Minutes of SPI-2 and EPI Study Group Meeting -- 9/11/95

Lohmeyer, John JLOHMEYE at cosmpdaero.ftcollinsco.ncr.com
Tue Sep 19 15:48:00 PDT 1995

Minutes of SPI-2 and EPI Study Group meeting in Bedford, NH 

Accredited Standards Committee*
X3, Information Technology
                                                   Doc. No.: X3T10/95-316r0
                                                       Date: September 19, 
                                                  Ref. Doc.:
                                                   Reply to: John Lohmeyer

To:         Membership of X3T10

From:       John Lohmeyer, Chair X3T10
            Larry Lamers, ViceChair X3T10

Subject:    Minutes of SPI-2 and EPI Study Group Meeting
            Bedford, NH -- September 11, 1995


1. Opening Remarks

2. Approval of Agenda

3. Attendance and Membership

4. Review  of  Results from August 14th meeting in Denver,  CO  (95-309r0)

5. Low Voltage / Low Power Differential Interface [Ham/Aloisi]

6. Hybrid Signaling? [McGrath]

7. Terminator Power Distribution [Aloisi/Ham]

8. Propagation Delay Length Limit [Ham]

9. Pin-out Definitions [Ham]

10. SPI-2 Architecture [Ham]

11. EPI Draft Document Outline (95-320) [Ham]

12. 50 Ohm Cable Specification (95-317r0) [McCall]

13. Should RST Signal Remain Single-Ended? [Gardner]

14. +/- Signal Skew [Jander]

15. Preliminary Test Results [Ham] (95-315)

16. Meeting Schedule

17. Adjournment

                              Results of Meeting

1.    Opening Remarks

John Lohmeyer, the X3T10 Chair, called the meeting to order at 9:00 a.m.,
Monday September 11, 1995.  He thanked Charles Monia of Digital Equipment
Corp. for hosting the meeting.

As is customary, the people attending introduced themselves and a copy of 
attendance list was circulated.  Copies of the draft agenda and general
information on X3T10 were made available to those attending.

2.    Approval of Agenda

The following items were deleted from the agenda.
8) Determining Differential vs. Singled-ended [Aloisi/Ham]

The following items were added to the agenda.
15) Skew on +/- signal [Jander]
16) Preliminary Testing results [Ham]

The agenda was approved with the changes noted above.

3.    Attendance and Membership

 Attendance at study group meetings does not count toward minimum attendance
 requirements for X3T10 membership.  Study group meetings are open to any
 person or organization directly and materially affected by X3T10's scope of

 The following people attended the meeting:

