This could get into a religious discussion. It boils down to personal
opinion, belief, and some emotion. With that said, here is my opinion.
1 - keep a separate directory tree for development and testing from
2 - my development tree is organized as working/vendor_name/part/buffer.
The buffer level might not be needed if the vendor gave you the IBIS file.
3 - my published directory is for general consumption, so it is organized by
function. I.e. si_lib/function/part.ibs. If you plan to have a very
extensive library, you might want to add another layer of directories.
Using IS, I can see two very good ways to deal with published IBIS files. A)
set your ICX_IBIS_SEARCH_PATH to all the si_lib/function directories, or B)
each design engineer can copy all the IBIS files that he is using to a
directory in the design directory tree and set the ICX_IBIS_SEARCH_PATH to
that one directory.
4 - make sure you have a good way to keep IBIS files in the same directory
with the EBD files that call them. On unix, you could use links very
effectively to keep one master copy of the IBIS file for easy maintenance.
I'd be interested to know what you come up with. It might change my IBIS
Sent: Wednesday, 01 September, 1999 6:15 PM
Subject: [SI-LIST] : IBIS Model Libraries
I have been given the task of starting and supporting a
process. The questions my supervisor wanted answered before
we made any
significant structural changes or additions in terms of data
or man power
How do other companies that use signal integrity
particular) organize their libraries of IBIS models?
What kinds of processes do they go through to obtain and
models and the information associated with them i.e. keeping
track of paths
to parts and which internal part number that models works
Is there some software package that does that?
Any enlightenment on either what is a good method or what is
a poor method
would be helpful. I have been finding that I spend a great
deal of time
determining what part an internal number is associated with
and then trying
to find the model if it even exists and then more time if it
creating that model. All of which spans many different
groups with many
different managers, different budgets, and different
**** To unsubscribe from si-list: send e-mail to
firstname.lastname@example.org. In the BODY of message put: UNSUBSCRIBE
si-list, for more help, put HELP. si-list archives are accessible at
**** To unsubscribe from si-list: send e-mail to email@example.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****