Re: [BP] Presentation
One thing to keep in mind is that we are pushing the system (eye is
closed going into the receiver and we've decided to expect that the
receiver will not cancel out cross talk). Because of this even the
reduced levels at Fs (10.3125) could make a difference. If we want the
informative channel recommendations to indicate a particular channel
will not work and the only difference is at the higher frequencies
around Fs then we need to recommend out to that point. If we don't want
Annex 69B to do that then we don't need it to reccommend out that far.
Do notice that the current reccommended IL goes from 50MHz to 15GHz,
which is nearly 1.5Fs!
Howard
Mellitz, Richard wrote:
>All that said, maybe we can raise the ICR limit small amount without
>going to 10GHz. However look at an expanded ILD up to 10GHz may make
>sense too. We are still working on the data to support what makes most
>sense. So stay tuned.
>... Rich
>
>-----Original Message-----
>From: DAmbrosia, John F [mailto:john.dambrosia@TYCOELECTRONICS.COM]
>Sent: Wednesday, April 05, 2006 9:42 AM
>To: STDS-802-3-BLADE@listserv.ieee.org
>Subject: Re: [BP] Presentation
>
>Charles,
>I am quite amused.
>
>First, I understand your comments regarding the Tx energy, however, we
>are talking about an informative channel model, not a Tx specification.
>I think the Tyco Case #7 is an excellent example of the sort of things
>that we find bothersome. It meets the specifications as we have them
>currently in place, but there are issues, and I would say Howard's
>approach of extending the frequency range can be a further way to filter
>channels, and should be considered.
>
>You know I have issues with Frankenstein channels. Rich has some
>concerns as well, especially when conditions are created that appear to
>be failing from the beginning. Marginal channels should be considered
>not failing channels.
>
>As I said before my key point of the presentation was to note that the
>ATCA channel data provided really isn't for ATCA because of the 2nd FEXT
>aggressor.
>
>John
>
>-----Original Message-----
>From: Charles Moore [mailto:charles.moore@avagotech.com]
>Sent: Wednesday, April 05, 2006 3:45 AM
>To: DAmbrosia, John F
>Cc: STDS-802-3-BLADE@listserv.ieee.org
>Subject: Re: [BP] Presentation
>
>john,
>
> I guess that it amounts to a question of clarification. The
>"Frankenstein" channels are an attempt to probe the boundaries of
>the channel spec. If they indicate that there could be channels
>which pass spec but will not work with any reasonable transceiver
>pair it may be a good idea to change the spec. If we are going to
>change the spec i do not want it to come in through the back door,
>i want it fully discussed first. Whether we are discussing changing
>the spec is what i want clarified.
>
> If we are discussing changing the spec, i will say that i
>am not too happy with extending the frequency range for the spec
>to 10.3125GHz, where there is no energy coming out of the Tx and
>where most channels near the edge of spec will have negligible gain.
>When i look at the roll off of the sinc(pi*T*f) function, speed
>limitations of the Tx, and the Rx, i think that any spec which
>relies on characteristics above 0.7*Fs (eg 7.2GHz) is likely to
>give unreliable results.
>
> charles
>
>DAmbrosia, John F wrote:
>
>
>>Charles,
>>That is correct. Is your question one of clarification? I believe
>>these limits are still fair game to comment on, given the on-going EIT
>>work and the subsequent impact it has on the channel models. (Adam -
>>
>>
>if
>
>
>>I am wrong, please correct me.)
>>
>>Howard did some good work, and with the exception of changing the Amax
>>equation to a fit of data as opposed to its current basis on material
>>properties provided by Joel, I am open-minded to the stuff he has
>>
>>
>still
>
>
>>proposed. True, it will put more burden on the channel
>>
>>
>implementation,
>
>
>>but we are suppose to be trying to divvy up the problem, and I would
>>state provide an informative model where false positives are limited.
>>
>>
>>John
>>
>>-----Original Message-----
>>From: Charles Moore [mailto:charles.moore@avagotech.com]
>>Sent: Tuesday, April 04, 2006 3:29 AM
>>To: STDS-802-3-BLADE@listserv.ieee.org
>>Subject: Re: [BP] Presentation
>>
>>john,
>>
>> I notice in your supporting slides that most of your comparisons
>>are to baumer_03_06. I do not recall that we voted to include those
>>specs into D2.4
>>
>> charles
>>
>>DAmbrosia, John F wrote:
>>
>>
>>
>>>All,
>>>
>>>I forwarded a thread to the reflector regarding issues with the ATCA
>>>channel data we had. This and other channel related issues that Rich
>>>and I have come across are shown in the following -
>>>
>>>___http://ieee802.org/3/ap/public/reference/dambrosia_r1_0306.pdf_
>>>
>>>Cheers!
>>>
>>>John D'Ambrosia
>>>
>>>
>>>
>>>
>>
>>
>>
>
>
>
>