The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Other CAD Tools >> Physical Verification, Extraction and Analysis >> Diva vs Dracula (which is more accurate ??)
https://designers-guide.org/forum/YaBB.pl?num=1147167746

Message started by Faisal on May 9^th, 2006, 2:42am

Title: Diva vs Dracula (which is more accurate ??)
Post by Faisal on May 9^th, 2006, 2:42am

Hi,

I am running DRC, LVS checks with Diva on my blocks and it indicates that there are no errors. I then do my post layout simulation with analog extracted netlist and the results are correct. However when I run DRACULA on the chip level, I find many errors e.g. unmatched device, unmatched node etc

What could be the possible reasons ?? Which tool is more trust worthy ?? Is this safe for tapeout ??

Regards,
Faisal

Title: Re: Diva vs Dracula (which is more accurate ??)
Post by bernd on May 9^th, 2006, 3:15am

1. All physical verification tools were using input files
so called 'command rules' written by humans.
As a result the result of your DRC run depends on the
command rule file used by the tool.

2a. Have you compared the top level result DIVA vs. DRACULA.
Your only talking about that you have check block level with
DIVA and top level with DRACULA.
Maybe the errors occurred are real errors introduced on top.
2b. DRACULA is using for DRC and LVS GDSII layout database and
CDL netlist whereas DIVA runs direct on Cadence DFII database.
Maybe there is an translation error which causes the errors.

3. Both tools are not longer state of the art, but this also
depends on the technology you are using, e.g. nobody will
use one of them below 130nm CMOS in the industry.

4. I would not do a tapeout if I don't know the errors and
have a reasonable explanation for them.

Bernd