         Name          S        Organization         Electronic Mail Address
 ---------------------- -- ------------------------- 
Mr. Norm Harris        P  Adaptec, Inc.             nharris at eng.adaptec.com
Mr. Lawrence J. Lamers A# Adaptec, Inc.             ljlamers at aol.com
Mr. Edward Fong        P  Amdahl Corp.              esf10 at amail.amdahl.com
Mr. Bob Atkinson       O  AMP, Inc.                 rdatkins at amp.com
Mr. Wayne E. Werner    V  AT&T                      wew at aluxs.att.com
Mr. Dennis R. Haynes   P  Burr-Brown Corp.          haynes_dennis at bbrown.com
Mr. Justin McEldowney  V  Burr-Brown Corp. 
         mceldowney_justin at bbrown.
Mr. Chris Millsaps     V  BusLogic                  chrism at buslogic.com
Mr. Louis Grantham     P  Dallas Semiconductor      grantham at dalsemi.com
Mr. Charles Monia      P  Digital Equipment Corp.   monia at shr.dec.com
Dr. William Ham        A# Digital Equipment Corp.   ham at subsys.enet.dec.com
Ms. Nancy Cheng        A# Hitachi Computer Products n_cheng at hitachi.com
Mr. Duncan Penman      P  IIX Consulting            penman at netcom.com
Mr. Dean Wallace       P  Linfinity Micro 
          75671.3443 at compuserve.com
Mr. Peter Gossler      O  NSM Jukebox GmbH 
         73503.3467 at compuserve.com
Dr. Akira James Miura  A  Panasonic Technologies,   miura at tadw.research.
                          Inc                       panasonic.com
Mr. Param Panesar      O  Pioneer Research
Mr. Skip Jones         P  QLogic Corp.              sk_jones at qlc.com
Mr. James McGrath      P  Quantum Corp.             JMCGRATH at QNTM.COM
Mr. Edward A. Gardner  A  Quantum Corp.             gardner at acm.org
Mr. David McCall       V  Quantum Corp.             dmccall at tdh.qntm.com
Mr. Richard Uber       V  Quantum Corp.             duber at tdh.qntm.com
Mr. Gene Milligan      P  Seagate Technology        Gene_Milligan at notes.
Mr. Brian N. Davis     A# Seagate Technology 
       brian_davis at notes.seagate
Mr. Gerald Houlder     A  Seagate Technology        Gerry_Houlder at notes.
Mr. Greg Alvey         V  Solution Technology       g.alvey at genie.geis.com
Mr. Robert N. Snively  P  Sun Microsystems Computer bob.snively at eng.sun.com
Mr. John Lohmeyer      P  Symbios Logic Inc. 
       john.lohmeyer at symbios.com
Mr. Rod DeKoning       V  Symbios Logic Inc.        rod.dekoning at symbios.com
Mr. Mark Jander        V  Symbios Logic Inc.        mark.jander at symbios.com
Mr. Tracy Spitler      V  Symbios Logic Inc. 
       tracy.spitler at symbios.com
Mr. Arlan P. Stone     A  UNISYS Corporation 
       arlan.stone at mv.unisys.com
Mr. Paul D. Aloisi     P  Unitrode Integrated       Aloisi at uicc.com
Mr. Matt Thomas        A  Unitrode Integrated       thomas at uicc.com
Mr. Tak Asami          A  Western Digital           asami at dt.wdc.com

35 People Present

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

4.    Review of Results from August 14th meeting in Denver, CO (95-309r0)

John went over the results of the LVDS signaling meeting held in Denver on
August 14th, highlighting the agreements reached.  The minutes are contained
in 95-309r0.  As is often the case, people not present for the Denver 
questioned several of the agreements reached there.  Several topics were re-

The revised power budget was apparently not documented, yet.

There was concern expressed that single-ended Fast-40 was eliminated from
consideration.  The reasons for not documenting single ended Fast-40 are 1)
too much effort required to document it; 2) the universal driver probably
cannot run at 40 megatransfers single-ended; and 3) it would likely have
extremely restrictive cable lengths and device counts.

There were objections to this approach, since the industry is likely to
develop single-ended Fast-40 whether or not it is documented; the topic was
deferred to a reflector discussion and a possible vote at the November 

5.    Low Voltage / Low Power Differential Interface [Ham/Aloisi]

Bill Ham reviewed his proposal for LVDS.  A more correct model of the
terminator was developed, the previous model turned out to be erroneous.

Mark Jander proposed that the Fast-80 hold times be reduced by 0.5 ns which
would result in 6.0 ns transmitter hold time and 2.0 ns receiver hold time.
This will simplify transmitter design be reducing the hold time to less than
50% of the cycle time (12.5 ns).  There was no objection to Mark's proposal
and Bill agreed to revise the SPI-2 electrical interface document.

6.    Hybrid Signaling? [McGrath]

Hybrid signaling is using single-ended for the low-speed control signals 
BSY, SEL, ATN, C/D, I/O, and MSG) and LVDS for REQ, ACK, data and parity.
This would save 7 pins on protocol chips by using one pin/signal on the 
signals instead of two pins/signal.

This was a controversial topic.  The Denver study group had rejected the
concept, but it was brought up again by its proponents (who had not attended
the Denver meeting).

