Re: PAM5 Objectives
Jamie,
If you're not interested in competing with the Serial PHY, then you are
competing with the WWDM PHY. Their solution already addresses 300 meters on any
MMF.
Since all IEEE PARs required multi-sourced parts, I would say that you need to
sign up at least 3 vendors in each of the following categories to authenticate
your claims that adding PAM5 to WWDM makes for a solution that is <1/2 the cost
of WWDM and meets the other objectives that the PAM SIG is assembling. Note that
these objectives are similar to those ascribed to by both the Serial and WWDM
camps. My proposed vendor categories for PAM5/4WDM are:
1) ASIC vendors;
2) Laser vendores;
3) Optoelectronic component (e.g. linear postamp, TIA, laser driver, etc.)
4) Transceiver module vendors;
5) Equipment vendors;
If you can't sign up three vendors in each category, then I would suggest that
your drop your PAM5/WWDM proposal in the face of lack of support and an
inability to meet the 1st PAR Criteria: Broad Market Potential.
What I'm getting at is that the cheapest 10 GbE solution MUST be a high volume
solution since volume will dictate cost. Broad Market Potential is required in
order toi achieve high volume.
As a Special Interest Group, we need to stop focusing on any specific proposal
and instead focus on developing PAM5 objectives, validating those objectives
with 10 GbE equipment vendors, and, it PAM5 passes muster there, developing a
CONSENSUS PAM5 solution.
Best Regards,
Rich
--
Jaime Kardontchik wrote:
>
> Rich,
>
> The objective should be to provide the cheapest 10 GbE solution,
> lowest power, etc... within essentially a building environment.
> It is here where "cheap" and "power consumption" is of
> paramount importance. I am not interested in competing with the
> serial guys (*) on their 2, 10 or 40 km objectives using SMF.
>
> Hence, I think it more appropriate to target the following fiber
> objectives of the 802.3ae:
>
> - minimum of 100 meters on installed MMF
> - minimum of 300 meters on new 2200 MHz*km MMF
>
> Notice, by the way, that although we cannot compete with the
> serial guys on link distance on SMF, we have a clear advantage
> over the serial guys on link distance over both the installed
> 160 MHz*km MMF and the new 2200 MHz*m MMF: the serial
> guys have given up completely on supporting the minimum
> objective on installed MMF and they can barely meet
> the 300 meters on new MMF, whereas we could easily go
> much higher than this.
>
> (*) serial guys defined here as 10.3 Gbaud, the standard symbol
> rate used by the serial SIG (Special Interest Group)
>
> Jaime
>
> Jaime E. Kardontchik
> Micro Linear
> San Jose, CA 95131
>
> Rich Taborek wrote:
>
> > Pat,
> >
> > A few minor conmments:
> >
> > 1) I agree with Jay's comments regarding power consumption
> > 2) PAM5 solutions must also be significantly cheaper than Serial (<1/2). The
> > first bullet says "installed base" solutions which alludes to 62.5 um MMF.
> > Serial doesn't address these. However, PAM5 should also address SMF at <1/2 cost
> > of Serial or WWDM.
> >
> > Best Regards,
> > Rich
-------------------------------------------------------
Richard Taborek Sr. Phone: 408-845-6102
Chief Technology Officer Cell: 408-832-3957
nSerial Corporation Fax: 408-845-6114
2500-5 Augustine Dr. mailto:rtaborek@xxxxxxxxxxx
Santa Clara, CA 95054 http://www.nSerial.com