Tuesday, June 20, 2006

PostScript driver

Another couple of tweaks to the PS driver to fix bugs. The code included some instances of the
string "\\" which were assumed to be interpreted by the compiler as "\". This doesn't work on all compilers so replaced this with char(92). Also, the prolog was being written with two instances of "%% Orientation", one specifying Landscape and the other Portrait. This was confusing Illustrator on the Mac.

Keywords: HEAsoft, pgplot

Tuesday, June 06, 2006

more XIS contamination models

After prompting by Aya Kubota I've made a model xiscoabh for the HXD nominal pointing position and xispcoab with an additional parameter giving the off-axis angle. Both these use the expression for column-depth vs off-axis angle from Dotani-san's e-mail.

Keywords: Suzaku

max number of GTIs in extractor

Someone on the Japanese side of the Suzaku team ran into the limit of 40960 rows in a GTI file ! I've increased this to 100,000 but it would be good to remove this limit completely and go to dynamically-allocated memory.

Keywords: HEAsoft, extractor, xselect

Monday, June 05, 2006

rdrmf5 and variable length records

There were two cancelling bugs in rdrmf4 and rdrmf5. The code to find whether a column was variable length had an error so never returned true. However, the attempt to then read all the data for a variable length column in one go is incorrect use of cfitsio, which requires such columns to be read one row at a time. Fixed both these bugs. Note that some tools still call rdrmf4 - these should be updated.

Keywords: HEAsoft


Lazendic et al. present results from the analysis of a 70 ksec Chandra HETG observation of Cas A and describe the techniques used on such a large and complex target.

Loewenstein looks at heavy metal creation in field and cluster galaxies and concludes that star formation is >3 times as efficient in clusters as in the field, mitigating the overcooling problem.

Bogdanov et al. propose that the tail in the spectrum of rotation-powered millisecond pulsars is due to Comptonization of the thermal polar cap emission.

Benford models jets and magnetic bubbles to argue that magnetic structures can persist long after the jet has turned off and this explains the bubbles seen in the ICM.

Miller et al. show from X-ray observations of GRO J1655 that the absorbing wind must be driven by magnetic forces internal to the disk.

Blundell et al. determine the low-energy cutoff for the energy of relativistic particles accelerated in radio lobe hot spots.

Krivonos et al. analyze Integral observations of the Galactic Ridge and show that in spectrum and spatial distribution it is consistent with a population of accreting magnetic white dwarf binaries.

Saunders et al. discuss the optimal sampling strategy for searching for periodic signals.

Pizzolato and Soker argue that the R-T instability in radio bubbles in cooling flow clusters is stabilised by the deceleration of the interface in the initial expansion of the bubble increasing the expected lifetime.

Hogan detects the production of gravitational wave bursts by light cosmic strings.

Friday, June 02, 2006

fakeit in v12

Changes to support use of Gaussian statistics when simulating the model part of the source file. If there is a pre-existing dataset then its value of is_Poisson is used when faking. If there is no pre-existing dataset then Poisson statistics are used.

This is required for Swift BAT simulations but also should be useful for other instruments which have non-Poisson statistics.

Keywords: xspec, HEAsoft

Thursday, June 01, 2006

PGPLOT PS driver

Made considerable simplification to the GRPSSY routine in psdriv.f. This was very baroque and I was able to reduce the code length by a factor of three and generate cleaner PS. This also fixed a nasty bug in the plot sens command in xspec.

Keywords: HEAsoft, pgplot