of view, as chipmakers have consistent advancement over technology and 12-inch wafer production, the growing amount of NAND Flash chips also help downstream players to adjust their strategy, and in turn help spur NAND Flash demand.

NAND Flash spot price recap, Feb 25 - Mar 3

In the SLC segment, 1Gb dropped by 5.9% to US$1.90; 2Gb dropped by 1.7% to US$2.85; 4Gb dropped by 9.5% to US$4.30, 8Gb dropped by 9.5% to US$8.56 and 16Gb dropped by 3% to US$19.13. In the MLC segment, 4Gb dropped by 6.3% to US$2.08; 8Gb dropped by 7.8% to US$2.72; 16Gb dropped by 5.2% to US$5.11 and 32Gb dropped by 3.5% to US$11.02.

For tables check out link

dramexchange.com



To: etchmeister who wrote (5670)3/24/2008 11:58:34 PM
From: etchmeister  Respond to of 5867
 
Analyst: Best all around for Infineon to kill Qimonda

he's got a point though don't kill it - refocus it ... Qimonda could sell its Inotera stake and focus on specialty DRAM etc. it's going to be hard to compete in commodity DRAM and it's going to drag down Nanya who needs a stronger partner.)
Bolaji Ojo
EE Times
(03/20/2008 1:26 PM EDT)

MANHASSET, N.Y. -- A semiconductor industry analyst has called on Infineon Technologies AG to shutter Qimonda AG, its majority-owned and money-losing DRAM division, arguing that such a move has become necessary to help the company focus better on re-energizing its remaining business units, as well as help ameliorate the general DRAM oversupply situation.

Malcolm Penn, CEO at London-based research firm Future Horizons, said Infineon would have a difficult time finding a buyer for its remaining 77.5 percent stake in Qimonda due to oversupply and severe pricing problems in the market.

"There is absolutely no commercial, market, business, technology or competitive reason for anyone to buy Qimonda," said Penn in a written response to questions from EE Times. "Even if you paid someone to take Qimonda off your hands, the recipient would have to be mad or foolhardy to consider it."

Infineon has said it wants to "significantly" reduce its interest in Qimonda below 50 percent by the first quarter of 2009, although executives at the semiconductor company admitted that selling their stake has so far been difficult.

The company sold a portion of its Qimonda shares last September, but noted that this was done at a net loss of 84 million euros, according to figures from a February 14 presentation by president and CEO Wolfgang Ziebart during Infineon's annual general meeting.

Ziebart said problems at Qimonda had a "negative effect on the Infineon Group's earnings" during the fiscal year ended Sept. 30, 2007 and "weighed heavily on group revenues, which slipped to 7.68 billion euros last year from 7.93 billion euros in fiscal 2006."

Qimonda's and Infineon's stock prices have suffered over the last 12 months, mainly as a result of problems at the memory supplier, but also due to outstanding issues within Infineon's communications-IC business.

Qimonda's American Depositary Share (ADS) price fell in intraday trading on Wednesday (March 20) to $4.10, down 76 percent from the 52-week high of $17.29, giving the company a market value of about $1.4 billion.

This puts Infineon's stake at the memory maker at approximately $1 billion before charges. Qimonda had approximately $1.5 billion in cash and short-term investments and long-term debt of $324 million at the end of its last fiscal year.

Liquidating Qimonda could help Infineon put a quick end to the losses at the company, although it would also have to deal with issues related to laying off the memory maker's more than 13,000 employees. Also, Infineon might not be easily able to dispose of the company's assets.

Qimonda's gross profit tumbled more than two-thirds, to $311 million in the fiscal year ended Sept. 30, 2007, down from $973 million in the previous fiscal year, as DRAM prices nosedived at a double-digit clip per quarter. Analysts estimate Qimonda's fiscal 2008 revenue would slip to $4.8 billion, down from $5.2 billion in fiscal 2007. They expect the company to report a net loss of $5.57 per share for the ongoing year, wider than the 0.97 cents per share loss it recorded for fiscal 2007.

