MAGIC Magic Mailing List
 
 
From: Thomas Zanon (zanon AT ece DOT cmu.edu)
Date: Thu Nov 13 2003 - 00:23:05 EST

  • Next message: Benjamin Cherian: ""Extra tokens" error while compile on RH9" 

    Hi Magic developer,
 I'm using the latest revision (rev 56) of version 7.2. I noticed that 
if I use 'scalegrid 1 9' in a layout
using .18um tech file that fet dimensions scale nicely, but capacitances 
do not (using extract do all;
extract all; exttospice). Here the .ext and .spice files before an after 
using scalegrid 1 9:

.ext before using scalegrid:
timestamp 1068697975
version 7.2
tech scmos
style TSMC0.18um(tsmc18)from:t11b
scale 1000 1 9
resistclasses 6700 7500 929000 929000 1 7700 7700 80 80 80 70 70 40
node "ps" 11 0 31 4 pdif 0 0 24 20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0
node "pd" 9 52.284 22 4 pdif 0 0 42 26 0 0 0 0 0 0 0 0 0 0 60 32 0 0 0 0 
0 0 0 0 0 0
node "nd" 8 52.284 5 4 ndif 42 26 0 0 0 0 0 0 0 0 0 0 0 0 60 32 0 0 0 0 
0 0 0 0 0 0
node "ns" 10 0 -1 4 ndif 24 20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0
node "pg" 46 51.5 29 1 p 0 0 0 0 0 0 0 0 0 0 24 28 0 0 0 0 0 0 0 0 0 0 0 
0 0 0
node "ng" 46 51.5 3 1 p 0 0 0 0 0 0 0 0 0 0 24 28 0 0 0 0 0 0 0 0 0 0 0 
0 0 0
cap "nd" "pd" 123.714
fet pfet 29 4 30 5 12 16 "Vdd!" "pg" 4 0 "pd" 6 0 "ps" 6 0
fet nfet 3 4 4 5 12 16 "Gnd!" "ng" 4 0 "ns" 6 0 "nd" 6 0

.ext after using scalegrid:
timestamp 1068697975
version 7.2
tech scmos
style TSMC0.18um(tsmc18)from:t11b
scale 1000 1 1
resistclasses 6700 7500 929000 929000 1 7700 7700 80 80 80 70 70 40
node "ps" 11 0 279 36 pdif 0 0 1944 180 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0
node "pd" 9 52.284 198 36 pdif 0 0 3402 234 0 0 0 0 0 0 0 0 0 0 4860 288 
0 0 0 0 0 0 0 0 0 0
node "nd" 8 52.284 45 36 ndif 3402 234 0 0 0 0 0 0 0 0 0 0 0 0 4860 288 
0 0 0 0 0 0 0 0 0 0
node "ns" 10 0 -9 36 ndif 1944 180 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0
node "pg" 46 51.5 261 9 p 0 0 0 0 0 0 0 0 0 0 1944 252 0 0 0 0 0 0 0 0 0 
0 0 0 0 0
node "ng" 46 51.5 27 9 p 0 0 0 0 0 0 0 0 0 0 1944 252 0 0 0 0 0 0 0 0 0 
0 0 0 0 0
cap "nd" "pd" 13.746
fet pfet 261 36 262 37 972 144 "Vdd!" "pg" 36 0 "pd" 54 0 "ps" 54 0
fet nfet 27 36 28 37 972 144 "Gnd!" "ng" 36 0 "ns" 54 0 "nd" 54 0

lscale changes fro 9 to 1 and so do the lengths and areas. the cap of 
the cap between nd and pd
reduced to a 1/9th of the value before using scalegrid, but cscale in 
the .ext file stays at 1, thus the
final cap value in the spice file is not identical to the cap value w/o 
using scalegrid. Am I doing something
wrong here or is this a bug?
Thanks a lot for any feedback
Thomas

    •  
     
    Questions? Contact Rajit Manohar     		cornell logo