[evla-sw-discuss] Simple subarrays: a modest proposal

[Michael Rupen]

>
> Just to jump in on the dumptrig stuff:
>
> It may help to not think of the dumptrig originating from a particular 
> Station board as needing anything to do with that station. Think about using 
> rack 8 to generate DTs for all the baseline boards and having the boards 
> source the DT needed for that sub arrays dump schedule.
>
> -Bruce
>

Yes, it's this that led us to think about getting more than 16 effective
DUMPTRIGs by generating different DUMPTRIGs on different Station Boards.
Or to access more than two "big" DUMPTRIGs by developing them on multiple
StBs.  Both seem complex but in principle do-able.

The question is whether there are restrictions on routing said DUMPTRIGs
from the StBs if we make up different DUMPTRIGs on different StBs. (I'm 
assuming that we're not crazy, er, clever enough to make up the DUMPTRIGs
for multiple subarrays on a single StB -- I realize this is possible but
my brain hurts.)  I don't _think_ there are important restrictions here
but I couldn't swear to it in court, without looking at the documentation
closely again.

Of course this is not directly relevant to the proposed way of handling 
subarrays, if we are happy with living with some restrictions which don't
really hurt the science, until we have the leisure to be certain we understand 
the possible constraints imposed by hardware & software (OK, probably firmware 
too).

            Michael