Dear gurus,
I am trying to run uvsfit but am getting this error message:
uvsfit: Revision 1.5, 2013/08/30 01:49:21 UTC
Reading the data ...
### Fatal Error: Buffer overflow
My input params are:
uvsfit% inp
Task: uvsfit
vis = ecdfs_allepochs_17_i.5500.uv
stokes =
line =
select =
object = point
spar = -0.01,0,0
fix = fxy
sspar = -0.7,0,0
sfix = 012
freqref =
out = foo.uv
options = residual
log =
The input uv file size is 395 MB, which doesn't seem that large.
I'm running the MIRIAD binary miriad-darwin_x86_64.tar.bz2 dated 27/8/2013 and run MACOSX 10.6.8
FYI I'm trying to inject a fake source into CABB 6cm data, following these instructions:
https://atcaforum.atnf.csiro.au/viewtopic.php?t=136
I have successfully added point sources into the uv data with uvmodel. It would be nice to be able to set the spectral index however, which I guess uvmodel defaults to 0?
cheers,
Minh
uvsfit error
Moderator: Mark.Wieringa
-
- ATCA Expert
- Posts: 297
- Joined: Mon Feb 08, 2010 1:37 pm
Re: uvsfit error
Hi Minh,
if you update you miriad version you should get uvsfit version 1.6. Unlike the previous version it doesn't try to hold all the data in memory when there is no fitting required, so you should be able to add fake sources to arbitrarily large datasets.
The point source option in uvmodel doesn't let you set the spectral index but you could generate an image with spectral index plane and feed that to uvmodel.
Cheers,
Mark
if you update you miriad version you should get uvsfit version 1.6. Unlike the previous version it doesn't try to hold all the data in memory when there is no fitting required, so you should be able to add fake sources to arbitrarily large datasets.
The point source option in uvmodel doesn't let you set the spectral index but you could generate an image with spectral index plane and feed that to uvmodel.
Cheers,
Mark