/*-----------------------------------------------------

/* SPU HLSW V7.0 - WHAT'S NEW !!

/*

/* Project: 	HSO/PACS, SPU HLSW

/* Contact:	spuhlsw@prip.tuwien.ac.at 

/* Date:	02-12-2003

/*-----------------------------------------------------



SPU HLSW 7.0 is positioned as an upgrade of SPU HLSW 6.0, containing the following new features:   

- Bug for On-board Calculation of Compression Ratio corrected  

- The Use of on-board glitch detection in Spectroscopy is Optional (Commanded)  

- Add the Option for Data Reduction in Photometry Using Signed/Unsigned Integers  

- Separate Lossless Compression Methods for Photometry than Spectroscopy



This will be explained in the specific sections that follow.



The required Decompression SW version is:



- Paladin6 (Standalone Decompression Software in ANSI C)

- Decompress6 (Ground Decompression Software in Java) The SPU HLSW Development Team.



Bug for On-board Calculation of Compression Ratio Corrected ------------------------------------------------------------

The calculation of the compression ratio for the previous SW version was performed using the original size of only 2 matrices (512 Detectors).

In this version we put the right original data size (480 KBytes) for the calculation of the compression ratio that is part of the SPU HLSW HK.



The Use of On-Board Glitch Detection in Spectroscopy is Optional (Commanded)

-----------------------------------------------------------------------------

HLSW uses glitch detection and removal in spectroscopy in default setting. 

There is an option added in this version where the user, for any reason (e.g. use of glitch detection on-ground to have more CPU power for Lossless Compression), may deselect the on-board glitch detection. HLSW will then calculate the sub-slopes for N-samples without glitch detection. Therefore, we can have lower CPU workload that can be exploited by the reduction of the length of the sub-ramp (more subslopes can be derived). The use of on-board glitch detection is commanded through the command 'write detector constant in spectroscopy'. 



Add the Option for Data Reduction in Photometry Using Signed/Unsigned Integers

------------------------------------------------------------------------------

HLSW handles signed integers for data reduction (averaging and integration) in photometry in default setting. This setting fits with the default mode in photometry (differential mode). There is an option added in this version where the user may use the on-board functions (for averaging and integration) that handle unsigned integers. The use of unsigned/signed integer handling functions commanded through the command 'write detector constant in photometry'. 



Separate Lossless Compression Methods for Photometry than Spectroscopy

----------------------------------------------------------------------

To have better compression performance in term of compression ratio, we separate the lossless coding algorithms used spectroscopy from that used in photometry. For Spectroscopy,  we still use the same algorithms as that in HLSW Version6 while in photometry we use actually 3 runs of RZIP and the arithmetic coding. As we have more time interval for compression in photometry than that in spectroscopy, we can afford to have several runs of RZIP to have higher compression ratio.





The following c/h files has been changed since HLSW version 6:

- average.c

- bol1_2.c

- bol3.c

- bol4.c

- dxs.c

- integ.c

- llc.c

- ramp_ft.c

- spec1_2.c

- spec3.c

- write.c

/* --- EOF ---