[ibis-interconn] Comments on Sparse matirx, "Non-sampled represetnation...", and Poert information
- From: "Walter Katz" <wkatz@xxxxxxxxxx>
- To: "IBIS-Interconnect" <ibis-interconn@xxxxxxxxxxxxx>
- Date: Wed, 15 Jul 2009 09:30:07 -0400
All,
First, I think we can all agree that Sparse Format, and Non-Sampled
Representation, and Port information should be treated as three independent
enhancements to Touchstone.
Comments on Sparse Format:
There are some detailed syntactical issues to be decided, but the real open
issue is whether the data should be organized as either:
<Vector #> : <Row #, Column #> <Row #, Column #> <Row #, Column #>
or
<Row #> : <Column #, Vector #> <Column #, Vector #> <Column #, Vector #>
I strongly recommend the ?<Row #> : <Column #, Vector #>? simply because
this is the organization that engineers will be more comfortable with.
Comment on Non-Sampled Representation
There are multiple methods of representing a transfer function. I believe
that Vladimir?s is a good one. I also believe that there are other
mathematical representations of a transfer function that are also valid
(e.g. Impulse Response, Pulse Response, Step Response, ?).
Port Information
Just for review, Port Information consists of the following:
1. Port Name associated with each Port Number
2. Des.PinNumber associated with each Port Number
3. Connectivity associated with Port Number Pairs
4. Differential Pair association with Port Number Pairs
5. References to Port Number should be able to use Port Name and
Des.PinNumber aliases.
Walter
Walter Katz
303.449-2308
Mobile 720.333-1107
wkatz@xxxxxxxxxx
www.sisoft.com
Other related posts:
