[ibis-macro] Re: Parameter Tree Format with views
- From: Bob Ross <bob@xxxxxxxxxxxxx>
- To: wkatz@xxxxxxxxxx
- Date: Wed, 02 Jul 2008 11:58:03 -0700
Hi Walter:
First Great Work on your example sets. Now that I have looked
more closely at the examples, I really like aspects of your
proposal.
Also, the examples clarify some functional and syntactical
alternative. Arpad also proposed going with a standardized
VHDL-AMS template as another alternative.
[I have not yet looked at your IBIS-like Syntax, but we have
another option, each with the advantage (or disadvantage) of
looking familar in some context.]
-----
Here are my more important comments at the top-level, most of
which should be viewed as business or acceptance concerns, not
technical barriers.
1. SIMPLICITY MODE
Is there a "simple" mode? From one of the examples, I would
also want this as an available syntax to match my own net-lists:
(I do not like requiring R= or R even if some tools support this.)
R1 7 17 1.7k (a Berkley SPICE subset, although I do
C1 U7.3 0 1.3p not know if U7.3 is valid)
T7 12 27 Z0=50 Td=200p
W7 12 27 18 54 Len=Length File=‘50Ohm.rlgc’ Model=‘50Ohm’
s7 12 27 18 54 File=’50Ohm.s4p’
X12 12 27 subckt=xyz file=’xyz.mod’
This is why we discussed a Berkeley SPICE or Simple Spice subset
that now consists of RLCK, EFGH plus T and maybe a few other
elements. We will never prevent using V, I and a few other
Berleley or comment elements whether or not we endorse them.
Note the simplicity mode is the issue with VHDL-AMS and perhaps
some of other choices due to syntax overload for simple cases.
2. MODULE-CENTRIC (possibly Big module-centric)
A primary need is for a MODULE description language. The descriptive
example us very useful to illustrate a net-list application, but
there conceptually a disturbing mixture of nice robust modules along
with single-element netlist connections.
Then the real syntax would be (with no attempt at node number matching)
XIBIS_SPICE 1 2 3 4 5 IBIS_SPICE_MODULE conisting of the
R1 C1 T7, W7 in your example)
XS7 1 3 5 2 7 9 S-Parameter_MODULE (your s7 module)
X12 12 27 1 2 7 9 10 subckt=xyz file=’xyz.mod’
(Berkeley Spice multiline, multistage
connector model off the web consising
of RLC F elements)
Some folks would convert this via
simulation into a pole-zero or S-param
format "views: for simulation efficiency
at the module-level. So you could
re-cast the X12 Module with several Views.
Even though SPICEs do this, we are mixing simple elements (R, L, C, T, etc) and
with module oriented elements. We should still focus on a few modules. This
may impact "View" or hierarchy of View (Views within View if supported). The
practical priority should be the module View.
While I am not offering an alternative, I want the top-level syntactical
orientation to be module-centric. We may have a top-level interconnect
connect language with a large number of references to other files including
IBIS and the internal content.
(EBD uses the R,L,C discrete or distibuted T-line equivalents as the
interconnect structure with path connection (vs nodal). IBIS also
provides I/O and some series elements. So the EBD is element centric.)
-----
This is where I am coming from regarding some of my nitpick comments
or pushing for optional Berkeley Spice-level symplicity.
Bob
<your e-mails NOT included here>
--
Bob Ross
Teraspeed Consulting Group LLC Teraspeed Labs
121 North River Drive 13610 SW Harness Lane
Narragansett, RI 02882 Beaverton, OR 97008
401-284-1827 503-430-1065
http://www.teraspeed.com 503-246-8048 Direct
bob@xxxxxxxxxxxxx
Teraspeed is a registered service mark of Teraspeed Consulting Group LLC
---------------------------------------------------------------------
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:
