[SI-LIST] Re: Correction: Re: IBIS-AMI Vendor Support Help

I would also agree. Non-linear should be accounted for. 

For the end-users, we also need to know all the assumptions under which
a model has been developed and which EDA tools support what
functionality of AMI (Current and upcoming specs).

Tks
Syed

-----Original Message-----
From: Dmitriev-Zdorov, Vladimir
[mailto:vladimir_dmitriev-zdorov@xxxxxxxxxx] 
Sent: Tuesday, February 09, 2010 10:19 AM
To: Syed Huq (huqs); si-list@xxxxxxxxxxxxx
Subject: RE: [SI-LIST] Re: Correction: Re: IBIS-AMI Vendor Support Help

Hi Syed,

Agree. We just have been debating this at DesignCon. Our point is that
support for non-linear buffers should be added as an option.
 
Vladimir


-----Original Message-----
From: Syed Huq (huqs) [mailto:huqs@xxxxxxxxx] 
Sent: Tuesday, February 09, 2010 11:10 AM
To: Dmitriev-Zdorov, Vladimir; si-list@xxxxxxxxxxxxx
Subject: RE: [SI-LIST] Re: Correction: Re: IBIS-AMI Vendor Support Help
Importance: Low

Hi Vladimir,

I believe that the assumption made where the Tx and Rx Analog model is
linear (I/V curves are straight lines and do not change in time) may be
true for one semiconductor vendor but may not hold ground for many
others. In that case, the modeling approach is flawed to begin with.

-Syed

-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
On Behalf Of Dmitriev-Zdorov, Vladimir
Sent: Tuesday, February 09, 2010 9:57 AM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: Correction: Re: IBIS-AMI Vendor Support Help


Syed,

A minor addition. The procedure outlined below (generating step and
impulse from detailed SPICE level simulation) provides an accurate
response that AMI models and EDA platform will update and then use in
time domain simulation by way of convolution. At the simulation stage,
we can use complete non-linear models of Tx/Rx analog buffers.

However, a validity of superposition (by convolution) of the generated
responses assumes that Tx and Rx analog models are linear and time
invariant (e.g. buffers' IV curves are straight lines and do not change
in time)

Vladimir 


------------------------------------------------------------------
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:
http://www.freelists.org/webpage/si-list

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


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

List archives are viewable at:     
                http://www.freelists.org/archives/si-list
 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu

Other related posts: