Arpad,
Thanks for the close read.
I will wait for others to comment and then send out a revised version.
Walter
Walter Katz
<mailto:wkatz@xxxxxxxxxx> wkatz@xxxxxxxxxx
Phone 303.449-2308
Mobile 303.335-6156
From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Muranyi, Arpad
Sent: Tuesday, September 12, 2017 6:19 PM
To: IBIS-ATM <ibis-macro@xxxxxxxxxxxxx>
Subject: [ibis-macro] Re: BIRD 158.6_draft6
Walter,
Thanks for preparing this draft. I have a few comments:
I am not sure what item #2 refers to In the "ANY OTHER BACKGROUND
INFORMATION",
but if it refers to the text that you just removed, we should get rid of
it.
The text under the Tx model drawing states that "the reserved parameter
Tx_R is optional".
While this is true, it does NOT say what happens if Tx_R is absent. I
know this is defined under
the parameter's description later, but I think we should state it here
that in this case a short
is assumed, so that the reader wouldn't be left hanging or forced to dig
through the rest of
the spec to find out.
Also, the text under the Tx drawing does not state what the meaning of the
small triangle
symbol is. The closest we come to those is "and the common reference node
Ref", but
it would be nice to state that these symbolize node0.
The same two comments also apply to the text below the Rx drawing. We
should state
what happens when the optional Rx_R is missing (open circuit), and might
not hurt to
say again that the small triangles represent connections to node0.
The last sentence in that paragraph doesn't sound right to me:
"For an Rx buffer, the receiver circuit defines the analog buffer model
between the
buffer terminals and a high impedance probe at the input to the Rx
Algorithmic model."
As far as I know, the Rx Algorithmic model's input is the channel's
impulse response.
A probe on the output of the Rx Touchstone model does not necessarily give
us an IR
that can be "probed" by the Algorithmic model directly. I think this
should be worded
in a different way, something along these lines:
"For an Rx buffer, the receiver circuit defines the analog buffer model
between the
buffer terminals and the high impedance input of the first signal
processing block of
the device."
We don't say anywhere what the slope of the stimulus should be for the
voltage
sources in the Tx model (if someone would choose to use these models to
generate
a step response, or in the [External Model] situation). We need to state
that this
should be the steepest possible edge, i.e. an ideal step function. I
think we mention
this somewhere else in the AMI portions of the spec, so you could maybe
refer to
that, or copy the same wording here.
Could you please incorporate these comments into the BIRD draft?
Thanks,
Arpad
=====================================================================
From: ibis-macro-bounce@xxxxxxxxxxxxx
<mailto:ibis-macro-bounce@xxxxxxxxxxxxx>
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Walter Katz
Sent: Tuesday, September 12, 2017 3:30 PM
To: IBIS-ATM <ibis-macro@xxxxxxxxxxxxx <mailto:ibis-macro@xxxxxxxxxxxxx> >
Subject: [ibis-macro] BIRD 158.6_draft6
All,
This BIRD_158.6_draft6.docx is based my BIRD_158.6_draft4.docx with
figures copied from Radek's BIRD_158.6_draft5.docx. I also removed an
additional paragraph that referenced Ts4file_Boundary.
Walter
Walter Katz
<mailto:wkatz@xxxxxxxxxx> wkatz@xxxxxxxxxx
Phone 303.449-2308
Mobile 303.335-6156