Date: 4 Jun 97 14:15:27 -0700 Subject: Draft minutes of 1394b Long-haul Low Cost(S100) May 28, 97 meeting From: "Takahiro FUJIMORI" To: cebrill@amp.com, stakagi@amp.com, kazu.hyakuktake@amp.com, ebergles@cel.com, sunnyl@us.ibm.com, mike_brown@ccm.ch.intel.com, gene_p_matter@ccm.fm.intel.com, rajiv_choudhary@ccm.jf.intel.com, eric_hannah@ccm.sc.intel.com, anc@hplb.hpl.hp.com, patelb@lucent.com, morikura@isl.mei.co.jp, Kazuto.Matsui@st.com, mbassler@molex.com, amiyachi@molex.com, mseto@molex.com, aaso@molex.com, anoda@molex.com, pyu@el.nec.com, negishi@kansai.nims.nec.co.jp, s-matsui@nims.nec.co.jp, kobaya@nims.nec.co.jp, tsugita@lsictmg.nec.co.jp, koyamam@std.abk.nec.co.jp, saito@optsys.cl.nec.co.jp, yamazaki@optsys.cl.nec.co.jp, new@optsys.cl.nec.co.jp, fujimori@arch.sony.co.jp, okawa@sm.sony.co.jp, takizuka@sm.sony.co.jp, sakurai@sslab.sony.co.jp, nabeken@sslab.sony.co.jp, tasaka@sslab.sony.co.jp, chizu@cv.sony.co.jp, dvj@ibm.net, colinws@bristol.st.com, dao-long.chen@symbios.com, mike.winchell@symbios.com, crn@msg.ti.com, inoue@ba2.so-net.o Cc: p1394b@fireflyinc.com X-Mailer: Cyberdog/2.0 MIMEEnabled: Y MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="Cyberdog-MixedBoundary-0012AD1C" Content-Transfer-Encoding: 7bit --Cyberdog-MixedBoundary-0012AD1C Content-Type: text/enriched; charset=macintosh Content-Transfer-Encoding: 8bit X-Fontfamily: Monaco X-Fontsize: 9 Dear all, The following is the draft minutes for the half day P1394b Long-haul Low Cost (S100) TG meeting in Tokyo, May 28, 1997. Please send me an e-mail if you find any comments/errors. Regards, Taka H. Fujimori fujimori@arch.sony.co.jp 1394b Long-haul Low Cost(S100) TG Chair ============================================================================ Minutes P1394b Long-haul Low Cost(S100) Task Group Kick-off Meeting, May 28, 1997, Sinagawa Ward, Tokyo, Japan Sony Osaki East Technology Center Taka Fujimori, chair Takayuki New, secretary(in Japanese Language) Agenda 0. Introductions Official Language Secretary History POR Repeater : Mr. Yamazaki 1. Requirements 2. Key Issues 2.1 Codings : Mr. Fujimori 2.2 Start-up(Beta Mode): Mr. Fujimori 2.3 POF Connector 2.4 SI POF 3. Open Issues GI-POF 4. Discussion 5. Schedule Standardization Editor 6. Action Items 7. Closing Next Meetings 0. Introductions *Official Language Taka Fujimori started the meeting and asked for language preference. As no one prefered English, Japanese language was chosen for the duration of this meeting. He mentioned that other than spoken language presentation slides, documents and minutes, etc. have to be in English. Sign up sheet was circulated. And then as a IEEE meeting convention, each participant introduced himself. Then Taka explained that the members of IEEE meetings are indivisuals, not organaization representatives. But the group may decide whether the voting should take place by indivisuals or by organizations. He will check the voting policy of p1394b group and this TG agreed to accept it. *Secretary The agenda of this meeting was shown and agreed. *History Taka reported that long-haul objective was added to p1394b WG at 1394TA Eindhoven this April. In the following p1394b Santa Clara meeting May 5-6, the long-haul Task Group had a first meeting and formed the following two groups as they found two distinct objectives: Long-haul Low Cost(S100) Task Group Long-haul High Speed(S800-) Task Group Motivations behind Long-haul Low Cost(S100) Task Group were described. First of all 4.5m/hop limit of 1394 is frequently said to be too short for applications outside a room. 100m reach allow us to use 1394 outdoors. Home Network standardization activities are also behind us. VESA/Home Network group has a requirement of min 50m andl min 50Mbps for room to room connection. It has a first internal proof-of-concept demo next month in San Jose, where MPEG source/sink devices, DV CamCorders, PCs and long distance 1394 would be demonstrated. DAVIC also placed Call for Proposals 8 for Home Network. As they had agreed to adopt 1394 for their Set Top Boxes, 1394 based home network are likely standardized as DAVIC spec 1.4 by early 1998. DAVIC might add ATM based home network next month. Their target is 100m and 200Mbps. *POF(Plastic Optical Fiber) Repeater Yamazaki-san presented the NEC's POF repeater, connecting a PC and a 1394 PC Camera with 70m SI-POF(Step Index POF), that was demonstrated sucessfully in WinHEC'97 this April. On May 20, '97, ATM Forum has approved 155Mbps POF(50m)/HPCF(Hard polymer clad silica optical fiber, 100m) PMD specification including PN connector based on JIS F07 connector. An F07 plug can be mated with a PN receptacle. This ATM Forum spec of POF/HPCF is good enough for S100 and both can be extended to S200/400. IEC accepted the spec as a Committee Draft, and would standardize it within one or one and a half years. There are more than five suppliers of PN connector in US and Japan and more than 15 suppliers of POF/HPCF Link. ATM Forum spec of 50m 156Mbps includes 2 splicing points. 