Judy, Andrew Vance's response to your query was one of the most lucid, reasoned, and well thought out arguments for owning Cymer that I have heard. Not only am I printing it out for all of my friends and family members to see, I am taking the liberty of posting it here for those too lazy to click on your link. Thanks for posting it.
(Most of these friends and family members now have nice profits and called me this weekend to see if they should sell. When I told them not to sell any until 100 and more until 200, they thought I was delirious. This will help.)
CYMI is almost the only game in town.
Last look, they had 85% of the market.
Competitors are few and far inbetween, with most no worth worrying about.
Competitors: Komatsu (Japan), USHIO (trying at least), Lamda Physik.
The onyl one to worry about, IMO, is Komatsu only becasue they are in Japan and could cut deals with both Nikon and Canon.
Major issue is the interface. For simplicity purposes, consider that the DUV Stepper or Scanner is made up of 3 parts. First is the stepper system itself (optics, wafer stage, reticle holder, etc), second is the DUV source like what CYMI provides, and last is the interface that joins both together. My understanding is that you jusat cannot swap out DUV sources from one vendor to the other without addressing interface issues.
Therefore, CYMI is almost locked into those companies that already have DUV systems installed. Subsequent systems will have the CYMI source specified to allow for equipment commonality and to eliminate a good deal of requalifications. Also, it makes sense to be sole source on this from a maintenance, spare parts, service, etc. perspective.
You do not want to screw around here as an end user. Life is difficult enough in lithography that you do not want, need, or have to inject another set of variables into your process. Also, due to the sophistication of these systems, I would also bet that corporations would also use a blanket policy for these systems. Replacement DUV Lasers and parts are exrtremely costly so you do not want to have to have duplicate inventories.
For new implementation of DUV technology, you could make an argument for the field being wide open. However, this does not really worry me since many of my fellow lithogrpahers are not big risk takers. No one wants to re-invent the litho tool and be the guinea pig for a CYMI competitor. So, if there is any loss of market share, it will come in Asia first and then slowly across the sea here. The other reason for little concern is that once the flood gates open on DUV (inventories are getting lower), their will be such a big rush that CYMI might not be able to meet the demand whgen it comes in. at Last count, CYMI couls produce close to 1000 units per year. They may need qa competitor to help fill the industry need that will come on strong all at once. Any competition would not have a negative impact on CYMI for quite awhile, IMO.
An added bonus: all of the present DUV systems out in the industry have older lasers in them that probably were not upgraded during the down cycle in the industry. CYMI has 2 or 3 newer generations of excimer lasers that can replace the existing excimer lasers. I would expect to see an initial pop in retrofitting those systems already out in the field to bring them up to snuff with the most state of the art capability.
Now as fars a EUV is concerned, I will briefly highlight the changes in litho technology.
20 years ago - UV technology was ghi broad band with the knowledge and physics that this would die out at close to 2 micron.
Since then, g line and i line lithogrpahy broke the 1.5u, 1.0u and 0.50u dimensional barriers. All through that time, there was a great need to look beyond this technology to excimer(DUV), e-beam direct write, and x-ray lithography.
Well, presently, they finally found the real barrier at 0.18u, the place where the above technologies come into play finally. However, before you jump into EUV, e-beam and x-ray or SCALPEL, keep in mind that they have already demonstrated the feasibility of 0.08u feature sizes at the University of TX at Austin, thereby guaranteeing at least 3 or 4 new advanced device manufacturing technologies on the same exposure tool platform.
Gone are the days of the 500K to $2000K exposure tools. We are now looking at $7000K per system and you know they will get their money's worth. DUV has long life for this industry, at least another 3-5 years before worrying about EUV in full production. It took more than 10 years to get DUV mainstream and even with a compression in the technology roadmap, I would watch EUV from afar, just to keep an eye out over my shoulder.
Andrew
RadarView Financial Newsletter |
