There are two common definitions of FEXT and NEXT. One definition of FEXT
is that the aggressor signal goes in the same directions as the victim
signal, and NEXT the aggressor goes in the opposite of the victim signal.
IEEE 802.3 has a different definition of FEXT and NEXT.
In 802.3, an aggressor is FEXT if the Tx is on the same chip and the same
I/O buffer model as the victim Tx and the aggressor Rx is on the same chip
and the same I/O buffer model as the victim Rx. If an aggressor is not
FEXT, then it is NEXT. Therefor in 802.3 FEXT aggressors are generally
synchronous and have the same amplitude and equalization as their victims,
while NEXT aggressors can have significantly different magnitude, phase
and direction as their victims.
Walter
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx]
On Behalf Of jun zhang
Sent: Wednesday, December 23, 2015 8:13 PM
To: Richard Allred <richard.allred@xxxxxxxxx>; si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: a question about the spacing differences between TX
to TX distance and TX to RX distance
Thank you very much for your reply.
Besides, for Serdes, sometimes, the total link will go across several
brds, cables. Two lanes may go parallel close to each other on one brd,
but not be neighbours in the other brds. Then how could we judge their
FEXT and NEXT risks?
Regards
On Wed, Dec 23, 2015 at 11:27 PM, Richard Allred
<richard.allred@xxxxxxxxx>
wrote:
FEXT xtalk is typically synchronous and thus has a fixed phaseposition.
relation while NEXT is asynchronous so it does not have fixed relative
This is quite clear if you apply peak distortion analysis to the link.wrote:
Richard Allred
On Wed, Dec 23, 2015 at 8:00 AM, jun zhang <zhangjun5960@xxxxxxxxx>
sender.
Hi Richard,
but Serdes lane has no requirement of length match and therefore FEXT
can also be in the same relative position in the victim eye, I think.
Regards
On Wed, Dec 23, 2015 at 11:13 AM, Richard Allred <
richard.allred@xxxxxxxxx> wrote:
I think the reason for the difference in the spacing is two fold.
One, NEXT is usually stronger than FEXT since it isn't attenuated as
much by the channel and two, the xtalk phase relation.
For two DQ signals in the same byte group, they are always going the
same direction and their edge transitions have a somewhat fixed
phase relation to each other. The crosstalk will always be in the
same relative position in the victim eye.This is far end xtalk (FEXT).
For two signals that go opposite directions, the phase relation of
the edge transitions are not fixed. This means that this near end
crosstalk can attack any position in the eye and so on average is
worse than a fixed phase relation.
If you allow me, I would say it is because:
FEXT is proportional to (Kc - Kl) while NEXT is proportional to
Kc+Kl where Kc=Capacitive coupling coeficient and Kl=Inductive
coupling coeficient.
Even more, NEXT prevails during all agressor signal travel so I
understad it is more sensitive than FEXT.
Lets see what more experienced people say.
Ivan Perino
On 22 dic 2015, at 22:25, Carson Au <carsona@xxxxxxxxxxxx> wrote:it
NEXT and FEXT.
This email is subject to copyright and the information in it is
confidential. This e-mail, its content and any files transmitted
with
are intended solely for the addressee/s and may be legallyaddressee/s
privileged and/or confidential. Access by any other person other
than the
is unauthorized without the express written permission of the
If
you have received this e-mail in error notify the sendermay
immediately by email, do not use the email or any attachment or
disclose them to any person, delete the email from your system and
destroy all copies you
have printed. Metamako LP does not guarantee that any email orattachment
is secure or free from viruses or other defects.wrote:
On 23 December 2015 at 12:17, jun zhang <zhangjun5960@xxxxxxxxx>
distance
I have a question about the spacing differences between TX to TX
whereand TX to RX distance. For example, in Intel PDG, TX to RX
distance requirement is stricter than TX to TX requirement. For
example, for SATA-Gen3, TX to TX: 7h for microstrip; TX to RX:
9h for microstrip
thinkh is the height of the mcrostrip layer.
If FEXT for TX to TX is -20db and NEXT for TX to RX is also
-20db, I
------------------------------------------------------------------their noise levels are the same. Then why does Intel give
stricter requirements for the distance between TX to RX?
--
best wishes,
Jun Zhang
-----------------------------------------------------------------
-
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject
field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject
field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject
field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject
field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
--
best wishes,
Jun Zhang