Bill Ham noted that allowing a hybrid solution would sacrifice common mode
noise rejection.  Frank Gasparik sent a foil to the meeting with Tracy 
illustrating that near-end crosstalk would be an issue as the higher power
single-ended signals couple into the LVDS signals.  The crosstalk foil 
controversial because several people argued that flat ribbon cable should 
be used with any kind of differential.  Others felt that flat ribbon cable
would get used anyway.  Gene Milligan suggested that the single-ended 
should use a new low-voltage single-ended, which had not been discussed yet
(and was not discussed further at the meeting).

John Lohmeyer was concerned that the architectural precedent of a hybrid
interface would be present a roadblock to future enhancements.  It would be
very difficult, if not impossible, to migrate to a pure LVDS interface after
setting a hybrid precedent -- we have no mechanism to sense the difference.

Can we live with the extra 7 pins?  The answer completely depends on package
sizes.  Some designs might be forced into a larger size, while others would
fit.  There was no resolution on this issue at the meeting.  It will be
discussed further at the October meetings.

7.    Terminator Power Distribution [Aloisi/Ham]

This topic was not covered.

8.    Propagation Delay Length Limit [Ham]

This topic was deferred until the SCSI-3 General Working Group meeting due 
insufficient time.

9.    Pin-out Definitions [Ham]

Bill Ham briefly showed a foil of an 8-bit version of an LVDS cable.  Other
than a minor error Bill noticed, there were no issues identified.

10.   SPI-2 Architecture [Ham]

Larry Lamers lead a discussion of the marketing impact of how the documents
are focused and written.  Should Fast-40 SE, Fast-40 LVDS,  Fast-20, and SPI
be integrated into a singular document.  A unified document will add at 
6-8 months to the develop for resolution of the issues arising from
integration.  The benefit is that a singular document exists and associated
issues should be resolved.  The negative is the added time to market (time 
forwarding) will make it appear that the advanced physical interfaces are 
yet stable.

Jim McGrath volunteered to host a marketing oriented meeting on packaging 
technical work.

11.   EPI Draft Document Outline (95-320) [Ham]

This topic was deferred until the SCSI-3 General Working Group meeting due 
insufficient time.

12.   50 Ohm Cable Specification (95-317r0) [McCall]

David McCall presented a proposal for SPI-2 to consider a reduced impedance
cable specification.  Recommendations include dropping differential 
to 75 ohms and allowing for a 60 ohm termination option.  The benefit is a
better match with backplane implementations.  The consequence is increased
power consumption in the transceivers.  No conclusions were reached.

13.   Should RST Signal Remain Single-Ended? [Gardner]

This topic was not covered.

14.   +/- Signal Skew [Jander]

Mark Jander requested that the maximum skew time between the + and - lines 
each LVDS signal be increased considerably from the current 50 ps
specification.  After some discussion, the group agreed to specify this 
as 500 ps.

15.   Preliminary Test Results [Ham] (95-315)PrePPPPP pppre

Bill Ham presented some preliminary test data based on a reasonable
approximation of the LVDS interface.  Operation is feasible at Fast-100 over
25-30 meters in point-to-point mode.  With multi-drop, Fast-80 may work at 
20 meters.

Intersymbol interference is a real issue as the speeds increase.  A 
scheme may be needed.  Bill plans to do more testing.

16.   Meeting Schedule

Two meetings interim meetings were scheduled at the meeting, one on SPI-2
technical issues and another on parallel SCSI marketing issues.  The 
meeting is not an X3T10 activity.  Unfortunately,  both meetings were re-
scheduled the next day at the SCSI-3 Working Group meeting due to conflicts.
The re-scheduled meetings are:

Friday, October 13, 1995 - SPI-2 Technical Meeting at the Denver Stouffer
Concourse Hotel hosted by Symbios

Monday, October 16, 1995 - SPI-2 Marketing Meeting in the San Jose area 
by Quantum

In addition to the above meetings, a SPI-2 Study Group meeting was scheduled
for 9:00-5:00 Monday November 6th in Palm Springs and a EPI Study Group
meeting was scheduled for 9:00-12:00 Tuesday November 7th in Palm Springs.

17.   Adjournment

The meeting was adjourned at 6:00 p.m. on Monday September 11, 1995.

More information about the T10 mailing list