I winn answer this, but we have been going back a forth for a while, so
perhaps someone else has a better answer than me.
-------- Original Message --------
Subject: Re: [daip] aips
Date: Thu, 10 Nov 2005 16:10:27 +0800
From: zzwu <zzwu at shao.ac.cn>
To: Amy Mioduszewski <amiodusz at nrao.edu>
References: <1127407309.4562.13.camel at wu> <4332EBC4.1070005 at nrao.edu>
<1127547532.4287.16.camel at wu> <433DCC16.10701 at nrao.edu>
<019001c5dfaf$dff91100$e61d7fca at zhongzum1bsaxx> <4368F96E.60700 at nrao.edu>
Dear professor Amy Mioduszewski:
I have send you an email about my solint :
solint =4 4853 good solutions 1307
failed solutions
solint=5 4853 good solutions 1307 failed solutions
solint=6 3562 good solutions 750failed solutions
solint=7 3562 good solutions 750failed solutions
solint=8 3562 good solutions 750failed solutions
solint=9 2095 good solutions 313failed solutions
solint =10 2095 good solutions 313failed solutions
solint =11 2095 good solutions 313failed solutions
solint =12 2095 good solutions 313failed solutions
and you said that I should choose as short as possible ,and calculate
the percentage of failed solution(failed soln/good soln),and choose the
solint based on the lowest percentage of failed solutions),I have tried
all the solint values ,and applied them to different cl tables and split
them seperately,but I didn't find nothing different ,from solint 4 5 6 7
8,the splited data is nearly the same,solint =9 10 11 12 15,the data is
nearly the same, and if selected solint 4 5 6 7 8,the dada will be
some less than the dada select solint 9 10 11 12 15,I use difmap imaging
the data,I compared them with the command vplot,and find that,can you
tell me how can it embody the sensitiviy and how the lower percentage
of failed solutions is good,than you very much.about image the data ,I
only you difmap,I think you must be familuar with that software,so I
want ask you that how can I make sure the structure is real,maybe it is
too gerenal question,but I do not know what is the reason ,I image
several times ,and modelfit it ,few time of them is the same ,how could
I be sure about that ,Thank you very much.
cheers
wu
> Hi Wu,
>
> I would choose as short as possible, that does not decrease your
> sensitivity so that you get extra failed solutions. So what you should
> do is calculate the percentage of failed solutions (failed soln/good
> soln), and choose the solint based on the lowest percentage of failed
> solutions.
>
> VLBAMPCL was fixed several weeks ago. If you are on the midnight job
> you should have the fix. If you are not you can run it in such a way
> that the bug will not be triggered (i.e., do not fill the entire
> ANTENNAS parameter with numbers, type ANTE 5,0 instead of ANTE 5).
>
> Cheers,
>
> Amy