P1394b High-Speed, Long-Haul PMD Meeting Thursday, October 23, 1997, Aston Wailea Resort, Maui, Hawaii Chairman: Colin Whitby-Strevens Secretary: Richard Churchill Reflector: p1394b@zayante.com Web site: http://zayante.com/p1394b Agenda 1. Welcome and Introductions 2. Approval of Agenda 3. Approval of minutes of last meeting (previously distributed) 3.1 Liaison from VESA 4. Report from S100 PMD Task Group 5. Consideration to allow optical connector late submission (Mini-MT) 6. Optical connector comparison matrix - review 7. Optical connector selection voting procedure 8. Optical transceiver parameters 9. Any other long-haul PMD issues 10. Future plans and schedule 11. AOB 12. Adjourn Meeting convened at 8:40 AM 1. Welcome and Introductions 2. Approval of Agenda The agenda was presented, and approved, with minor revisions and additions, without objection. 3. Approval of minutes of last meeting The minutes of the previous meeting were distributed via the reflector, and were approved without objection. 3.1 Liaison from VESA (Dick Scheel) Liaison report to IEEE P1394b 10/23/97 Dick Scheel / Sony dicks@lsi.sel.sony.com VESA Home Network Committee has expressed a concern about the optical fiber medium chosen by P1394b. 50 micron glass fiber is not a standard in the US (TIA Standards). 6.25 micron is in the current standard. DAVIC Home Network subgroup sent a liaison letter to this committee [Assorted observations were made regarding the fact that 50 micron fiber is not presently standardized, but that it is under consideration for such. Colin observed that part of what we are about is to "future proof" the system we specify, so that it supports the speeds we envision. There was a discussion of the liaison letter Colin and David Wooten are pre- paring for VESA, DAVIC, TIA and others.] [Observation was made that other groups are to meet within a couple of weeks at which decisions regarding whether to adopt 50 micron fiber. Colin said that we might recommend that these groups strike 62.5 micron fiber from their standards.] [Colin clarified for the minutes that we are recommending 50 micron fiber for adoption. This is more a forward looking matter, as 62.5 micron is optimized for past and present applications, but not for the future. A suggestion that we in some way deal with the question of how p1394b should deal with existing 62.5 fiber. Guidance is necessary. Do we even want to open the question?] Action item: Discussion of whether or not to include 62.5 micron fiber was claimed by Colin. [A dongle might work, but transceiver issues are involved, and thus at the higher speeds a mixed fiber environment may not work. We must think about this, and be prepared to discuss at the next meeting.] 4. Report from S100 PMD Task Group Minutes of Sept. 2, and Oct. ?? meetings <<< See group web site for Sharp documents >>> [Colin and others pointed out that this is a liaison report, and the additional material regarding connectors can only be in that context. A presentation of OMJ connector did not deal with 1394 bandwidths. Question was raised regarding what the aggregate bandwidth of the con- figuration presented via slide. Colin pointed out again that this is a liaison report, and the answers to many questions are not appropriate, as the material was previously discussed in other venues.] [Question regarding 1394 common mode signaling on the OMJ connector is a study project. Current shipments in Japan are on the order of 1 M per month.] <<< See group web site for Sony documents >>> [Question was asked as to whether the low-speed, long-haul group intends to specify one, or more than one, connector(s), and one, or more than one, medi[um|a].] [Questions regarding whether the low-speed group intends to convert 1394 to a different plug type ... Not decided.] 