Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

RE: [10GBASE-CX4] Re: Terminology




I believe Peter has stated the situation very well.

Thanks,

Dan

-----Original Message-----
From: Peter Bradshaw [mailto:pbradshaw@bitblitz.com]
Sent: Monday, February 10, 2003 9:04 PM
To: ddprocurve@antelecom.net; larry rennie; DOVE,DANIEL J
(HP-Roseville,ex1); Oleynick, Gary; Howard A. Baumer; Zev Roth;
10GBASE-CX4 (E-mail)
Subject: RE: [10GBASE-CX4] Re: Terminology


Slight expansion on Dan's remarks.

'Pre-emphasis' was chosen for the processing BEFORE the cable channel, in
preference to 'de-emphasis', because of prior use of this term. Since such
pre-emphasis is externally measurable, a specification is reasonable and
(probably) required, and indeed being worked on.

'Equalization' was largely pushed to the receiver, though there was some
discussion of the use of (passive) equalized cables. Adaptive equalization
implies some adapting engine; whether receiver equalization be 'active'
(presumably high frequency boost) or 'passive' (low frequency attenuation?)
was left up to the implementer, as far as the spec is concerned. Since
receiver equalization is (in general) normally not externally measurable,
except indirectly, the current expectation is that the receiver will have
only a performance (BER with worst case TX and channel) specification (apart
from return loss).

At least that is my recollection from the discussion at the meeting, leading
to the voting listed by Dan.  

Peter Bradshaw

> -----Original Message-----
> From: ddprocurve@antelecom.net [mailto:ddprocurve@antelecom.net]
> Sent: Thursday, February 06, 2003 7:35 PM
> To: larry rennie; DOVE,DANIEL J (HP-Roseville,ex1); 'Oleynick, Gary';
> Howard A. Baumer; Zev Roth; 10GBASE-CX4 (E-mail)
> Subject: Re: [10GBASE-CX4] Re: Terminology
> 
> 
> 
> Hello all,
> 
> Thanks for your patience.
> 
> From the December Meeting...
> 
> Pre-emphasis 21,Pre-distortion 5, TX Equalization 8
> 
> I just knew we had addressed this before. On to newer and
> better things....
> 
> Regards,
> 
> Dan
> >OK.
> >
> >Larry
> >
> >"DOVE,DANIEL J (HP-Roseville,ex1)" wrote:
> >
> >> Hello all,
> >>
> >> I am getting dizzy!
> >>
> >> Rather than go through this loop again, I will take the
> >> chair's perogative to request that we halt the discussion
> >> until I can find the appropriate decision in our minutes
> >> and remind everyone what it was. OK?
> >>
> >> Thanks,
> >>
> >> Dan
> >>
> >> -----Original Message-----
> >> From: Oleynick, Gary [mailto:GOleynick@fciconnect.com]
> >> Sent: Wednesday, February 05, 2003 9:50 AM
> >> To: 'larry rennie'; Howard A. Baumer
> >> Cc: Zev Roth; 10GBASE-CX4 (E-mail); Oleynick, Gary
> >> Subject: RE: [10GBASE-CX4] Re: Terminology
> >>
> >> All,
> >>
> >> I thought adjustments at the Rx end were referred to as 
> "adaptive or
> active
> >> equalization" and not pre/de-emphasis?
> >>
> >> Regards,
> >>
> >> Gary
> >>
> >> -----Original Message-----
> >> From: larry rennie [mailto:Larry.Rennie@nsc.com]
> >> Sent: Wednesday, February 05, 2003 9:17 AM
> >> To: Howard A. Baumer
> >> Cc: Zev Roth; 10GBASE-CX4 (E-mail)
> >> Subject: [10GBASE-CX4] Re: Terminology
> >>
> >> Thanks for the reply Howard.  I just wanted to make sure 
> that in our
> >> definition that neither term implied that it is a function 
> of the TX or
> RX
> >> exclusively, i.e., you can have pre-empahsis or 
> de-emphasis at the TX or
> RX
> >> end.
> >>
> >> Regards,
> >>
> >> Larry
> >>
> >> "Howard A. Baumer" wrote:
> >>
> >> > Larry,
> >> >     I thought at one point along the way we desided to call it
> >> > de-emphasis.  I know this hasn't caught on that well as 
> even us here
> at
> >> > Broadcom call it pre-emphasis at times.  I don't think this
> nomenclature
> >> > is that big of a deal.  The more important definition is how the
> >> > emphasis is computed and then communicated to the group.  On this
> point
> >> > the group desided to always speak in terms of 
> (1-Vlow/Vhigh), where
> Vlow
> >> > is the low value of the step response and Vhigh is the 
> value of the
> >> > peak.
> >> >
> >> > Howard
> >> >
> >> > larry rennie wrote:
> >> >
> >> > > Howard and Zev,
> >> > >
> >> > > In looking at some Dec 2002 presentations, I notice 
> that the 2-tap
> >> > > filter Broadcom calls a  tx "de-emphasis" filter is called a tx
> >> > > "pre-empahsis" filter by Mysticom (slide 5 of
> >> > > MysticomCX4_Dec0602:6.pdf). What is the terminology we 
> have decided
> >> > > upon?
> >> > >
> >> > > Regards,
> >> > >
> >> > > Larry
> >
> >
> >
>