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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: andreas_wonisch who wrote (71773)	2/17/2002 2:51:54 PM
From: fyodor_	Respond to of 275872

Thanks for the link. -fyo

To: andreas_wonisch who wrote (71773)	2/17/2002 5:56:15 PM
From: dale_laroy	Respond to of 275872

Thanks for the link. I feel that a few things should be considered. >He (Deppe) felt Fab 30 was transitioning well to 0.13-micron using phase shift masks with the existing ASML 248-nanometer exposure tools that were installed from the beginning of the plant. "In fact, we are hoping we can get to 0.09-micron (processing) using phase shift masks with our current lithography tools," he added.< AMD was confident they could get to 0.13-micron using phase shift masks with ASML 248-nanometer exposure tools. Shortly after they started the transition to 0.18-micron, they anticipated being able to beat Intel to 130nm because SVG's lack of 248nm optics with the required numerical aperature would force Intel to use 193nm lasers at the 130nm node. AMD claimed they could get to 130nm in early 2001 using ASML's phase shifting optics. Later in 2000, after Fab30 began to ramp, AMD stated that the transition to 130nm would be delayed because of the need to use 193nm lasers for some of the mask steps. It now seems that after about a one year delay versus what they initially claimed, they do not need any 193nm lasers after all. As I see it, the major differences between then and now is that then they were "confident" of being able to reach the 130nm node with 248nm laser tools, while now they are "hoping" to reach 90nm using 248nm tools. On the plus side, AMD's lack of confidence could mean that they have better formulated contingency plans should they be unable to reach 90nm using 248nm laser tools. On the negative side, if they really are counting on reaching 90nm using 248nm tools the delay versus when they are claiming to be able to transition to 90nm could be greater than the one year delay getting to 130nm. >AMD has no target date when new 157-nm wavelength lithography tools would be introduced into Dresden. If the present exposure tools can be extended to 0.09-micron processing using phase shift masks, then perhaps only test equipment at 193-nm might be installed. Deppe said AMD hopes 193-nm lithography using phase shift masks could then be extended to 0.06-micron processing "That would mean we could skip 157-nm tools entirely, which have been targeted for 0.07 and 0.06-micron (processing)," he added. Like most technologists, Deppe also believes Extreme Ultraviolet (EUV) Next Generation Lithography will be needed for the 0.045-micron semiconductor node.< Let us see now. Like most technologists Deppe believes that EUV Next Generation Lithography will be needed for the 45nm node. This would imply that the 157nm tools will not be adequate for the 45nm node. So it would seem that reaching the 65nm node using 193nm tools would be a long shot. And, even if it is possible to reach the 65nm node using 193nm tools, this would be end of the line for Fab30 because the cleanroom at Fab30 is not adequate for the 45nm node even if it could be reached without EUV equipment. AMD currently thinks that Fab30 will be able to reach the 65nm node, but they are unsure because "the 65nm node may need to use EUV tools". I am quite certain that Fab30 could be upgraded to EUV tools if needed, so it would seem a different variable is the real reason for AMD's uncertainty. Let us reorganize and analyse the following paragraph: >AMD has no target date when new 157-nm wavelength lithography tools would be introduced into Dresden. If the present exposure tools can be extended to 0.09-micron processing using phase shift masks, then perhaps only test equipment at 193-nm might be installed. Deppe said AMD hopes 193-nm lithography using phase shift masks could then be extended to 0.06-micron processing< Addressing the third sentence first we have: >Deppe said that AMD hopes 193-nm lithography using phase shift masks could then be extended to 0.06-micron processing< We have two options here. AMD could be hoping to be able to use inexpensive 193nm tools to equip the 300mm JV fab, or AMD could be hoping to use 193nm tools at Fab30 to get it to 65nm processing. Stepping back a sentence to: >If the present exposure tools can be extended to 0.09-micron processing using phase shift masks, then perhaps only test equipment at 193-nm might be installed.< The prior sentence established that this is talking about Dresden, so this establishes that, if the present tools can be extended to 90nm, AMD probably will not be installing 193nm tools at Fab30, which means no 65nm production using 193nm tools at Dresden. Moving back to: >AMD has no target date when new 157-nm wavelength lithography tools would be introduced into Dresden.< This is a clear indication that the possibility of Dresden being upgraded to 157nm lithography tools exists, otherwise the statement would be to the effect that 157nm lithography tools will not be used at Dresden. Instead, they have no target date. The best interpretation of this is that, if 248nm lithography tools are not adequate for 90nm production, AMD will be installing 157nm lithography tools to insure that they can get down to 65nm at Fab30. And, most likely these 157nm lithography tools will be installed at the 90nm node. Thus we have two potential scenarios: A) Fab30 can be upgraded to 90nm production using 248nm lithography equipment, in which case Fab30 will not be upgraded to 65nm production. Also, in this case AMD will be targeting using inexpensive 193nm lithography tools for the initial ramp of the 300mm JV fab at 65nm. B) Fab30 can not be upgraded to 90nm production using 248nm lithography equipment, in which case it will be upgraded to 90nm production using 157nm lithography equipment and later be upgraded to 65nm production using 157nm lithography equipment. Also, in this case the 300mm JV fab would probably be ramped more slowly (not a problem with Fab30 converted to 65nm production) using more expensive EUV equipment.