5. Consideration to allow optical connector late submission (Mini-MT) Moved by Richard Churchill, and seconded by Jack Hollins, that we add the mini-MT connector to the matrix. [Does AMP intend to withdraw the proposal previously presented? No.] [Comment regarding opposition due to the lack of a presentation the previous month. Colin - We must decide on the information available, and if the presentation was not made that is simply a reduction of the information (or of the distribution of information). Comment regarding the stance that it should not be considered. Colin - We should make our decision on the basis of as wide a range as possible, but this is his personal view.] Votes were 26 for, 4 opposed. Motion was carried. Mike Brown moved (seconded by Jim Busse) that we allow AMP to make an approximately 10 minute presentation. [Further comments in opposition.] Votes for were 14, opposed were 18, and the motion fails. 6. Optical connector comparison matrix - review Comparison Matrix - 1 <<< Table >>> STANDARDS DOCUMENTS Fiber jack is best described in TIA. [We should point to an international document, if it contains the needed information. Thus, IEC would be preferable, if that description is sufficient.] F07/PN, use IEC Direct use TR41 VF-45, PN ..., FO-6.3 LC IEC Mini-MT IEC [Comments about standardization -- We might select something no-one else will eventually settle on. Most standards are broad, and not limited to a particular connector. These other standards often describe several or all connectors claiming viability. Past IEEE groups have sited "incomplete" documents, substituting the appropriate reference at a later date. There is often much "work in parallel."] PATENT LETTERS to IEEE FibreJack - yes [copy needed for Colin and David Wooten. Must have a final formal letter to IEEE in hand for final decision. May need to deal with this on individual basis, with each concerned party requesting from each vendor.] F07/PN - in process Direct - yes LC - in process Mini-MT - ??? [If the connector has specific fiber requirements, these should be indicated, and any patent disclosures related to the fiber in question made.] [Note that we cannot quote prices, but can look at open market pricing. We may also look at comparative costs, sans monetary numbers.] Pricing (comparative) [All vendors made rather complete comparable comparisons ... pretty much "apples to apples" in character.] [Colin checked prices at Fry's Electronics at the October TA meeting in Phoenix, and quoted prices (which are thus "open market" prices as of the day and vendor in question. The numbers in dollars are not quoted here, and the reader is directed to independent pricing if they are interested. Initial comparison indicates a "near wash" on the cost.] [Proposed that we strike certain lines of the table, dealing with the various costs. Several comment that these cannot really be considered here, and that there are anti-trust questions. We end up being precise and violating anti-trust rules, or make the numbers fuzzy and have little to work on the basis of. We are not dealing with something that critical, here, and anyone who wants to create an anti-trust issue based upon the vague information here is more interested in anti-trust, not in the information, which is far to vague to form a real basis for decision. Removing it does not eliminate the information from consideration. Moved by ??? and seconded by Lee Wilson that we strike these lines from the matrix. [Will word-smithing this to indicate "approximate equivalence" with UTP-5 satisfy the concerned parties? Both indicate that it will. Secretary notes that removing from the matrix does not mean the information will be removed from the minutes. The minutes must be as accurate as may be managed, and thus these lines will be contained in the minutes. We should state that they are conjecture, and pre- sented by the individual proposers based upon independent means and method.] Motion died. Moved by Dan Wiencek, and seconded by Bob Sharp, that all the lines referring to individual component costs be deleted from the matrix. The motion carried with 33 yeas and one nay. [We need a baseline system cost? Proposers should provide contacts for committee members to obtain individual prices from.] Action Item -- Colin will discuss with the proposers if there can be further clarification of the issues involved. 