Minutes of SPI-2 Working Group -- May 5, 1997

John Lohmeyer John.Lohmeyer at symbios.com
Mon May 12 15:27:26 PDT 1997

* From the SCSI Reflector (scsi at symbios.com), posted by:
* John Lohmeyer <John.Lohmeyer at Symbios.com>
Minutes of SPI-2 Working Group                                 T10/97-175r0

Accredited Standards Committee*
National Committee for Information Technology Standards (NCITS)

                                             Doc. No.: T10/97-175r0
                                                 Date: May 12, 1997
                                              Project: 1142-D
                                            Ref. Doc.:
                                             Reply to: John Lohmeyer

To:         Membership of T10

From:       Ralph Weber, Secretary T10
            Larry Lamers, Vice-chair T10
            John Lohmeyer, Chair T10

Subject:    Minutes of SPI-2 Working Group
            May 5, 1997 -- Natick, MA


1. Opening Remarks

2. Approval of Agenda

3. Attendance and Membership

4. Single-ended termination (96-245r2 and 97-174r0) [Wallace, Aloisi]

5. Proposed clarification to Fig. 24 (97-115r1) [Moore]

6. Integration Issues [Lamers]

7. Bus Set Delay Reduction (97-116) [Ham]

8. Hot-Plugging Data (97-144r0) [Ham]

9. REQ/ACK Glitch Filters [Ham]

10. LVD backplane testing (Fast-40 and Fast-80) [Ham]

11. Ultra3 Topics
   11.1 Requirements for Fast-100 operation (97-179r0) [Ham]
   11.2 Symmetrical Driver Margin Improvement (97-172r0, 97-178r0 slides 1-4 &
      6-12, 97-181r0) [Bastiani/Bridgewater]
   11.3 Driver Power Consumption (97-178r0 slide 5) [Bastiani]
   11.4 Cable Specification Changes [Lamers]
   11.5 First Pulse Compensation - Inter-Symbol Interference (ISI) [Lamers]
   11.6 Node Capacitance Reduction (97-181r0) [Bridgewater]
   11.7 Modeling Assumptions [Schneider]
   11.8 Figure 25 Timing for Ultra-3 [Bastiani]

12. Meeting Schedule

13. Adjournment

                              Results of Meeting

1.    Opening Remarks

John Lohmeyer, the T10 Chair, called the Cinco de Mayo meeting to order at
9:00 a.m., Monday May 5, 1997.  He thanked Charles Monia of Digital Equipment
for hosting and arranging the meeting.

As is customary, the people attending introduced themselves and a copy of the
attendance list was circulated.

2.    Approval of Agenda

The agenda was approved with the following additions and changes:

      4. Changing driver modes when hot plugging (96-270r1) [Penokie] - was

3.    Attendance and Membership

Attendance at working group meetings does not count toward minimum attendance
requirements for T10 membership.  Working group meetings are open to any
person or organization directly and materially affected by T10's scope of
work.  The following people attended the meeting:

         Name          S        Organization         Electronic Mail Address
---------------------- -- ------------------------- -------------------------
Mr. Vincent Bastiani   A# Adaptec, Inc.             bastiani at corp.adaptec.com
Mr. Lawrence J. Lamers A  Adaptec, Inc.             ljlamers at aol.com
Mr. Wally Bridgewater  V  Adaptec, Inc.             wally at eng.adaptec.com
Mr. Richard Wagner     P  Cable Design Technologies rwagner at montrose-cdt.com
Mr. Siegfried Schmalz  R  Dallas Semiconductor      siegfried.schmalz at dalsemi
Mr. Greg McSorley      P  Data General / Clariion   greg_mcsorley at dgc.ceo.dg.
Dr. William Ham        A# Digital Equipment Corp.   ham at subsys.enet.dec.com
Mr. Chris Nieves       P  Fujitsu                   cnieves at fcpa.fujitsu.com
Mr. Don Vohar          A# Fujitsu (FCPA)            dvohar at fcpa.fujitsu.com
Mr. George Penokie     P  IBM Corp.                 gop at rchvmp3.vnet.ibm.com
Mr. Dean Wallace       P  Linfinity Micro           75671.3443 at compuserve.com
Mr. Louis Grantham     A  Linfinity Micro           lgdatcom at ix.netcom.com
Mr. Alan Littlewood    O  LSI Logic                 alanl at lsil.com
Mr. Skip Jones         P  QLogic Corp.              sk_jones at qlc.com
Mr. Ting Li Chan       A  QLogic Corp.              t_chan at qlc.com
Mr. James McGrath      P  Quantum Corp.             JMCGRATH at QNTM.COM
Mr. Richard Uber       V  Quantum Corp.             duber at tdh.qntm.com
Mr. Dana Hall          V  Quantum Corp.             dhall at tdh.qntm.com
Mr. Gene Milligan      P  Seagate Technology        Gene_Milligan at notes.
Mr. Gerald Houlder     A  Seagate Technology        Gerry_Houlder at notes.
Mr. Daniel (Dan) F.    O  Seagate Technology        daniel_smith at notes.
Smith                                               seagate.com
Mr. Dave Guss          P  Silicon Systems, Inc.     dave.guss at tus.ssi1.com
Mr. Vit Novak          A  Sun Microsystems, Inc.    vit.novak at sun.com
Mr. John Lohmeyer      P  Symbios Logic Inc.        john.lohmeyer at symbios.com
Mr. Ralph O. Weber     A  Symbios Logic Inc.        roweber at acm.org
Mr. Kenneth J. Hallam  P  UNISYS Corporation        ken.hallam at mv.unisys.com
Mr. Paul D. Aloisi     P  Unitrode Corporation      aloisi at unitrode.com
Mr. Gregory Kapraun    V  Western Digital Corp.     kapraun at wdroc.wdc.com

28 People Present

Status Key:  P    -  Principal
             A,A# -  Alternate
             O    -  Observer
             L    -  Liaison
             V    -  Visitor

4.    Single-ended termination (96-245r2 and 97-174r0) [Wallace, Aloisi]

In reviewing document 97-174r0, Paul Aloisi called the group's attention power
dissipation requirements for single-ended terminators (1.44 watts for narrow
and 1.9 watts for wide).  Several questions were raised regarding whether such
requirements can appear in a standard, the definition of dissipated power, and
other issues.

Paul, Bill Ham, Dean Wallace, and Louis Grantham discussed the Single-Ended
Terminator IV curve to be included in SPI-2.  Gene Milligan noted that he
started this discussion several meetings ago by requesting that the SCSI-2
terminator be included in SPI-2.  The group discussed many aspects of the IV

At one point, Dean Wallace characterized a difference of opinion between
Unitrode and Linfinity regarding the width of the allowable region in the IV
curve, noting that Linfinity was more lenient than Unitrode.  This led to a
discussion of writing tighter specifications for drivers and whether the group
intends to go faster with single-ended.  Several IV curves were compared.

Dean and Paul agreed to take the advice of the group, conduct some additional
testing of terminators presently sold on the general market, and produce a
final-form draft proposal (to be numbered 96-245r3) for new terminator
requirements and changes to existing terminator requirements in SPI-2 that
includes specific wording (instead of statements of principles).

The group agreed that the next working group meeting will consider the node
capacitance measuring scheme for measuring the node capacitance at another
voltage, because the current scheme does not fully account for the possibility
that highest capacitance might exist at a voltage other than the one used in
the current measuring scheme.

5.    Proposed clarification to Fig. 24 (97-115r1) [Moore]

Larry Lamers reported that Richard Moore says 30 mV will give much better
interoperability than 60 mV and that the previously agreed changes in wording
are necessary to avoid misinterpretations.  John Lohmeyer noted that the April
meeting agreed to make changes that include 60 mV and the addition of words
clarifying how signals depicted by the figure are to be tested.  Bill Ham
expressed the belief that the additional wording is editorial, because it
expresses the long-held opinion of the group.  John was concerned that the
existence of keywords such as "may" and "shall" in the additional wording made
the changes arguably substantive.  Larry agreed to work the issues off-line
and to ask for agenda time at future meetings only if additional discussion is
necessary.  Otherwise, the group recommended that the plenary adopt the
changes in Figure 24 as agreed at the April meeting.