A recovery in DRAM pricing will not occur until the current oversupply situation is corrected, according to Future Horizons' Penn, who suggests one or more suppliers should exit the business.

"The industry as a whole would benefit enormously if Qimonda was simply to die," he said. "The current problem in DRAM is oversupply. Taking out a competitor will go a long way to helping that situation."



To: etchmeister who wrote (5670)8/1/2008 8:12:30 PM
From: etchmeister  Respond to of 5867
 
Qimonda could sell its Inotera stake and focus on specialty DRAM etc. it's going to be hard to compete in commodity DRAM and it's going to drag down Nanya who needs a stronger partner.
1.) Nanya is lagging in regards to die size 8F2 vesus 6F2
2.) Copper will be mandatory @ 65nm and below (at least according to DRAM interconnect roadmap); Micron is the only memory guy that runs copper in production for some time
3.) Micron, Sammy and Hynicks all settled on stack capacitor versus trench capacitor; not sure about Elpida but Qimonda seems to be the only one that still banks on trench
4.) Nanya got leap frogged 512Mb versus 1Gb; what I recall backend testing is significantly less for higher density


Done deal - now it's long awaited.
Pay attention ...
Message 24366350
They going to retool BIG time to copper;
this is one of those BIG projects LRCX and Varian have "penciled" in going forward...

Micron nearing deal to buy Qimonda JV stake

Mark LaPedus
EE Times
(08/01/2008 10:20 AM EDT)

SAN JOSE, Calif. -- Micron Technology Inc.'s long-awaited move to buy Qimonda AG's stake in Inotera Memories Inc. could be near, according to an analyst.

Rumors are running rampant about Micron's intentions. Taiwan's Nanya Technology Corp. and Germany's Qimonda have a joint DRAM manufacturing venture in Taiwan called Inotera. That venture appears to be in limbo.

Micron recently entered into a separate R&D and DRAM venture with Taiwanese memory rival Nanya, which, turn, raises questions about the Qimonda-Nanya alliance.

That venture, MeiYa Technology Corp., is DRAM joint venture between Micron and Nanya. It appears Micron will also take over the Inotera venture with Nanya.

''Our checks continue to suggest that some form of deal that will allow Micron to take Qimonda's place in the Inotera manufacturing joint venture with Nanya is close to being consummated,'' said Daniel Berenbaum, an analyst with SG Cowen Securities Inc., in a report.

''Even if Micron has to purchase Qimonda's shares directly at a big premium to market price, it would allow a capex cut, lower unit costs, increase DRAM bit growth, and provide a royalty revenue stream,'' he said.

Next week, Micron is expected to host its analysts day. According to Berenbaum, Micron is expected to address the following issues:

*Industry supply/demand dynamics for memory. ''We expect both NAND and DRAM to remain in oversupply through H109,'' he said.

*The long-awaited CMOS image sensor spinoff, ''which may finally be close to getting done,'' he said. The Boise, Idaho, chip maker recently set the stage for a possible spinout of its CMOS image sensor business by transforming its sensor business unit into an independent division. That entity, Aptina Imaging, will be free to seek manufacturing alliances with outside foundries. The captive business unit had relied solely on Micron for its manufacturing needs.

*Options and ramp plans for the IM Flash NAND JV with Intel. IM Flash is a joint NAND production venture between Micron and Intel. ''Intel appears disappointed with profitability of JV and may be looking to exit or reduce its stake,'' he said.

Rumors are running rampant that Seagate Technology may buy Intel's stake in IM Flash. On one hand, Intel's margins are taking a hit due to the NAND downturn.

But Intel and Micron have recently leapfrogged the competition and claimed the technical lead in the NAND flash memory market. Reportedly moving ahead of Toshiba, Samsung and others in the process race, the Intel-Micron duo have introduced the industry's first sub-40-nm NAND flash device, by rolling out 34-nm, 32-gigabit (Gbit) multi-level cell (MLC) chip. Previously, the leading-edge NAND device from Intel and Micron was a 50-nm part.