70m is possible with no splicing. He addressed the necesity of a new power class in Self_ID packet for non power repeating power consumer, that should be standardized in p1394a meeting. Taka mentioned p1394a may be finished by this August. We should propose this new power class either quickly to p1394a or p1394b might be the place to define it. Yamazaki-san also showed a result of good eye pattern after 100m transmition with GI-POF and DVD Laser Diode. Takagi-san raised a cost concern of large diameter Photo Detector running at Gbps rate. 1. Requirements Yamazaki-san presented the following Requirement Criteria 1. Architecture -- PMII&1394b-n -- He raised a concept that the connection(ie, 1394) between a 1394 node and a POF repeater, should be the PMII. Taka commented that it may not necessarily be PMII. The intention of PMII is that 1394b PHYs and 1394b PMDs share the same interface, so that PMDs can be changed. He also address two kinds of repeaters: unintelligent one and intelligent one with PHY logic, Node_ID/self_ID, etc. -- Home Network Modeling -- He came up with a model of home network with STAR wired 4 long distance 1394 hops and 8 short 1394 hops, serving two rooms. 2. PHY Delay Under the model of home network, PHY_Delay was caluculated to show degradation of bus performance. If PHY_Delay is less than or equal to 460ns and the gap count is 63, then 4 long distance hops and 8 short hops can be supported with the max gap count 63. 1394-1995 has no spec of minimum of MAX_ARB_STATE_TIME. 1.1us + margin was suggested. Also the current defference of 300ns between ROOT_CONTENTION_FAST and ROOT_CONTENTION_SLOW are not large enough. 3. Maximum Speed The group should agree on a maximum speed requirement. 4. Start up To start up in Beta mode at the maximum baud rate which is acceptable to both ends. To start up with no user intervention or configuration. 5. Block Coding Compared 4 possibel coding schemes: 4B/5B(+NRZI) by NEC, 4B/5B(+NRZI) by SONY, 8B/10B by IBM, 8B/10B by HP should be studied intensively. 6. Powering ID(P1394a issue) An addition of the POWER_CLASS code in Self-ID packet, that should be addressed in p1394a WG. 1. A repeater which is powered from the cable and not repeat power. Taka reported the requirements of Long-haul Group of p1394b meeting in Santa Clara. Regarding min 50m/hop, a question was raised whether "100m prefered" is necessary or not. Taka mentioned that installed UTP Cat 5 follows EIA/TIA recommendation which is 100m, therefore this group should include 100m in the requirements. Yamazaki-san said UTP Cat 5 may face difficulty in passing FCC Class B requirements and might end up with shortening length. Chair decided to take a vote. Motion: Min 50m/hop is good enough and remove "100m prefered" out of the requirement. 12 Yes, 1 No, 2 Abstention, and the motion carried. Regarding kinds of medium, POF and UTP Cat5 were thought to be in the scope. Morikura-san proposed to work on HPCF as well. The group agreed to work on those three media unless members propose another medium. Maximum Speed was then discussed. Since Long-haul high speed TG is focusing on S800 and above, this group decided to adddress 1394-1995, i.e., S100 S200 and S400. FCC Class B compliance, Amateure installable including installation guideline and installation test, and IEC 825-1 eye safety were all agreed by the group. A clarification was requested whether FDA, CDRA are necessary in addition to IEC825-1. As a result the group agreed on the following Requirements: 1. min 50m per hop 2. UTP-5, POF and HPCF 3. Speed of upto S400 4. Facilitate FCC Class B emission commpliance 5. Amateur installable Installation Guidelines, Installation Test 6. Eye Safety IEC825-1 (The group didn't have enough expertise of FDA and CDRA. Clarification for the necesity of FDA and CDRA were requested.) 2. Key Issues 2.1 Codings Taka presented a coding scheme of 4B/5B based long distance 1394 and a 1394 control code mapping method onto 4B/5B control codes. He concluded the presentation as: 1. The scheme is complete and there's no max speed limit. 2. Sony & Hyundai submitted a joint proposal to DAVIC. IPR statement sould be available to the group by this August. 3. 