News & Analysis
Comment
mark.lapedus
HSBC says GlobalFoundries (GF) won't be a real factor in the foundry market ...
double-o-nothing
GlobalFoundries is trying to become, in a few years, what TSMC became after many ...
GlobalFoundries puts rivals on notice, tips 20-nm process
Mark Lapedus
9/1/2010 9:25 PM EDT
No major surprises on process roadmap
Meanwhile, at its technology event, GlobalFoundries outlined its process roadmap, but there were no major surprises on that front, said Dean Freeman, an analyst with Gartner Inc. "They are on track" in terms of their process roadmap, he said.
Perhaps the only surprise was the disclosure of a new 20-nm process on the roadmap. GlobalFoundries declined to elaborate on the 20-nm technology, nor would it discuss the timetable for delivery.
Regarding its process roadmap, GlobalFoundries is more aggressive than most had thought. For example, on the processor side of the business, where it serves AMD, GlobalFoundries' roadmap "has been the most aggressive it has ever been," said Doug Freedman, an analyst from Gleacher & Co.
At present, GlobalFoundries has been ramping up 65-nm (and above) processes at Chartered's Fab 7 facility in Singapore. The Fab 7 facility will ramp "a little bit" of the company's 40-nm process, Grose said.
GlobalFoundries' Fab 1 facility in Dresden, Germany, will become the high-volume fab for the company's 45- and 40-nm processes. That fab is ramping to 80,000 wafers a month in the facility.
Within the fab, the company is shipping a 40-nm bulk process, based on a low power offering. The Dresden fab is also offering a 45-nm SOI process, mainly for AMD's processors. By the first half of 2011, GlobalFoundries will bring up a 32-nm SOI with its initial high-k/metal-gate (HKMG) scheme. The 32-nm process is also tuned for AMD's processors.
Surprisingly, even before 32-nm, GlobalFoundries will tape out a 28-nm bulk CMOS process with HKMG. The 28-nm process, which will initially be a high-performance offering, is expected to tape out by the end of 2010. Both 32- and 28-m HKMG offerings are based on a gate-first scheme.
At the event, the company announced the addition of a new technology offering based on its 28-nm HKMG technology. Scheduled to begin risk production in Q4 2011, the 28-nm High Performance Plus (HPP) technology provides a performance boost of as much as 10 percent over the company’s current 28nm High Performance (HP) offering.
The Dresden fab is starting work to help develop 22-nm CMOS process and will run the process in volume. It is not clear whether the 22-nm will include a departure from the gate-first HKMG technology.
In the future, Fab 1 will also become the home of a new 20-nm pilot line. GlobalFoundries' wafer fab under construction in New York state, Fab 8, would run the 22-nm production and more advanced nodes.
Construction for Fab 8 started in July of 2009. The fab will have 60,000 wafer starts per month once it goes into full production. Production is expected to go online in 2012.
With risk production set to begin in the second half of 2012, the company is well on its way to delivering 22- and 20-nm technology to customers for product introduction in 2013. The 20-nm technology offerings will come in two varieties: a High Performance (HP) technology designed for wired applications such as servers and media processors, and a 20-nm Super Low Power (SLP) technology designed for power-sensitive mobile applications.
Meanwhile, at its technology event, GlobalFoundries outlined its process roadmap, but there were no major surprises on that front, said Dean Freeman, an analyst with Gartner Inc. "They are on track" in terms of their process roadmap, he said.
Perhaps the only surprise was the disclosure of a new 20-nm process on the roadmap. GlobalFoundries declined to elaborate on the 20-nm technology, nor would it discuss the timetable for delivery.
Regarding its process roadmap, GlobalFoundries is more aggressive than most had thought. For example, on the processor side of the business, where it serves AMD, GlobalFoundries' roadmap "has been the most aggressive it has ever been," said Doug Freedman, an analyst from Gleacher & Co.
At present, GlobalFoundries has been ramping up 65-nm (and above) processes at Chartered's Fab 7 facility in Singapore. The Fab 7 facility will ramp "a little bit" of the company's 40-nm process, Grose said.
GlobalFoundries' Fab 1 facility in Dresden, Germany, will become the high-volume fab for the company's 45- and 40-nm processes. That fab is ramping to 80,000 wafers a month in the facility.