Separately, the company also announced today the appointment of Mark Adams to the position of vice president of worldwide sales. Adams, formerly Micron's vice president of Digital Media, joined Micron in 2006 with the company's acquisition of Lexar Media Inc. Adams replaces Mike Sadler, who has accepted a new role in the office of the president for MeiYa.



To: etchmeister who wrote (5670)10/13/2008 7:24:02 PM
From: etchmeister  Read Replies (1) | Respond to of 5867
 
Remember where you red it first...six months ago.
And what do I get for following memory closely - I get "static"!!! :0)
Germany builds great cars but failed in as a chip maker or equipment maker.
Back 20+ years ago some people said semiconductors are Siemens most expensive "hobby".
I think Germen like to build things at 110% and test it over and over again but in reality often things have to pushed out the door before being perfect and fix things later on the fly.
But that usually does not bode well with German engineering sense.
The Germen also spend a lot of time to figure out why things are working before "releasing" it, in the US "release" it in case it's working - ask questions later.

Qimonda could sell its Inotera stake and focus on specialty DRAM etc. it's going to be hard to compete in commodity DRAM and it's going to drag down Nanya who needs a stronger partner.
1.) Nanya is lagging in regards to die size 8F2 vesus 6F2
2.) Copper will be mandatory @ 65nm and below (at least according to DRAM interconnect roadmap); Micron is the only memory guy that runs copper in production for some time
3.) Micron, Sammy and Hynicks all settled on stack capacitor versus trench capacitor; not sure about Elpida but Qimonda seems to be the only one that still banks on trench
4.) Nanya got leap frogged 512Mb versus 1Gb; what I recall backend testing is significantly less for higher density


Qimonda sells Inotera stake, cutting 3,000 jobs
Monday October 13, 1:54 pm ET
Germany's Qimonda sells stake in Inotera to Micron for $400 million; cuts 3,000 jobs

BERLIN (AP) -- German memory-chip maker Qimonda AG is cutting 3,000 jobs and selling its stake in Inotera -- a joint venture with Taiwan's Nanya Technology Corp. -- to Micron Technology Inc. for $400 million.

ADVERTISEMENT
Qimonda, which is majority owned by Infineon Technologies AG, said the cash deal to sell its 35.6 percent stake in Inotera Memories Inc. is part of a wider restructuring and cost-cutting program. The company said it is trying to become "leaner and more focused" because of a severe downturn in the memory chip industry.

Prices of flash memory chips have fallen this year because of an oversupply in the market.

Qimonda is not the first to be hit by such pricing pressures -- last week, Micron itself said it would cut about 15 percent of its work force. Micron said flash-memory oversupply had driven the selling price for chips below manufacturing costs.

Qimonda expects to record a one-time loss of about 300 million euros ($403 million) on its investment in Inotera.

"The sale of our stake in Inotera is a key step in Qimonda's restructuring, helping to give us a cash influx and sharpening our focus," Chief Executive Kin Wah Loh said in a statement. "We plan to concentrate our efforts on selected market segments where we can best leverage our innovative technologies," he added, saying, "we will rationalize our manufacturing footprint and streamline our operational and personnel structures."

Qimonda plans to stop manufacturing chips at one of its two facilities in Richmond, Va., by January. A factory in Dresden, Germany, will be shut down by the end of next March.

The company also said it will reduce research, development and administrative expenses and staff -- mainly in Munich, Dresden and Raleigh, N.C. -- in moves that will affect some 3,000 employees.

The company said the plans call for some 1,500 job cuts in Germany and 1,500 in the United States. Qimonda said it had a total of 12,200 employees as of September.

Qimonda expects to take restructuring charges of some 50 million euros ($67 million) in the next quarter, and sees the program resulting in annual cost savings of about 450 million euros ($604 million) once implemented.

Also on Monday, Qimonda said its chief financial officer, Michael Majerus, is stepping down and will be replaced by Chief Operating Officer Thomas Seifert on an interim basis.

Qimonda said Majerus' resignation "comes at his own wish for personal reasons."

(This version corrects that not all production will cease in Richmond; the company has two facilities there.)