Date: January 22, 1991 To: X3T9.2 Membership From: Lawrence J. Lamers, X3T9.2 Secretary John B. Lohmeyer, X3T9.2 Chairman Subject: January 14, 1991 Low-Power Differential Working Group Meeting John Lohmeyer called the meeting to order at 1:00 p.m., Monday, January 14, 1991. He thanked Ken Post of Future Domain for hosting and arranging the meeting. As is customary, the people attending introduced themselves. A copy of the of the X3T9.2 membership list was circulated for attendance and corrections. The following people attended the meeting: Name Organization ------------------------------ ------------------------------ Mr. Robert Kellert Cirrus Logic Inc. Mr. Douglas Hagerman Digital Equipment Corp. Mr. Robert Liu Fujitsu America, Inc. Mr. Kenneth Post Future Domain Mr. Kurt Chan Hewlett Packard Co. Ms. Carrie Claflin Hewlett Packard Co. Mr. Bill Rentelmber Linear Technology Corp. Mr. Robert Reay Linear Technology Corp. Mr. James F. Mahar Linear Technology Corp. Mr. Lawrence J. Lamers Maxtor Corp. Mr. Gary Murdock National Semiconductor Mr. Thai Nguyen National Semiconductor Mr. Don Chin National Semiconductor Mr. John Lohmeyer NCR Corp. Mr. Luke A. Perkins NCR Corp. Mr. Gene Milligan Seagate Technology Mr. Robert L. Simpson Sony Corp. of America Mr. Robert N. Snively Sun Microsystems, Inc. Mr. Erik Jessen Western Digital 19 People Present Unfortunately, many of the agenda items from the Austin meeting in October could not be covered because no one from T.I. was present. Luke Perkins came with a detailed analysis of the three-resistor terminator (90-196). His conclusion was that a 650-135-650 ohm termination is feasible. An eight-bit low-power differential protocol chip could drive such a termination. Sixteen-bit low-power differential probably would require a second chip because of power dissipation limits. The alternative direction of using a single shunt resistor was also discussed. This method has two distinct advantages: 1) no terminator power is required and 2) it requires less power to drive since there is no uncertainty about the terminator power voltage level. The disadvantage is that it is not compatible with SCSI-2 differential devices. There was also some concern that the receiver circuits would be more complex and would have reduced noise margin. Gary Murdock discussed his proposal (90-177) to assign the low-power differential signals so that they would be more compatible with the single ended pin assignments. This would permit a device to operate in either mode depending on which kind of cable is connected. Mixing such devices with SCSI- 2 differential devices would require converter devices to shift the signal assignments. Gary said he expected preliminary test results on some low-power differential devices by the February meeting. On Tuesday, during the SCSI Working Group meeting, it was pointed out that the Small Form Factor group is currently planning to only support single-ended devices. Some of the ground lines have been reclaimed in order to include power and device ID lines. Unless this direction changes, it will not be possible to use either kind of differential device. This could have a significant impact on the potential market size for low-power differential devices. While it might be possible to design low-power differential so that it addresses the migration issues of both single ended and RS-485 differential, compromises would have to be made which would render the solution less than optimal for either group's needs. It is probably appropriate that the plenary group revisit the issues and lend marketing guidance (again). Action Items 1. Look at higher speed options - Mike Bartlett 2. Compatible with existing differential - Mike Bartlett 3. Are bias resistors needed and what happens if they are not there - Mike Bartlett 4. Investigate fail-safe inputs on single resistor approach - Luke Perkins and Gary Murdock Agenda for March 1991 Meeting 1. Spice of > 10 Megatransfers per second 2. Review National's testing results 3. Review new proposals Next Meeting The next working group meeting is being hosted by Dal Allan of ENDL at the Sunnyvale Hilton Inn beginning at 1:00 p.m. (see document 90-197 for hotel information).