[SI-LIST] use of PVT corner models for various speed-grades

  • From: "Zabinski, Patrick" <zabinski.patrick@xxxxxxxx>
  • To: <si-list@xxxxxxxxxxxxx>
  • Date: Thu, 30 Nov 2006 13:01:30 -0600

We're analyzing SI aspects of a handful of commercial FPGAs, and we're
running into some difficulties with the available PVT (process, voltage,

temperature) models.  I'd like to hear how folks deal with these issues.

First some background....

When selecting one of the latest FPGAs, they are binned in different
speed grades ranging from slow to nominal to fast (each vendor
has its own numbering system for speed grades, so I'll use these
generic terms).  Theoretically, the "fast" parts are selected through
testing as having the fastest performance.  As such, we can infer
that they have the fastest process relative to the other speed grade
parts.  Similarly for the nominal and slow grade parts.

In contrast, the models from the FPGA vendors include three PVT corners
per IBIS spec: minimum, typical, and maximum.   These models represent
the entire span of process, voltage, and temperature.  As such, the=20
minimum models represent the slowest-grade process, lowest-voltage,
and highest temp.  At the other end, the maximum models represent the
fastest-grade process, highest-voltage, and lowest temp.

Now the challenge...

For our application, we are using the fastest speed grade parts, which
would suggest the parts have the fastest process.  And we will
be operating the parts over the full voltage and temperature ranges.

Yet, models are not provided that cover the "fast" process and
full range of voltage and temp.

The models we have are IBIS based and encrypted HSPICE.  As such,
we are not able to manually control the PVT parameters, because
they are either implied through the models or buried within
the encrypted libraries.

How are you dealing with this disparity between available models
and the limited-process window of the parts?

Does any chip vendor offer PVT corner models specific to speed grades?
Beyond FPGAs, the memories we're using are in a similar situation.  It
is that common to get just one set of models to represent all possible
speed grades of parts? =20

Thanks,
Pat



------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List FAQ wiki page is located at:
                http://si-list.org/wiki/wiki.pl?Si-List_FAQ

List technical documents are available at:
                http://www.si-list.org

List archives are viewable at:     
                //www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu
  

Other related posts:

  • » [SI-LIST] use of PVT corner models for various speed-grades