From: Richard A. Schumacher (email@example.com)
Date: Thu Nov 18 1999 - 15:21:17 PST
Of course. The difference is that there's no umbrella SPICE
organization. Everyone understands that "SPICE" is not a brand
name and that any SPICE model could be junk. IBIS, on the other
hand, is being held up as an imprimatur of usefulness, which
surely must imply correctness.
If the notion of IBIS is at all worthwhile, then vendors should
not be allowed to call a model an IBIS model unless and until
they are willing to attest to or demonstrate the model's
correctness, at least in some minimal set of defined situations.
Passing the golden parser and having every pin correctly
simulate the vendor's own standard load for that pin would be
a good start.
> The same can also be true for SPICE models.
> Arpad Muranyi
> Intel Corporation
> -----Original Message-----
> From: Richard A. Schumacher [mailto:firstname.lastname@example.org]
> Sent: Thursday, November 18, 1999 1:20 PM
> To: email@example.com
> Subject: Re: FW: [SI-LIST] : IBIS datasheets for PCI and DDR
> Well, not necessarily when the vendor's IBIS model is flawed.
> This can occur in such a way that the error doesn't turn up
> until a board is routed and timed, or even until a prototype
> is built and tested. Better no model at all (which causes
> one sig int engineer to work on building one) than one which
> causes a dozen designers to propagate an error.
> Perhaps vendors should be prohibited from using the IBIS name
> unless they certify the accuracy of their model or attest
> to it with measurements. Until then, Caveat Emptor.
**** To unsubscribe from si-list: send e-mail to firstname.lastname@example.org. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****
This archive was generated by hypermail 2b29 : Tue Feb 29 2000 - 11:39:00 PST