# SAS 3.0 Question

Newman, Harvey Harvey.Newman at lsi.com
Wed Apr 2 12:49:27 PDT 2014

```Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1404021_f.htm">HTML-formatted message</a>
Attachment #1: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1404021_image001.png">image001.png</a>

Hi Paul,
Regards,
Harvey Newman
LSI Corporation
Operations & Central Engineering - SerDes Architecture
(408) 433-4028
From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf Of Paul
Wassenberg
Sent: Wednesday, April 02, 2014 11:45 AM
To: T10 at t10.org
Subject: SAS 3.0 Question
Hello T10 members,
We have a question regarding the Tx EQ coefficients request response as shown
in the table below. We understand that the test is to measure the voltage
variation according to the coefficients request as Vx(k+1)-Vx(k), where x is
HL, 1, 2 and 3.
However, there is difference in opinion within Marvell on how to interpret
Vx(k+1)-Vx(k).
Group 1 says: Vx(k+1)-Vx(k) is the voltage variation of Vx where k represents
the value before the coefficient request and k+1 represents the value after
after the coefficient request. Therefore, each coeff request has to be
compared with its prior case. For example, Dec/Dec/Hold request should be
compared with Hold/Hold/Hold request and Inc/Inc/Hold request should be
compared with Dec/Dec/Hold request and so on.
Group 2 says: Vx(k+1)-Vx(k) is the voltage variation of Vx where k represents
the initial condition and k+1 represents the value after the coefficient
request. Therefore, based on the Table 45 below, every single coeff request
has to be compared with initial condition, which is Hold/Hold/Hold request.
For example, Dec/Dec/Hold request should be compared with Hold/Hold/Hold
request and Inc/Inc/Hold request should be compared with Hold/Hold/Hold
request and so on.
Who is correct?
Thanks,
Paul Wassenberg
Marvell
[cid:image001.png at 01CF4E71.F9F49360]

```