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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Yousef who wrote (31711)	4/11/1998 1:03:00 PM
From: Ali Chen	Read Replies (1) \| Respond to of 1573683

Yosef, <I HAVE been able to hit the gigahertz range with these specs.> For a 1-gate device, right? Ali

To: Yousef who wrote (31711)	4/12/1998 10:24:00 PM
From: Maxwell	Respond to of 1573683

Yousef: <<This is absolutely WRONG ... A FET and backend interconnect can be simply modeled as a voltage variable current source with a series Resistance and parallel Capacitance ... simple RC circuit driven by a current source. In this circuit, Cpacitance is what determines how high the frequency can go. This assumption is valid as long as the FET does not become de-biased (Vd stays > Vdsat).>> The capacitance in your model is actually 2 types of capacitance lumped together. One is a contact capacitance which occurs everytime there is a junction with contact resistance. You can change this by using different processes such as copper dual damascene versus aluminum tungsten vias. The other capacitance is due to nearby metal lines, intra-level and inter-level. You can do a simple model by taking 2 metal lines with spacing L. Send a signal of frequency F onto one wire and look at the induced signal on the other wire. Pick a certain threshold which you call noise and find F where the induced signal becomes significant. That frequency will be a function of L and K where K is the dielectric constant of the insulator between the 2 metal lines. Once you do it you will be surprised what F is. F(L,K) <<I HAVE been able to hit the gigahertz range with these specs.>> Aluminum metal lines with conventional k=4 oxide and metal line spacing of 0.4um will have no problem hitting 1GHz. Thus I don't doubt your specs. When you get in the 2GHz and beyond watch out. Maxwell