( ESNUG 433 Item 6 ) -------------------------------------------- [10/20/04]
Subject: ( SNUG 04 #17 ) Nassda HSIM vs. Avanti HSPICE Power Discrepancies
> I've found the accuracy of timing numbers obtained from Nassda HSIM is
> usually within 5% of difference compared with HSPICE. Its power numbers
> have a larger discrepancy compared to other tools, especially the peak
> current numbers. But I don't know which one is closer to correct answers.
> My leakage simulation show pretty close results as HSPICE. My concern is
> that the memory bitline separation accuracy is sometimes not good enough.
> It could report BL separation more than 10 mV different from HSPICE (which
> is a big deal to us.) But I didn't dig into this to find out the reason.
> HSIM has a lot of accuracy control parameters. Maybe I can get better
> results by adjusting those parameters.
>
> - Dechang Sun
> LSI Logic
From: Reynaldo Hernandez <reynaldo=user domain=nassda spot calm>
Hi John,
Regarding the bitline separation issue, Dechang may find increasing the
localized precision in the sense amp will help improve the correlation to
HSPICE. Basic commands like HSIMALLOWEDDV and HSIMANALOG can be used. If
current precision is critical, HSIMENHANCEIDDQ, a new command in HSIM 5.0,
can also be used. An example:
.hsimparam subckt=my_sense_amp
+ hsimenhanceiddq=1
+ hsimanalog=2
+ hsimalloweddv=0.01
HSIM, like all simulators, can trade off precision and performance. I
suggest the following to address his peak current concerns. When doing
instantaneous current analysis (printing/measurements) use the
".param HSIMLUMPCAP=0" command. This command makes the simulator bypass
lumping together certain capacitors for faster simulation, and can improve
the precision of the peak current values related to vdd or gnd type nodes.
Note: with or without this command, the precision of the average current
shouldn't change; only minor variations in peak current analysis should be
seen.
- Reynaldo Hernandez
Nassda Corporation Santa Clara, CA
|
|
