Re: Vih, Vil, Vmeas etc..

From: D. C. Sessions <dc.sessions@tempe.vlsi.com>
Date: Wed Mar 25 1998 - 07:24:58 PST

JEAN-CHRISTOPHE_PAUTRAT@HP-France-om5.om.hp.com wrote:

> The [Model] statement only allows 1 value for Vih, Vil, Vmeas, Cref,
> Vref, Rref. Whereas the PCI or AGP, for example, specify Vih,Vil &
> Vmeas as a ratio of the power supply. Also, the test conditions can be
> different between rising and falling edges (exple : Cref=10pF for all
> AGP max timings and 0pF for min timings).

AGP inputs are actually differential, and that's how they should be
characterized.

> I think ibis spec should include typ/min/max statement for all these
> values plus a 2 test load conditions for rising and falling edges.

As for the test loads, capacitive test loads are evil. The whole idea
of IBIS is that if you know what the driver characteristics really are
you can derive the behavior for a wide range of loads.

-- 
D. C. Sessions
dc.sessions@tempe.vlsi.com
 
Received on Wed Mar 25 07:28:40 1998

This archive was generated by hypermail 2.1.8 : Fri Jun 03 2011 - 09:53:46 PDT