( ESNUG 355 Item 5 ) --------------------------------------------- [7/26/00]
From: John Lee <jolly@avanticorp.com>
Subject: ( DAC 00 #46 ) The Anon User On Parasitic Extraction Out-To-Lunch
John -
That one anon review on parasitic extraction ( Item 46 in the DAC'00 Trip
Report) was terrible! I have to imagine that this person isn't active in
the backend world --- too many strangely wrong comments and thoughts. Have
someone who is in the backend read it, and I'm sure they'll laugh or cry.
But thanks for the great web site and mailer! I definitely got a lot out of
the other reviews.
For the record, our Avanti party was definitely the best, but perhaps that
is because I had 1 beer too many.
- John Lee
Avanti Fremont, CA
---- ---- ---- ---- ---- ---- ----
From: Hank Walker <walker@cs.tamu.edu>
John,
Nobody does full-chip 3-D extraction. What they do is in effect pattern
recognition and stitch together some 3-D results, maybe with parametric
tweaks. This inherently has an accuracy problem at the boundaries of
patterns. There are true 3-D capacitance extractors around now that are
100x faster than FastCap, which itself is much faster than Raphael, with
only a small error. These are not fast enough for full-chip extraction,
but certainly for cell extraction or maybe segments of a critical net or
regular array (e.g. memory).
- Hank Walker
Texas A&M University College Station, TX
|
|