6.    Integration Issues [Lamers]

Gene Milligan noted that the group is losing track of what integration issues
have reached agreement regarding changes in SPI-2.  He noted that no formal
list has been maintained, except possibly in meeting minutes.

John Lohmeyer noted that the plenary motion taken nine months ago to integrate
SCSI-2, SPI, Fast-20, and SIP has not been addressed.  The committee discussed
how to address the plenary requirements.  The group also discussed how best to
handle the growing list of Ultra3 discussion topics.  Gene Milligan stated his
objection to approving a SPI-3 project proposal prior to near completion of
the integrated SPI-2 document.

7.    Bus Set Delay Reduction (97-116) [Ham]

Bill Ham expressed the belief that previous meetings had agreed to this
change.  Larry Lamers reviewed his disagreement with the proposal.  The
working group recommended that the plenary vote on 97-116 (if the document can
be found).

8.    Hot-Plugging Data (97-144r0) [Ham]

Bill Ham reported that he has not collected any additional data on hot
plugging.  Bill discussed his hot plugging data with Wally Bridgewater.  Bill
agreed that this item should be removed from future agendas.

9.    REQ/ACK Glitch Filters [Ham]

Bill Ham agreed to provide a detailed written proposal.  George Penokie asked
that this agenda item be kept until he can get the specific information to his
engineers and collect their responses.

10.   LVD backplane testing (Fast-40 and Fast-80) [Ham]

Bill Ham asked that this item be removed from future agendas.

11.   Ultra3 Topics

11.1  Requirements for Fast-100 operation (97-179r0) [Ham]

Bill Ham presented a list of problems with operation at 100 mega-transfers per
second and showed some signal traces taken with pseudo-random data.  He
expressed surprise that the results with pseudo-random data were worse than
the results from `worst-case' patterns in some of his experiments.  Bill noted
that for the first time, ISI issues are becoming more pronounced.  He plans to
do future testing with pseudo-random test patterns.

Bill pointed out a low-frequency sinusoidal waveform superimposed on the
signal that he was at a loss to explain.  Dean Wallace said that this was
aliasing from the terminator.  Dean expressed a desire to give a more specific
description of the problem, but he felt that more testing and research will be
needed to add detail to the description.

During the discussion, Bill noted that he selected Fast-100 because his Fast-
80 tests showed no serious problems.  Bill concluded that transfer rates above
Fast-80 have serious problems because the cable is not intended for such high
frequencies.  Gene Milligan, John Lohmeyer and George Penokie suggested that
several other approaches could be considered, such as: other types of
termination, equalization, and encoding (all of these are techniques used by
existing serial interfaces).  Using both edges of the REQ/ACK signals was also
discussed, and Bill noted that such a change would necessarily result in the
waveform being more symmetric.

The group discussed encoding schemes and the challenges of encoding several
parallel lines.  Some felt that encoding would not present serious
difficulties for silicon designers.  Others felt that encoding would overly
difficult to fit in precious chip real estate.  Two encoding schemes were
mentioned; 8b/10b and 4b/5b. The group spent some time conducting self-
education regarding the mechanics, implications, advantages, and disadvantages
of encoding.  John asked group members to begin discussing encoding with their
engineering teams.

Bill presented the proposal that faster speeds be attained using regular
pulses and using both edges of the REQ/ACK pulses.  The group discussed the
meaning of Bill's proposal.  Bill expressed confidence that Fast-100 would be
possible based solely on switching to using both edges.

