[ibis-macro] Re: IBIS model for SerDes IO - It is not achoice between AMS and AMI, it is a choice between AMS and IV-VT curves
- From: "Muranyi, Arpad" <Arpad_Muranyi@xxxxxxxxxx>
- To: <ibis-macro@xxxxxxxxxxxxx>
- Date: Wed, 22 Oct 2008 06:41:55 -0700
Walter,
I am sorry, but I have to disagree.
You seem to (correctly) distinguish between the analog
(B-element) models and the digital algorithmic (AMI)
models in your response. However, while it is true
that AMI models do require an additional (separate)
analog model to go along with them so that the channel
response could be simulated with the analog Tx and Rx
models connected to the ends of the interconnect (to
get the correct reflection coefficients, i.e. the correct
channel response), this doesn't mean that in *-AMS you
couldn't write a SINGLE model that has an analog as well
as a digital description of the Tx and/or Rx to do all
of the above with a single model.
Both AMS languages (Verilog-AMS and VHDL-AMS) have mixed
signal capabilities, and the beauty of that is that in
the analog sections of the model you can describe what
you called the B-element (I-V and V-t curve based or
any other technique) and in the digital sections you
can describe what we have in the AMI models today, and
all this can be done in one model.
Of course, one could separate the analog and digital
portions of the model even in the *-AMS world, but I
can't think of any compelling reasons that this would
necessary.
Arpad
===========================================================
-----Original Message-----
From: Walter Katz [mailto:wkatz@xxxxxxxxxx]
Sent: Wednesday, October 22, 2008 8:28 AM
To: akshaye.sama@xxxxxx; Muranyi, Arpad
Cc: ibis-macro@xxxxxxxxxxxxx
Subject: RE: [ibis-macro] Re: IBIS model for SerDes IO - It is not achoice
between AMS and AMI, it is a choice between AMS and IV-VT curves
Akshaye,
There is no choice between AMI and AMS. The choice for a Tx (or Rx) is
whether to use IBIS "B" element (IV and VT curves), or an AMS model written
in Verilog or VHDL, or a Spice transistor subckt for the analog simulation
of the channel. AMI models describe how to process the analog waveform at
the receiver, independent if the analog waveform was generated using IBIS
"B" element, AMS Verilog or VHDL models, or Hspice models.
So the answer to your question is that an Rx or a Tx can have a "B" element
and an AMI model, or an AMS external model and an AMI model, or an Hspice
transistor model and an AMI model. Similarly, the TX can have a "B" element
and an AMI model, or an AMS external model and an AMI model, or an Hspice
transistor model and an AMI model.
Walter
---------------------------------------------------------------------
IBIS Macro website : http://www.eda.org/pub/ibis/macromodel_wip/
IBIS Macro reflector: //www.freelists.org/list/ibis-macro
To unsubscribe send an email:
To: ibis-macro-request@xxxxxxxxxxxxx
Subject: unsubscribe
Other related posts:
