[ibis-macro] Important survey for the future of IBIS - please respond
- From: "Muranyi, Arpad" <Arpad_Muranyi@xxxxxxxxxx>
- To: 'IBIS-ATM' <ibis-macro@xxxxxxxxxxxxx>, "mike@xxxxxxxxxxx" <mike@xxxxxxxxxxx>, Walter Katz <wkatz@xxxxxxxxxx>
- Date: Wed, 23 Oct 2013 18:35:07 +0000
Hello everyone,
The IBIS-ATM task group would like to conduct an informal survey
in order to be able to determine how we should go about defining
solutions for the package and on-die interconnect modeling needs.
We would like to encourage everyone to answer the questions below
so that we could get a good representation from the IBIS community
and the industry in general.
We will treat this survey as an informal vote, which means if
multiple responses are submitted from the same company, we will
only count the response once, and if conflicting answers are
submitted from different employees of the same company, we will
try to resolve the differences by personal email correspondence.
If you wish to keep your response confidential, please email to
to Mike LaBonte mike@xxxxxxxxxxx only, but if you don't mind
revealing who you are and what you think, you may also reply to
this email reflector.
Please review carefully and answer the 8 questions below with a
Yes, No or Abstain in the supplied spreadsheet at the end of this
e-mail. Make sure that the e-mail goes out as HTML, so that it
is easy to fill in the spreadsheet.
Note, that by the words "Component concept" in the questions we
are inferring that .ibs files would be enhanced to support an
alternative way of defining an IBIS Component that consists of
a package and a single die (or possibly stacked dice), but this
may not necessarily be achieved by modifying the [Component]
keyword.
Please help in steering the IBIS specification in the right direction
by sending your answers back to us.
Questions:
==========
#1) Shall the Component concept in IBIS be enhanced to support multiple
(stacked)
memory dice in a component?
#2) Shall the Component concept in IBIS be enhanced to support multiple
(stacked)
non-memory dice in a component?
#3) Shall the Component concept in IBIS be enhanced to support a different
number
of supply pins and die pads for a component?
#4) Shall the Component concept in IBIS be enhanced to support any one pin to
be
connected to multiple instances of the same buffer (on the same or
different dice)?
#5) Shall the Component concept in IBIS be enhanced to support any one pin to
be
connected to instances of different buffers (on the same or different
dice)?
#6) Shall the Component concept in IBIS be enhanced to support multiple pins
to be
connected to a single buffer?
#7) Shall there be a way for IBIS package and on-die interconnect models to be
associated
with IBIS [Model]s?
#8) Shall there be a way for the ports of crosstalk IBIS package and on-die
interconnect
models to be associated with IBIS Model Types Input (FEXT) and Output
(NEXT)?
Question
Yes
No
Abstain
1) Stacked Memory
2) Stacked Non-Memory
3) Supply # Pins != # Die Pads
4) 1 Pin --> N Identical Buffers
5) 1 Pin --> N Different Buffers
6) N Pins --> 1 Buffer
7) Interconnect Models by [Model]
8) Interconnect Models by Model_type
Your feedback is much appreciated.
Thanks,
Arpad
=====================================================================
Other related posts:
