[Almacal] Re: [mmaimcal] Meeting today

[Stéphane Guilloteau]

A 08:42 11/05/2005 -0400, Al Wootten a écrit :
>Hi Stéphane
>
>Thanks.  Indeed that is a good point.  I'll talk to Skip about this but
>I think it is really the magnitude which causes problems.  It is no problem
>at the moment because they are using a x6 multiplier for B9.  However, I
>am told that there is a x5 solution 'almost ready' for implementation, and
>that for this it is very difficult to switch by 0.03%.  Since we cannot
>ascertain lock at 10ms anyway it would be quite reasonable to relax that
>somewhat.
Well, the 0.03 % shift problem is really for band 3 and below: how does 
Band 3 perform in this respect ?
One solution would be to split the spec as
         - XX ms for a shift of 0.05 % for Band 3 and below
         - YY ms for a shift of 0.02 % for Band 4 and above

         Cheers,

                 Stéphane