Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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 : Cymer (CYMI) -- Ignore unavailable to you. Want to Upgrade?

To: Maxwell who wrote (11564)	12/15/1997 4:47:00 PM
From: Czechsinthemail	Read Replies (4) \| Respond to of 25960

Maxwell, Thanks for the informative post on critical layers. One question that keeps coming up for me, particularly around DRAM, is the economics of 0.25 micron tools. Since we seem to be in an environment of weak DRAM prices, is there enough margin in the increased yield of 0.25 micron to support major investments in new equipment at this point? Much of the discussion swirling around the Korean companies seems to hinge around whether they can't afford the 0.25 equipment or whether they can't afford not to afford it. I'd appreciate any information you or others could contribute regarding the economies of 0.25 production, since I think that it is important in trying to get a sense of what marginal decisions to purchase or defer purchase semi companies are likely to make. Thanks, Baird

To: Maxwell who wrote (11564)	12/15/1997 8:35:00 PM
From: James Word	Respond to of 25960

Maxwell, Tulvio RE: Critical/Non-critical The definition of what constitutes a critical layer is dependent on the device being manufactured. The devices that my company works on have about 8 critical layers and maybe 10 to 12 non-critical. Overlay requirements in addition to linewidth determine the cut-off point. My company, as well as many others, buy tools specifically for handling non-critical layers, those with >0.8um CDs and >0.2um overlay tolerance. Most stepper manufacturers make a tool that is specifically designed for non-crits. Usually these tools can print 2 to 4 stepper fields in 1 shot, thus dramatically increasing throughput. BTW, I was the one who predicted decades of life for DUV. Here's why. Right now the definition for critical layer may be anything >0.8um linewidth. That is ~ 3X larger than current leading edge critical dimensions (CDs). What happens when leading edge CDs are <0.1um? Obviously 248nm DUV will not print those dimensions. But it will probably still be capable of printing the non-critical dimensions which will probably be anything >0.3um linewidths. Right now, Ultratech stepper is the leader in non-critical tools. Their leading edge tool is an i-line 2:1 stepper I believe. Who knows, maybe in 5-10 years their leading edge tool will have to be a 248nm DUV stepper or scanner, something that we consider state of the art by todays standards "The reason that the industry still using mix-match stuff is that they have to utilize the tools they already paid for. It only make sense." We do this too, utilizing our old g-line systems for extremely non-critical. They have a resolution of ~0.6um but the overlay is no better than 0.3um. Not too good. When we build a new fab, we will no longer buy g-line, but will buy some non-crit i-line tools. "In short term buying i-line along with .25um steppers is the best model. In longer term like 2-3 years down the road, buying all .25um excimer steppers is the best choice since these tools will be alive and well in the future. " It depends. Critical level i-line tools are ~$4mil. Non-critical i-line tools can be had for <$3mil. The cost advantage is even greater than 3:4 since the productivity of the non-critical tools is much greater than the critical tools. Remember, new fabs are being built all the time. They are going to be filled with NEW equipment, and many companies utilize the mix and match approach. This may be bad for Cymer short term since nobody is going to build an ALL DUV fab. But long term it is a good thing for them since DUV will probably still be used during the "X-Ray" (or whatever) generation in a mix and match capacity to reduce cost. James Word