[SI-LIST] Re: Length matching of source synchronous busses.
- From: "Jeremy Plunkett" <jeremy@xxxxxxxxxxxxxxx>
- To: si-list@xxxxxxxxxxxxx
- Date: Tue, 11 Mar 2003 02:20:11 -0800
Sorry for the repeat post, just trying to see if I can get rid of those
silly equal signs by setting Outlook to plain text, no encoding...
-----Original Message-----
From: Jeremy Plunkett [mailto:jeremy@xxxxxxxxxxxxxxx]
Sent: Tuesday, March 11, 2003 1:55 AM
To: scott@xxxxxxxxxxxxx; si-list@xxxxxxxxxxxxx
Subject: RE: [SI-LIST] Re: Length matching of source synchronous busses.
Scott,
It may be that our in-house layout engineers are exceptionally talented (or
possibly exceptionally pliable), but I have yet to have any heavy objects
thrown at me for requiring a 25mil matching spec within a source-synchronous
group. And I can testify that they work under the same degree of time
pressure as our customer's layout people (if not greater).
Here's my perspective: the person setting the length matching spec tries to
find the "knee of the curve" where the additional gain in timing margin from
tightening the matching spec is no longer worth the additional time required
from the layout engineer. In discussions with the aforementioned
talented/pliable layout guys we have in-house, we concluded that once you
are going through the process of adding serpentines on pretty much every
signal in a bus(which is required for package length compensation in any
case), setting the length of the serpentines to match the total length
within 25mils is only slightly more time-consuming that adjusting them to
match within 100mils. In fact, in our experience getting them to match
within 5 mil is only slightly harder that 25 mils (and some of our customers
choose to do this); the reason we don't ask for it is that there isn't much
of a point, since as you and others have noted there are plenty of other
sources of delay variation in the board that are not reflected in the length
extraction such as the pad connections, fiberglass weave directions,
temperature gradients, resistor placement, etc etc.
If customers found this requirement to be especially irksome, I would expect
to receive some significant pushback on the subject, but that has not been
the case.
In situations where a matching requirement (or any part of our topology
recommendations) is making it difficult to route a bus, we do receive
feedback from our own layout engineers as well as customers and we are very
careful to take all factors into account in making the tradeoff.
regards,
Jeremy
------------------------------------------------------------------
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 archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
