[ibis-macro] Re: A technical question on UI
- From: "Walter Katz" <wkatz@xxxxxxxxxx>
- To: <radek_biernacki@xxxxxxxxxxx>, <Arpad_Muranyi@xxxxxxxxxx>, <ibis-macro@xxxxxxxxxxxxx>
- Date: Wed, 18 Apr 2012 19:02:31 -0400 (EDT)
All,
I agree with Arpad, and Radek. I think the following words should be added
to 5.1:
Any Model Specific Parameter of Type UI (this shall applied to values in
Table columns of Type UI) shall be a Time in units of bit_time. If this
parameter is (Usage In), then it shall be passed as is when passed as a
value to the DLL. If this parameter is (Usage Out), then it shall be
returned in units of bit_time when passed as a value from the DLL.
I hope we do not need a BIRD for this.
Walter
-----Original Message-----
From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of
radek_biernacki@xxxxxxxxxxx
Sent: Wednesday, April 18, 2012 6:53 PM
To: Arpad_Muranyi@xxxxxxxxxx; ibis-macro@xxxxxxxxxxxxx
Subject: [ibis-macro] Re: A technical question on UI
Hi Arpad,
I am as surprised by this interpretation as you are.
Actually, even for the reserved parameters I do not recall any requirement
to convert a value specified in terms of UI to absolute seconds. In other
words if the model creator designates a parameter as of type UI the model
should expect the fraction of UI (time normalized w.r.t. the bit time) as
its input.
Otherwise, why would we need Type UI at all?
Radek
-----Original Message-----
From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Muranyi, Arpad
Sent: Wednesday, April 18, 2012 3:09 PM
To: IBIS-ATM
Subject: [ibis-macro] A technical question on UI
Hello everyone,
An interesting question was raised in the Editorial Committee meeting this
morning, and I got the AR to write an email on this reflector to solicit
for some comments and feedback.
The question is revolving around parameters which are expressed in terms
of UI (unit interval). A statement was made in the Editorial Committee
meeting this morning, that if a parameter is a Reserved_Parameter, the
tool will know what to do with it because all Reserved_Parameters are
completely described in the specification.
However, if the parameter is a Model_Specific parameter, the EDA tool will
have to convert the value from UI to time and pass the time value to the
AMI executable model.
For example, if the bit time is 400e-12 (ps) and the UI value is
0.5 for a certain parameter, the EDA tool would have to pass
200e-12 to the DLL, and not 0.5.
(Please correct me if this summary is not accurate).
This surprised me, because I don't recall anything in the IBIS-AMI
specification that states that the EDA tool has to convert parameters
expressed in UI to time and pass the time value to the DLL. If I were
writing the simulation software, I would just pass 0.5 to the DLL.
Does anyone have any comments on how this is supposed to be handled by the
EDA tool?
Once we have consensus on that question, the next question will be whether
we can correct/clarify this as part of the Editorial Committee, or whether
a BIRD is necessary to address this question.
Thanks in advance for the comments,
Arpad
======================================================================
---------------------------------------------------------------------
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
---------------------------------------------------------------------
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
---------------------------------------------------------------------
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:
