X3T10 comment resolution for SSA-PH1

scheible at vnet.ibm.com scheible at vnet.ibm.com
Wed Feb 28 20:06:48 PST 1996


* From the SCSI Reflector, posted by:
* scheible at VNET.IBM.COM
*
Sorry for the duplication, copied to the SSA and SCSI reflectors.

Accredited Standards Committee
X3, Information Processing Systems
                                         Doc:  X3T10/96-131r0
                                         Date: February 28, 1996
                                         Project:  X3T10.1/1145D
                                         Ref Doc.: SSA-PH1 rev 8
                                         Reply to: John Scheible
To:      X3T10 Membership
From:    John Scheible
Subject: X3T10 comment resolution on SSA-PH1 rev 8 (96-008)

AMP Comments on SSA-PH1 attached to Yes ballot (transcribed by John
Lohmeyer)

AMP-1) Comment #1. pg. 6, Fig 3 Test Points, Node Port Driver M1
     Connector, Node Port Receiver M4 Connector
     M1 & M4 are not necessarily external connectors, therefore 2 more
     test points need to be identified at the external connectors.
     Test points are defined in the specification from the driver to
     the receiver & not at the external connectors.  Need to break up
     the specification & specify it at the external connectors because
     that's where all the external "cable assy houses" have control
     over.

     RESPONSE: Accepted as a rewrite of clause 7.5 (in rev 8a).  The
     test points specify the requirements of the entire Complex Port
     Connection, without a separate budget for the external cable.
     Add the following implementers note following paragraph 2 of
     clause 7.5.3 in rev 8a (Complex Port Connection).
     Implementer's note x: It is the responsibility of the system
     integrater to work with their suppliers to insure the Complex
     Port Connection meets the requirements of SSA-PH1.

     An explaination is as follows:

     The SSA-PH1 document does not specify a different requirement at
     the internal device connector from that at the external enclosure
     connector because there are means offered to use the same
     requirements at both connectors.  Unfortunately Rev 8 did not
     include these means explicitly but the intent is to use the same
     requirements for both the internal and external connectors (or
     any other connectors deemed to be interoperable junction points
     in the port connection path).

     In order to achieve this condition one either needs cables with
     negligible signal degradation between the internal and external
     connectors or one needs to use port connection coupler circuits
     (retiming repeaters for example) at the affected connectors.  The
     requirements on the cable assemblies are governed by performance
     specifications and multiple cable assemblies are allowed within
     the same port connection.  This means that SSA cable assemblies
     usually need to be better than the minimum requirements to allow
     for such complex connections.  A short internal cable assembly is
     a minor perturbation if it meets the transmission line properties
     specified for all cable assemblies.  The choice of external or
     internal connectors is arbitrary.  The implementer must choose
     the interoperable junction points in his market or system.

AMP-2) Figure 18
     The 5.62 dimension should go to datum F as called out in the IEEE
     1394 specification.  5.62 +/- 0.05 should go to datum F.
     RESPONSE: Accepted


Digital Comments on SSA-PH1 attached to No ballot

Digital voted "No" on the referenced letter ballot for a number of
technical and editorial reasons.  Even though Digital is intimately
involved in the editing of this document some critical content is not
present in rev 8.  Some of this content was formally approved by the
plenary for inclusion in the letter ballot.  Other content has
developed in recent SSA working groups and through other oversights.
The following changes were approved by the December 13, 1995 plenary
to be incorporated into SSA-PH1 Rev 8:
  1) Move connector performance requirements to normative annexes
  2) Add figures describing line segments, port connections, port
     connection segments, port connection couplers, and complex port
     connections
  3) Change line segment termination requirements to use 750 ps rise
     time instead of slew rates
  4) Move equalizer requirements to cable assembly section
Item 1 was incorporated in rev 8.
Item 2 was not done in Rev 8.  Item 2 constitutes important
architectural technical content concerning the physical link.
Item 3 was not done in rev 8.  Item 3 constitutes important technical
requirements for the line segment termination requirements.
Item 4 was not done in rev 8.  The statements relating to equalization
are technically in error in its present location in rev 8 under
complex port connections.
I will attempt to draft proposal wording and figures to address these
comments before the next plenary meeting.

RESPONSE: Accepted, Digital provided input to Rev 8a for all the
points not covered in Rev 8.

DEC-1) New technical information concerning the requirements for
     line segment termination has shown that the specifications under
     section 7.3 are not valid.  It is necessary to incorporate
     exceptions to the impedance limits for certain time duration's in
     order to allow existing devices to comply.

     RESPONSE: Accepted, Rev 8a contains the correct information

DEC-2) There are numerous instances where the terminology in Item 2
     is not incorporated into rev 8.