20 year life-span (as per FOTP) Fibre Jack - yes F07 - not tested Direct - yes VF-45 - tested to TIA environmentals [Would like to see the assumptions on this. What was the analysis regarding bends, etc. Information will be presented on the Web site. What number was used for the pull force test? Will be posted.] LC - yes - "1000 repetitions" [These are not reliability indicators ... We might be better served by "durability" than "life span" indications. We can test relative durability ... this is one of several FOTP tests, and consider environmental endurance, as per FOTP. Didn't you want an indication of "need to upgrade"? We worked on the basis of this in part. This information needs to be put up on the web site. This is a different issue ... What will requirements be in twenty years? We wanted this cable to be able to take 4 GBaud signals 20 years after installation? Is this meaningless because everyone will say yes? Provide same kind of data ...] Mini-MT - Installation cable termination complexity (describe) [May need to indicate level of skill required, and steps, etc. May be best to indicate "As demonstrated" here.] Maintenance required Fibre Jack - no cleaning F07/PN - wipe endface/no Direct - none [No real difference, as all recommend wiping/cleaning before each connection ...] Maintenance advised [Much the same as above] [HFE is recommended by 3M, because of non-conductance, etc. 3M does not consider it safe to spray alcohol into a transceiver, related to differences in implementation and flexibility] Protection from Dust [Dust caps attached to housing? Most are not, and thus can be lost easily. Indicate that they are attached or intrinsic. There is nothing inherent in NOT providing such if customers desire.] Comparison Matrix - 2 <<< Table >>> Duplex, asummetric, differentiated, (keying, etc.) [Either could be added or present in all.] Support speeds from 125 MBaud up to 4 GBaud Suitable for equipment dongles and wall plates Enables user assembled patch cords (NB not a requirement) [Want to remove this? Likely yes. Strike line!] Size compared with RJ-45 [Size was simply a point of reference?] Worst-case loss (dB) Optical performance complies to TIA 568 requirements [TIA 570 may be more appropriate here. This is the one that refers to the residential environment.] Suitable for MMF [Need to be more specific to 50 micron/125 micron OD fiber.] [] Suitable for POF Eye Safety issues [Assume class 1 transceiver. With that assumption, the line may be stricken.] Physical Safety issues [During installation and after ... Pulling through the phone, it is more likely a professional ... Question on amateur model for installation ... Is this really practical, due to cost. These things may be expensive. This brings up a question of whether this is an advantage for POF. Tools may be rentable. This may also be considered simply a metric -- How hard is it to install, and thus how much installation expertise and ex- pense will be required.] Comparison Matrix - 3 <<< Table >>> What is available for GOF What is available for POF What is under development for GOF What is under development for POF Performance after 500 cycles Performance after 1500 cycles Independent Testing (including round robin info) Number of vendors of the proposed connector Transceiver integration information 7. Optical conector selection voting procedure Optical connector selection voting procedure - We're governed by Roberts Rules of Order - NOT well suited to multiple choice decisions - Propose - A series of multple choice straw votes - result will be to identify a single candidate - A formal motion to endorse the selection of the candidate - Straw votes - process of successive elimination - Each voter selcts one proposal to eliminate - count the elimination votes for each proposal - eliminate one proposal (or more than one in the event of a tie) - Repeat the process several times, each time eliminating another candidate. [Clarification of voting rules ... Who is eligible? Anyone registered at and attending one or both of the two meetings immediately preceeding a formal vote, AND registered and in attendance at the meeting where the formal vote is to be taken is eligible to participate in that vote. There are no proxies.] 