4B/5B code is proven by 100Mase-T Ethrenet. Cost is minimal. Pass FCC Class B 4. The verilog code of this method will be availabe. Yamazaki-san proposed a through study of 4 sckemes, 4B/5B NEC and SONY, 8B/10B IBM and HP, by comparing delay, gate count, IPR, etc. He also suggested issuing "Call for Proposal". Taka is going to check whether issuing a CFP is possible. The group agreed to work on codings intensively, as it is always a controvertial point. 2.2 Start-up(Beta Mode) Taka briefly explaind the start-up mode proposed by Colin in the last p1394b Santa Clara meeting May 5-6. It is addressing an issue of finding max common PLL speed between PHYs of various speeds. Basically this beta mode takes place after power on and before 1394 protocols, thus 1394 protocols need not be changed at all. First both side talks at S100 PLL speed and exchange speed capabilities starting from S100 then to S200. If both side are S200 capable, then they switch PLL speed from S100 to S200. Again both side exchange speed capabilities by starting from S100, S200 to S400. If they find that both PHYs are S400 capable, then they switch PLL speed from S200 to S400. This process itarates until they find max common PLL speed. After finishing this Beta Mode, all the 1394 protocols including Bus_Reset, Tree_ID, Self_ID and so forth are executed. He said this beta mode is an open issue in our group. 2.3 POF Connector Sakurai-san proposed a small POF connector which is suitable for small consumer devices. It's 14mm Wide and 8mm High with a 5mm separation of fibers. Reverse insertion is also prevented. A question about crosstalk was raised especially at S400 rate. 2.4 SI POF No presentation was made. 3. Open Issues GI-POF Taka announced that the Ichikawa-san of Mitsubishi Rayon Co., working on GI-POF, has agreed to give p1394b group a GI-POF presentation at this August Meeting in Hawaii. 4. Discussion S100 Long Distance PHY also has to have start up mode? The answer is YES to keep interoperability with higher speed Long Distance PHYs. Do we need to consider devices with directly attached long disance 1394 or just stardard-long distance repeater should be in the scope? The conclusion is we should have directly attached long distance devices in our scope, because consumer devices may adopted it as is an optical interface of consumer audio devices today. 1394 control code mapping onto HP's 8B/10B code is not available today. Therefore it may be difficult to study it intensively. To meet the assuming schedule of '97/E we need to make a decision in a couple of meetings. This group has to specify POF specification at S200/250Mbaud and S400/500Mbauud, since no other standardization body is addressing this frequencies. The spec can be futher forwarded to IEC. 5. Schedule Standardization The group decided to split the schedule into two phases: Vote by '97/E to meet DAVIC schedule Vote by '98/E to achieve S400. Also performance goals for PMD and PHY are separetly determined as follow: PMD PHY Phase 1 '97/E: S200 S400 Phase 2 '98/E: S400 -- Note: UTP is only at S100 Editors: Provisional PHY editor Taka Fujimori PMD editor TBD 6. Action items PMD POF S200 ----- Yamazaki-san PMD HPCF ----- Morikura-san PMD UTP Cat 5 Emission ----- Takizuka-san PHY Coding Intensive Study ----- Taka Fujimori Start-up ----- Taka Fujimori 7. Closing Tentative meeting Schedule July 8 location TBD, chair asked for a hosting company September 2 location TBD, chair asked for a hosting company The meeting is ajurned. ============================================================================= Attendees: Taka Fujimori Sony +81-3-5488-6353 fujimori@arch.sony.co.jp Hiroshi Takizuka Sony +81-466-30-4050 takizuka@sm.sony.co.jp Susumu Morikura Matsushita +81-6-906-4876 morikura@isl.mei.co.jp Michihiko Sakurai Sony +81-3-5448-5362 sakurai@sslab.sony.co.jp Akira Aso Molex +81-462-65-2328 aaso@molex.com Masashi Seto Molex +81-462-61-4500 mseto@molex.com Kenji Watanabe Sony +81-3-5448-5362 nabeken@sslab.sony.co.jp Susumu Tosaka Sony +81-3-5448-5362 tasaka@sslab.sony.co.jp Akira Koyama NEC +81-471-85-7483 koyamam@std.abk,nec.co.jp Shuntaro Yamazaki NEC +81-44-856-2082 yamazaki@optsys.cl.nec.co.jp Shinichi Takagi AMP +81-44-900-5012 stakagi@amp.com Tatsuo Inoue Arch Tech +81-3-5545-7813 inoue@ba2.so-net,or.jp Kazuto Matsui SGS-Thomson +81-3-3280-4128 Kazuto.Matsui@st.com Chizu Kobayashi Sony +81-3-5488-6353 chizu@cv.sony.co.jp Tomoki Saito NEC +81-44-856-2082 saito@optsys.cl.nec.co.jp Takayuki Nyu NEC +81-44-856-2082 new@optsys.cl.nec.co.jp Kazunari Hyakutake AMP +81-44-844-8077 Kazu.hyakuktake@amp.com --Cyberdog-MixedBoundary-0012AD1C--