     RESPONSE: Accepted, Rev 8a incorporates these changes

DEC-3) Annex G does not contain the information developed in Botley
     relating to printed wiring board design.

     RESPONSE: Accepted, Include 96a126r0.

DEC-4) Line fault detection is not a required feature of SSA but is
     listed as an "essential feature" in section 1.1.

     RESPONSE: Accepted, Clause 1.1 has been rewritten in rev 8a to
     include only essential characteristics of SSA-PH1.

DEC-5) Several statements in section 1.1 are not supported by
     statements in the body of the document.  While these statements
     may be true for some implementations they are not "essential
     characteristics" of the technology described in SSA-PH1.

     RESPONSE: Accepted, Clause 1.1 has been rewritten in rev 8a to
     include only essential characteristics of SSA-PH1.

DEC-6) Changes are required to Figure 1 to extend the line driver
     and line receiver to the connector.

     RESPONSE: Accepted.  Figure 1 is modified in Rev 8a.

DEC 7) The difference between Table 4 and Table 5 is not clear.
     The terms "test" and "operating" need some explanation.

     RESPONSE: Accepted, clarifying text is added to rev 8a.

DEC-8) The difference between Table 7 and Table 8 is not clear.
     The terms "test" and "operating" need some explanation.

     RESPONSE: Accepted, clarifying text is added to rev 8a.

DEC-9) The colors need to be removed from fig 31.

     RESPONSE:  Accepted, rev 8a has a changed figure with the colors
     removed.

DEC-10)Figure C.4 needs to have the term "Table 30" removed.

     RESPONSE:  Accepted  rev 8a will have this figure changed.

DEC-11)There are several figures needed to accurately describe the
     terms in item 2 above.

     RESPONSE:  Accepted Rev 8a includes these figures.


FSI Comments on SSA-PH1 attached to No ballot

Gary Stephens faxed the following comment (transcribed by John
Lohmeyer):  "This standard provides too little functional improvement
to warrant its standardization.  The 40 MB/S version provides the
correct level with SSA-S3P as a base.  See note with SSA-TL1 and SSA-
S2P votes."  The note Gary refers to says that "due to severe personal
events, a proper typed and numbered response was not possible.  I will
attempt to do so before the March meeting."

     RESPONSE: Thank you for your input.  There are many companies
     that feel otherwise.


IBM Comments on SSA-PH1 attached to Yes ballot

IBM-1)    Some of the definition refinements from December meeting
were not included in the document.

     RESPONSE: Accepted see DEC comments for actions.


Milligan (Seagate) Comments on SSA-PH1 attached to No ballot

SEA-1) The patent statement has been useful information for the
     committee participants. However now that the SSA-PH1 is being
     forwarded, the patent statement should be replaced with the
     standard X3 patent statement for the case where patent claims
     have been made and offered in accordance with the ANSI patent
     policy. In particular the specific citing of claims should be
     removed.
     I recognize that X3T11 has left such statements in some of their
     forwarded standards. But this is inappropriate since the
     committee should not take any position on the validity of the
     claims made.

     RESPONSE: Accepted, new patent statement incorporated

SEA-2) If an ISO/IEC submittal is intended, it would be preferable
     (and necessary for the IS) to replace item (h) in Clause 2  and
     the ANSI/EIA citations in Tables B.1 through B.7 with equivalent
     IEC specifications.

     RESPONSE: Deferred, IEC 512 is a heavily modified version of
     ANSI/EIA 364.  Before going to ISO we plan to convert the
     ANSI/EIA references to IEC 512 references.  Since it is a non-
     trivial exercise, it will not be done prior to public review.

SEA-3) It is not acceptable to include "??" as normative reference.

     RESPONSE: Accepted, the proper references and titles will be
     added.  Also references in Table C.4 (BP3 and BP5) were incorrect
     and have been modified.


Editorial comments that are technical changes

EDI-1)    Corrections are needed to Annex C to clarify certain
connector tests.

RESPONSE: Make the following changes:
  1) Change 5000 insertions to 500 insertions in tests BP2, BP5 and
     DP1.
  2) Replace the maximum resferences of 35 m and 20 m to the following
     "Maximum 15 m change from initial contact resistence measured in Table
     C.2 test P3." in tests AP5, AP6, AP13, BP4, CP3 and DP3.
  3) Remove test BP5.5 in Table C.4
  4) Change part reference of BP3 to part 11 and add the document to
     the normative references.


Sincerely,

John Scheible
Voice:    (512) 823-8208
FAX:      (512) 823-0758
Email:    Scheible at vnet.ibm.com




More information about the T10 mailing list