8. Optical transceiver parameters 9. Any other long-haul PMD issues 10. Future plans and schedule December p1394b meeting details --- December 2 & 3, 1997, Fort Lauderdale ???? North, Fort Lauderdale, FL 11. AOB a. Discussion of whether or not to include 62.5 micron fiber by Colin. b. Discussion with the proposers if there can be further clarification of cost questions by Colin. c. Additional information to be provided by the various proposers in a timely manner for continued consideration of their proposals, to be posted to the ftp site and to the reflector. 12. Adjourn ============================================================================= Attendance List Richard Churchill 281-514-6984 richard.churchill@compaq.com Prashant Kanhere 510-668-1773 prashant@macrodesigns.com Dick Scheel 408-982-5834 dicks@lsi.sel.sony.com Brad Saunders 714-221-6513 bradley.saunders@rss.rockwell.com Steve Bard 503-264-2923 steve_bard@ccm.intel.com Stephen Finch 714-573-6808 steve.finch@tus.ssi1.com Stephen Cowen 303-772-9729 stephen.cowen@st.com David Rekieta 972-480-6466 drekieta@ti.com Richard Baker 512-425-6205 artb@msg.tx.slr.com Dave Cook 512-984-3699 ddcook@mmm.com Juan Pulido 512-984-5188 jmpulido@mmm.com Dao-Long Chen 970-223-5100 x9461 dao-long.chen@symbios.com Michael Shinkarovsky 610-712-2938 mshinkarovsky@lucent Farrukh Latif 610-712-7546 flatif@lucent.com Peng Zhang 972-480-3109 pzhang@ti.com Ken Taylor 508-347-3309 foken@aol.com Jim Kevern 17-986-5701 james.kevern@amp.com Charles Brill 717-592-6198 cebrill@amp.com John Ta 714-573-6957 john.ta@tus.ssi1.com Jack Hollins 408-957-2309 jack_hollins@eng.adaptec.com Ju-Ching Tang 510-623-8300 x5189 jctang@corp.cirrus.com Weidong Jiang 408-922-9783 wjiang@fmi.fujitsu.com Masood Sharif 732-957-5479 mshariff@lucent.com Derek Imschweiler 717-235-7512 dimschweiler@stwconn.com Bob Scharf 717-235-7512 rscharf@stwconn.com Karl Bergstrom 800-888-0211 x4563 mpeg007@aol.com Khorvash Sefidvash 714-474-7095 khorvash.sefidvash@symbios.com Oleg Awsienko 602-554-9666 oleg_awsienko@ccm.intel.com John L. Hill 717-592-6375 jhill@amp.com Bill Pherigo 970-229-3586 wlp@fc.hp.com Sushant Verman 416-620-7400 sushant@lsil.com Bill Prouty 916-785-4631 bprouty@hp.com Dave Brunker 630-527-2622 dbrunker@molex.com Max Bassler 630-527-4490 mbassler@molex.com Palanisamy Mohanraj 602-554-4243 palanisamy_mohanraj@ccm.intel.com Dave LaFollette 408-765-2587 dalafolle@mipos2.sc.intel.com Yasumasu Hasegawa +81 22 347 1128 yasumasa@ffm.fujifilm.co.jp David Wooten 281-518-7231 david.wooten@compaq.com Bill Northey 717-938-2119 northewa@bergelect.com Jerry Hauck 408-765-5528 jerry_hauck@ccm.sc.intel.com Martin Sodos 408-969-4683 msodos@issiusa.com Takashi Matsui +81 76 274 2440 takashi_matsui@ibm.net Dan Landeck 408-922-9522 dlandeck@fujitsu.fmi.com Dave Instone +44 1705 486363 dinstone@uk.xyratex.com Nelson Arata 408-721-4979 nelson.arata@nsc.com James Gay 512-891-2218 jimg@oakhill.sps.mot.com Jim Busse 415-528-3810 jimb@ccgate.sj.nec.com Peter Teng 408-588-5555 peter_teng@el.nec.com Kenji Watanabe +81 3 5448 3537 nabeken@sslab.sony.co.jp Shuntaro Yamazaki +81 44 856 2082 yamazaki@ccm.cl.nec.co.jp Takatoshi Mizoguchi +81 745-65-1161 mizoguti@shinjo.sharp.co.jp Hirohisa Wakai +81 743 65 4529 wakai@slab.sharp.co.jp Don Wiencek 708-532-1800 dcw@panduit.com Mike Borgman 708-532-1800 mjbo@anduit.com Mike Brown 602-554-3713 mike_brown@ccm.intel.com Ron Soderstrom 507-253-6290 rons@vnet.ibm.com Lee Wilson 512-838-6569 leewils@austin.ibm.com Colin Whitby-Strevens +44 1454 611500 colinws@bristol.st.com Vilas Bhade 408-777-4723 vilas@wipro.com Jim Doyle 602-554-2051 jdoyle@sedona.intel.com Ed Butler 602-554-0751 edward_butler@ccm.intel.com Mike Winchell 970-225-4807 mike.winchell@symbios.com Bill Duckwall 408-461-4902 duck@zayante.com Haim Shafir 408-343-0192 hshafir@ix.netcom.com There were 64 attendees who signed the attendance list.