The group discussed placing new requirements on host software to recognize
that devices may report an ability to operate at speeds higher than the cable-
plant will really support.  It is possible that both the host and the target
will be capable of operating at higher speeds but that high-speed operation is
somehow inhibited by the cable or other interconnect components.  In such
cases, the host should negotiate a lower speed with the device.  This process
was compared to the fall-back speed negotiation process used by modems.  It
was recognized that a failure to communicate at the higher speed will often
appear to be a bus hang during a data phase- this condition may not be easy to
detect and recover.

11.2  Symmetrical Driver Margin Improvement (97-172r0, 97-178r0 slides 1-4
& 6-
12, 97-181r0) [Bastiani/Bridgewater]

Vince Bastiani presented cable loss data (97-178r0 slides 1-4).  Questions
were raised regarding nature of the loads used in the multiple loads data.
Also, the group discussed potential reasons for the loss, e.g. skin effects,
stranded verses solid conductor, and of course, just plain cable attenuation.

Vince then presented analysis of driver and receiver properties based on the
boundary conditions established for several design properties in SPI-2 (97-
172r0).  Concerns were raised about how the model considers only the
fundamental signal and not the harmonics.  However, there was agreement that
losses can present significant problems for faster transfer rates.

Wally Bridgewater presented empirical data concerning signal loss on a biased
bus (97-178r0 slides 6-12).  After discussing the data, Bill Ham repeated his
conclusion that operating in an environment with high attenuation
is undesirable.

Wally presented a list of ten differences between symmetric and asymmetric
signaling (97-181r0), which the group discussed.

11.3  Driver Power Consumption (97-178r0 slide 5) [Bastiani]

Vince Bastiani presented information supporting the conclusion that better
signals can be produced with less power in an unbiased bus.  Questions were
raised regarding the dependence of the results on the tolerance around the
common mode voltage.

11.4  Cable Specification Changes [Lamers]

The group discussed changes in the cable that might be beneficial for faster
transfer rates (particularly lower attenuation cables), but eventually agreed
that proposals should be sought from cable vendors.

11.5  First Pulse Compensation - Inter-Symbol Interference (ISI) [Lamers]

Larry Lamers noted his desire that ISI issues be discussed.  He asked that
this item be kept on the agenda for the next meeting.

11.6  Node Capacitance Reduction (97-181r0) [Bridgewater]

Wally Bridgewater presented model and analog circuit information to support
his contention that a dual receiver (one biased and one unbiased) can be built
that has only a 0.5 pF increase in capacitance over the unbiased receivers
presently in use.

Concerns were raised regarding the effect of switchable terminators on node
capacitance. The possibility was discussed of changing system requirements to
eliminate the need (or possibility) for switchable terminators on disk drives.

11.7  Modeling Assumptions [Schneider]

Several people asked that the spice model presented by Frank Gasparik be
distributed to the group via e-mail.  John promised to investigate meeting the
request.  Questions were raised about whether the model addresses high
frequency signaling issues.

11.8  Figure 25 Timing for Ultra-3 [Bastiani]

Vince Bastiani asked that this item be deferred to the next meeting.

12.   Meeting Schedule

The next general SPI-2 working group meeting is scheduled for Monday July 14,
1997 in Colorado Springs, CO.

Larry Lamers noted that he will request the scheduling of a meeting to discuss
parallel SCSI protocol-specific topics in late June.

The group agreed to add the following items to the agenda for the July

      Capacitance Measurement at 2 Volts [Wallace/Ham]
      Universal Backplane Annex (9x-yyyrz) [Wallace]
      Cable Company's Presentations for Low-Attenuation Cables [`Ham']

13.   Adjournment

The meeting was adjourned at 5:30 p.m. on Monday May 5, 1997.

John Lohmeyer                 E-Mail: john.lohmeyer at symbios.com
Symbios Logic Inc.             Voice: 719-533-7560
4420 ArrowsWest Dr.              Fax: 719-533-7036
Colo Spgs, CO 80907-3444    SCSI BBS: 719-533-7950 300--28800 baud

* For SCSI Reflector information, send a message with
* 'info scsi' (no quotes) in the message body to majordomo at symbios.com

More information about the T10 mailing list