Within the fab, the company is shipping a 40-nm bulk process, based on a low power offering. The Dresden fab is also offering a 45-nm SOI process, mainly for AMD's processors. By the first half of 2011, GlobalFoundries will bring up a 32-nm SOI with its initial high-k/metal-gate (HKMG) scheme. The 32-nm process is also tuned for AMD's processors.
Surprisingly, even before 32-nm, GlobalFoundries will tape out a 28-nm bulk CMOS process with HKMG. The 28-nm process, which will initially be a high-performance offering, is expected to tape out by the end of 2010. Both 32- and 28-m HKMG offerings are based on a gate-first scheme.
At the event, the company announced the addition of a new technology offering based on its 28-nm HKMG technology. Scheduled to begin risk production in Q4 2011, the 28-nm High Performance Plus (HPP) technology provides a performance boost of as much as 10 percent over the company’s current 28nm High Performance (HP) offering.
The Dresden fab is starting work to help develop 22-nm CMOS process and will run the process in volume. It is not clear whether the 22-nm will include a departure from the gate-first HKMG technology.
In the future, Fab 1 will also become the home of a new 20-nm pilot line. GlobalFoundries' wafer fab under construction in New York state, Fab 8, would run the 22-nm production and more advanced nodes.
Construction for Fab 8 started in July of 2009. The fab will have 60,000 wafer starts per month once it goes into full production. Production is expected to go online in 2012.
With risk production set to begin in the second half of 2012, the company is well on its way to delivering 22- and 20-nm technology to customers for product introduction in 2013. The 20-nm technology offerings will come in two varieties: a High Performance (HP) technology designed for wired applications such as servers and media processors, and a 20-nm Super Low Power (SLP) technology designed for power-sensitive mobile applications.
|
|
|
|
|
Navigate to related information
Neo1
9/2/2010 2:19 AM EDT
Good news, their 20nm process could offer the best bet for companies other than intel to remain competitive in the market without depending too much on TSMC. But this foundry consolidation is imminent. Samsung will not be holding on to their fabs for too long either.
Sign in to Reply
unknown multiplier
9/2/2010 2:29 AM EDT
Very brave of GF, good for them.
Sign in to Reply
mark.lapedus
9/2/2010 3:06 AM EDT
Will GF make it or fall flat on its face? Comments out there?
Sign in to Reply
goafrit
9/2/2010 9:22 AM EDT
I think they will fail because building factories is not a good model. This is capital intensive with minimal returns. I see no good value in this strategy.
Sign in to Reply
CamilleK
9/2/2010 3:49 AM EDT
@mark.lapedus: to answer your question, they have a good chance to make it considering that the theme today was collaboration with partners like the news today with ARM that they taped out a dual core ARM cortex A9 in 28nm (http://www.tradingmarkets.com/news/press-release/armh_globalfoundries-launches-industry-s-first-28nm-arm-cortex-a9-processor-platform-with-gate-first-high-1145834.html). They also will develop Flash technology with Freescale, also announced today. The 28nm 'Gate first' HKMG technology will deliver 10-20% improved die size and cost over gate last, and allows true scaling 100% density increase over 40nm. Given that they are building capacity, have resources, lined up partners, are fully behind 3D TSV, are on their way to have test chips in 2H2011 in 20nm, it looks like they have the right plan and they can make it if they execute.
Sign in to Reply
greenpattern
9/2/2010 4:54 AM EDT
TSMC is also shooting for 20 nm in second half of 2012 (risk production).
Sign in to Reply
greenpattern
9/2/2010 12:18 PM EDT
I read that tsmc will not go planar transistor below 20 nm. Will Global follow suit?
Sign in to Reply
3D Guy
9/2/2010 10:24 AM EDT
I like the fact that Global Foundries is trying to change the foundry world... it is good for our industry since it will cause their competitors to innovate too.
But its an execution nightmare. See what they have on their table: 45nm SOI, 40nm bulk, 32nm SOI, 28nm bulk, 22nm (SOI?), 20nm bulk, and a major part of these in three flavours: high performance, low-power and high-performance plus. Adding to this, they are trying to develop MEMS, NVM, analog, etc.
When people are struggling to get one 40nm technology up and running with good yield, these guys are being too ambitious! I wish them luck, but if I were the CEO there, would keep the company focused on a few things and do these things well.
