IT/02-569r

Accredited Standards Committee, INCITS, Information Technology
INCITS Secretariat, Information Technology Industry Council (ITI)
1250 Eye St. NW, Suite 200, Washington, DC 20005
Telephone 202-737-8888; Fax 202-638-4922; Email: incits@itic.org
Date: July 25, 2002
Project: 1249-D
Ref. Doc:  None
Reply to:  Deborah J. Donovan, INCITS Secretariat
Phone: 202/626-5746
Email: ddonovan@itic.org

Register of Public Review Comments on INCITS 338, Information Technology - High - Performance Parallel Interface - 6400 Mbit/s Optical Specification (HIPPI-6400-OPT).  The 3rd public review period is from  June 7, 2002 to July 22, 2002.


If the Technical Committee action is to accept in whole or in part a proposal contained in the comment, the changes should be sent to the Coordinator of Standards Processing together with any TC comments supporting the change. If the TC action is to reject in whole or in part proposals contained in the comment, the response should provide the rationale for the rejection.

The comment should be discussed at the next TC meeting, and if not definitively responded to at once, an interim acknowledgment should be sent along with an estimated date of action.

The final response to the commentor must include the provision that the commentor has twenty working days from the postmark of the technical committees response to indicate, in a written statement, acceptance or rejection of the TC response.


Public Review Comment #1 for 3rd Public Review of INCITS 338

From:  Brown Gair D DLVA  [BrownGD@NSWC.NAVY.MIL]
Sent:  Monday, July 1, 2002 5:28 PM
To:  Donovan, Deborah; 'psa@ansi.org'
Cc:  'Don Tolmie'
Subject:  IT - High-Performance Parallel Interface - 6400 Mbit/s Optical Specifications (HIPPI-6400-OPT)
 

All,

The following comments are submitted on the ballot of the High-Performance
Parallel Interface - 6400 Mbit/s Optical Specifications (HIPPI-6400-OPT).
These comments are provided from the Telecommunications Industry Association
(TIA) Subcommittee FO 2.2.

If you have any questions please contact me.

Thank you for the opportunity to participate in this standardization effort.

Gair Brown
Chair, TIA FO 2.2

****************************************************************************
 

Technical Comments

Table 4 - An encircled flux specification requirement on the transmitter
should be added.  Recommended values are

 "Encircled Flux: < 25% @ a radius of 4.5 um
   > 75% @ a radius of 15 um"

These transmitter specifications are in harmony with those for Gigabit
Ethernet transceivers and will allow improved link performance at 300 m and
longer distances.   Refer to TIA-TSB-62-20 for additional information.
 
 

Table 4 - Add a specification for RML fiber bandwidth.  An overfill
bandwidth specification is not sufficient to guarantee link performance for
62.5 um multimode optical fiber used with VCSEL type optical sources.  The
RML bandwidth specification has been accepted as a new multimode fiber
specification parameter, which when coupled with transmitter encircled flux
specifications, leads to enhanced and predictable multimode fiber optic link
performance.  A RML bandwidth of 385 MHz-km can be specified for 62.5-micron
fiber.  Refer to TIA-TSB-62-20 for additional information.
 
 

Section 6 -  In multiple places, minimum receiver sensitivity is referred
to, which is no longer defined in this document.  The reference should be to
minimum optical modulation amplitude.  The last paragraph also refers solely
to Table 4.  This reference should be to Table 4 and Table 5.
 
 

Section 8.1 - Rewrite this section to refer to the RML bandwidth.  The RML
bandwidth specification, when combined with an encircled flux specification
for an optical transmitter, leads to enhanced and predictable multimode
fiber optic link performance.
 
 

Annex A section A.2 - The use of a statistical loss budgeting approach for
links with low numbers of components is not valid.  It is not clear that
most implementations of this standard will include enough connections to
enable the valid use of statistical loss budgeting methods.  Recommend that
statistical loss budgeting be removed from the document or that warnings be
included that statistical loss budgeting methods are not considered valid
for links containing less than 5 items.  If the transmitter encircled flux
specification and the RML bandwidth specification are incorporated into the
document, the optical power penalties will drastically decrease, which would
allow the use of worst case power budgets in this section.
 
 

Annex B section B.3 - There are several concerns with the method used to
measure cable skew.   Most notably are the following:

  1. The test method is not consistent with some
methods/equipment currently being utilized to characterize this parameter in
the industry.
  2.     The use of an overfill launch condition combines the
"inherent" cable skew with differential modal delay effects.  Since
differential modal delay effects are source launch condition dependent, skew
results measured using this method will not correlate across different
transmission technologies utilizing different launch conditions.

There are a great number of technical issues that are inherently involved in
the choice of a multimode optical fiber cable skew measurement method.  TIA
FO 2.2 has initiated work to develop a standard multimode optical fiber
cable skew measurement method.  A draft measurement procedure will likely be
discussed at the June FO 2.2 meeting.  If you would like the assistance of
TIA FO 2.2 in developing a possible replacement to section B.3 please
contact the Chair of FO 2.2 - Mr. Gair Brown (540-653-1579).
 
 

Editorial comments
 

1) 3.3: Gbit/s, Mbit/s, , psec (elsewhere, e.g., in 8.3.2 and 8.3.3), GBaud,
and MBaud violate SI terminology. The correct symbols are Gb/s, Mb/s, ps,
etc.
(we understand your desire to clarify the meanings, but it is not acceptable
engineering practice. Megabits, for example, is either Mb or megabits.
Other abbreviations are incorrect.)

2) The title should therefore be modified accordingly.

3) 3.1.34 "optional": This adjective is defined as a noun. Write either
"optional characteristics" as the term being defined, or else preface your
current definition with "Referring to".

4) Hyphenate compound adjectives like "data dependent" (3.1.20).

5) 3.1.39: Requires "that", not "which", being restrictive.
 
 
 

***********************************************************************
Naval Surface Warfare Center           browngd@nswc.navy.mil
Code B35                                        phone: 540-653-1579
17320 Dahlgren Road                       fax: 540-653-8673
Building 1500 Room 107
Dahlgren, VA 22448-5100

https://fiberoptics.nswc.navy.mil