Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!cmcl2!nrl-cmf!ames!ucbcad!ucbvax!XEROX.COM!"Nahid_Ansari.ElSegundo"
From: "Nahid_Ansari.ElSegundo"@XEROX.COM
Newsgroups: comp.lsi
Subject: SPICE2G6 BUG
Message-ID: <871106-173908-1924@Xerox>
Date: Fri, 6-Nov-87 18:06:18 EST
Article-I.D.: Xerox.871106-173908-1924
Posted: Fri Nov  6 18:06:18 1987
Date-Received: Sun, 8-Nov-87 23:34:20 EST
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The ARPA Internet
Lines: 16


Regarding the difference in narrow channel effect between the Berkeley
version of SPICE2G6 and that of University of Washington,  we at Xerox
have also found the Berkeley version to be incorrect.  The narrow
channel factor should be  *inversely* dependent on the  *channel width*.
Therefore, the correct equation is:

     FACTOR=0.125D0*FNARRW*TWOPI*EPSSIL/ (COX*XW)

instead of

     FACTOR=0.125D0*FNARRW*TWOPI*EPSSIL/COX*XL


Nahid Ansari
Xerox, El Segundo