[evla-sw-discuss] OSRO OPT - Integration Time

Bryan Butler bbutler at nrao.edu
Thu Oct 15 16:53:37 EDT 2009




Michael Rupen wrote, On 10/15/09 13:45:
> 
>> In Tuesday's SSS meeting i promised i would ask some questions of a 
>> wider audience.  This
>> is the 2nd of 3 emails.
>>
>> On the OSRO page the following is said about integration times:
>>       "These capabilities will be provided with integration times no 
>> shorter than 1
>>       second..."
>>
>> In the OPT for the VLA correlator integration times were presented to 
>> the user as a
>> dropdown list of only a handful of valid values.  Some questions for 
>> the OSRO user
>> interface of the OPT:
>>  1. Will we display a discrete set of valid values or allow users to 
>> type in values?
> 
> Probably easiest to allow users to type it in.  I think we should 
> restrict this to multiples of one second however (i.e., make this an 
> integer).
> 

agreed with michael here - allow them to type it in, but round it to 
integer seconds (with a note as such, because undoubtedly folks will 
like to type in something like 2.5).

> 
>>  2. One second is given above as a minimum time.  Should we 
>> nonetheless allow input of
>>     values < 1s and issue a warning, or truly prohibit it?
> 
> Truly prohibit it.  We do NOT want to deal with short integration times 
> for OSRO -- it stresses the CBE, and ups the data rate.
> 

for OSRO, yes, prohibit it.  for RSRO and ECSO we need to allow it of 
course.  one potential problem i see is that the OPT will be the only 
mechanism (other than hand-crafting scripts) for driving the array, and 
if staff scientists want to use standard OSRO modes, but with short 
dumps, this will cause them some amount of grief.  that grief is that 
they will have to hand-edit scripts to modify the correlator dump time. 
  which means they will have to have access to the model2script program, 
run it, and then modify the script, and submit the *script* for running 
on the array, rather than using the standard mechanism.  staff will need 
access to model2script anyway, but i just point it out as a potential 
hiccup in completely disallowing integration times < 1 sec for OSRO setups.





More information about the evla-sw-discuss mailing list