Sign in to Reply
mark.lapedus
9/2/2010 1:18 PM EDT
I agree. I don't know how they are going to balance their foundry work for AMD and the rest of the industry. A double headache. Of course, they got a rich parent.
Sign in to Reply
Ratgebber
9/2/2010 4:40 PM EDT
PR and yield do not have to correlate on the ITRS roadmap.
Sign in to Reply
gutiea
9/2/2010 7:26 PM EDT
The technical side, as proposed, seems to be nearly impossible to even visualize. Focus is not the strong point of this company, it also suggests management believes there is an endless supply of sacrificial capital. Perhaps the UAE investors are willing to spend unlimited amount of resources?
The financial aspect, based in losing money in every single product Globalfoundries ship, with losses exceeding 75% of its revenue, and a lousy balance sheet; is a complete delusion unless they can sell ATIC on funding them to the end of the times. Have they already done so?
What will TSMC do?
It does not look good for Globalfoundries. They seem to be imploding; or just a quarter away from it.
Sign in to Reply
KB3001
9/2/2010 10:02 PM EDT
I believe GF will ultimately lead the way in the foundry business. Their owners (ATIC) have deep pockets and have no shareholders to worry about :-) They can take the short term losses for the sake of long term dominance.
Sign in to Reply
FH1
9/3/2010 9:44 AM EDT
Dear Goafit ... "I think they will fail because building factories is not a good model. This is capital intensive with minimal returns. I see no good value in this strategy" ... an interesting philosophy. So just who exactly is supposed to build factories and make the chips ??? Are you suggesting the world defaults to no foundries, no IDMs ... no wafer factories ??? Bottom line is when there's no competition left, wafer prices will skyrocket making them great on returns and good value to have.
Dear Noe 1 ... "Samsung will not be holding on to their fabs for too long either" ... in your dreams. My guess is Samsung is the only chip competitor that really keeps TSMC awake at night.
As for ATIC ... fine words and nice fanfare but words and fanfare's are the easy part; the proof of the pudding's will be in the eating. What ATIC really needs is TSMC's prowess, Samsung's ruthlessness and Intel's execution; everything else is an expected and necessary given.
Sign in to Reply
resistion
9/3/2010 10:57 AM EDT
Like UMC, GF has a handicap in that it must rely on outside mask suppliers. I don't think they can advance easily to subsequent nodes as smoothly as TSMC, Samsung, Intel. In fact, it may make sense for them to consider maskless or even direct write instead of something mask-intensive like EUV.
Sign in to Reply
Pablito
9/3/2010 2:17 PM EDT
test
Sign in to Reply
Luis Sanchez
9/3/2010 5:01 PM EDT
Quite interesting to read you all guys,
one really learns about the IC industry just by glancing over your comments.
So I see that Samsung is strong in chip manufacturing and there are many non-foundry vendors who depend on the makers.
And there are different techs like 40nm, till 20nm.
I would like to know what is the ROI of the semiconductor business is this what we might call hi value added? Perhaps so since at the end the makers are selling little pieces of sillicon but with a high agregated value right? If we think of it... is like making stones speak, calculate and the sort...
Sign in to Reply
CamilleK
9/4/2010 1:15 AM EDT
@resistion: I am not sure I agree that GF using outside mask suppliers is a handicap. It is in my opinion providing a key differentiation in the sense that they can offer competitive rates from various mask suppliers and they do not even need to invest in the internal development of mask making and they can use the capital to work on the process R&D. In addition to this customers may feel that they are overpaying in terms of masks if they have no alternative options but going through the foundry's captive mask shops.
Sign in to Reply
double-o-nothing
9/4/2010 7:57 AM EDT
GlobalFoundries is trying to become, in a few years, what TSMC became after many nodes. That is a tall order. Going to double patterning at 22/20 nm also not easy for them to accept, I bet.
Sign in to Reply
mark.lapedus
9/7/2010 1:35 AM EDT
HSBC says GlobalFoundries (GF) won't be a real factor in the foundry market until 2011. I agree. I am still not convinced about them. Sometimes, I think they are really lost. I am not sure they have made the transition as an IDM under AMD to the pure-play foundry business. They got deep pockets, but they don't have a clue right now. They may figure it -- perhaps sooner than later. But for now, I see red ink and slower-than-expected sales for the foreseeable future.
Sign in to Reply