| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
| Good morning Ramana, 1.- By using 970 ~ 990 nm wavelength range, we are not against the Broad Market Potential of the CSD. There are contributions presented in the TF and it has been stated many times in our discussions that it is much easier the fabrication of reliable devices at longer wavelengths and any vendor with 940nm in its portfolio today will be able to produce 980nm devices. The number of devices fabricated today at 940nm is several orders of magnitude the production of 850nm. 2.- I deducted you have reasons to think that the paper below support your position of the new math for reliability assessment of VCSEL. I copy here the URL so anybody can read it and judge (it is open access). It would be very interesting that you can explain the specific parts of the paper that support your position. 3.- The option of two bands (on wideband wavelength and other single wavelength) is even worse for the market than just one wide 840 ~ 990. On top of that, we would be defining two solutions for the same problem without arguments behind, which in my opinion is against CSD Distinct Identity. I have asked many times about the testing topic. Suppose that we change TX and RX characteristics to be 840 ~ 990 nm. Which wavelength or wavelengths should be used for SRS testing? It is up to the implementor? Should we specify in the 802.3cz amendment which one for interoperability? Is the industry to accept just single wavelength reference TX in the market to make ECU and Vehicle level validations? All these questions are about interoperability and no one in this forum is giving opinion. If we accept 840 ~ 990 nm, next point will be to guarantee interoperability by test methods. We cannot leave this gap in our specification. 840 ~ 950 has the same problem to solve of 840 ~ 980 about test specification. 4.- About SWDM. As far as I know, these systems are MSA vs 802.3. If I am wrong, could you give me references to specific clauses in 802.3 or running projects where multimode SWDM are specified so I can read how the SRS test is specified. BTW, in my opinion BiDi systems are different, because specific different wavelengths (with given tolerances) are used in each direction, so it is clear how SRS has to be implemented. Ref TX for TR and RT PHY have to use different wavelength. On the other hand, IMHO, the 802.3db, clause 167.8.14, has the specification gap that I want to avoid in 802.3cz. If I read 802.3db spec I would understand that for VR types, full wavelength range of the RX has to be tested under SRS test method, which means having many Ref TX operating in different wavelengths. Other can say, no, I just test at 850nm. The specification is saying nothing, which for me is a gap. Rubén Pérez-Aranda, KDPOF
To unsubscribe from the STDS-802-3-OMEGA list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-OMEGA&A=1 |