[ibis-macro] Re: What is a [Model], I now understand the issue
- From: "Muranyi, Arpad" <Arpad_Muranyi@xxxxxxxxxx>
- To: IBIS-ATM <ibis-macro@xxxxxxxxxxxxx>
- Date: Fri, 11 Jan 2013 04:20:06 +0000
Walter,
I don't think the solution is that obvious. There are other options
to consider aside from defining the analog models in the .ami file.
The [Model] or [External Model] keywords could potentially be modified
or a new [AMI Model] keyword could be invented for the .ibs file, or
the AMI DLL could also carry its own analog model definition (as it
was pointed out by Greg's question not too long ago). We have yet to
discuss the pros and cons of each of these possibilities (or any other
ones we didn't think of yet) in order to determine the most preferable
solution...
My presentation in the last ATM meeting was basically a "problem statement".
We have yet to come up with proposals on how we can solve the problem.
I would like to encourage everyone to think about possible solutions
and the pros and cons of each so we could find our way out of this
situation.
Thanks,
Arpad
=======================================================================
From: ibis-macro-bounce@xxxxxxxxxxxxx [mailto:ibis-macro-bounce@xxxxxxxxxxxxx]
On Behalf Of Walter Katz
Sent: Thursday, January 10, 2013 7:54 PM
To: IBIS-ATM
Subject: [ibis-macro] What is a [Model], I now understand the issue
All,
Here is the difference between Arpad's understanding of a [Model] and my
understanding of a [Model].
Arpad believes a [Model] is essentially a subckt with ports PuRef, PdRef,
PcRef, GcRef, Pad, Z (for input), Stimulus (for Output), Enable (for I/O and
3-state), and the IV and VT curves are implemented in a proprietary simulator
as it chooses. (e.g. the HSPICE B element).
I believe that a [Model] contains the results of specific measurements (lab or
simulation), that a simulator can choose to use in any way.
I also agree that the common usage of IBIS [Model]s have been in accordance
with Arpad's view. If we want to proceed with the notion that common usage
defines and constrains what [Model] is, then of course representing 4 port AMI
analog models in [Model] is definitely incorrect. This applies both to adding a
subparameter Tstonefile in [Model], or using a 4 port IBIS-ISS subckt in
[External Model].
We must accept the requirement that IC Vendors are commonly requiring 4 port
Touchstone models. So where in IBIS is this information to be stored?
The solution is obvious. Since these 4 port models are limited to AMI models,
and they cannot fit into [Model] or [External Model], then it is only natural
to define them inside of the .ami file!
Walter
Walter Katz
wkatz@xxxxxxxxxx<mailto:wkatz@xxxxxxxxxx>
Phone 303.449-2308
Mobile 303.335-6156
Other related posts:
