SCALE Software Coordinator
This topic has been migrated from the SCALE 6 notebook.
Date: Wed Jun 23 16:50:00 2010
I am new to SCALE 6 but have used all the previous versions for many years. I use many of the individual codes in standalone mode, but I have minimal experience with the automated sequences (and I hope to keep it that way since I am somewhat of a control freak and I like to know exactly what goes into and comes out of every code -- and running the codes in standalone mode gives me a better level of understanding...)
The first step in most of my analyses is to use AJAX to extract the data of interest from the large SCALE master libraries -- and then I work with the smaller subset to do additional processing and analysis as needed for a particular study. The problem is the routine AJAX sequence that I have used for years will not work on the SCALE 6 master libraries. I checked several paths, as follows:
1. Both the SCALE 6 and SCALE 5.1 versions of AJAX work fine with the SCALE 5 master libraries, but they both fail while trying to read the SCALE 6 libraries.
2. I have tried several libraries: the 44 and 238 group ENDFB-V libraries in SCALE 5 and the separate VITAMIN-B6 library (DLC-184) all work (but not so for any of the SCALE 6 libraries).
My questions are: Should this behavior be expected? Was there a format change that I am not aware of? The fact that AJAX in SCALE 6 reads the SCALE 5 libraries is confusing -- which makes me think there was no format change? However, the SCALE 6 sample sequence util.input to test out many of the AMPX utilities seems to work fine with the SCALE 6 master libraries (tried several), but this starts out with a "csas-mg parm=chk" run and then calls AJAX and several other utilities. I am not sure how the first step affects the master library read by AJAX in this sequence (but it works just fine)...
Well, with the above observations, does anyone have some suggestions to help resolve my issue (preferrably one that does not involve using csas-mg or other control modules).
Thanks for your help...
John White
SCALE Software Coordinator
John,
Endianess describes the bit order of data in a binary file. Some operating systems default to the most significant bit first (big endian) and some default to the least significant bit first (little endian). Either way, the data get read into memory, but if the wrong bit order is assumed, the data will be meaningless. Here’s the Wikipedia page on endianess if you’re interested: http://en.wikipedia.org/wiki/Endianness
Fortran allows a user to force an endianess condition. For historical reasons, SCALE data files are formatted as big endian.
In earlier versions, SCALE operated as little endian on Windows after reading the big endian data files. Now the scale6 command file automatically exports units 21, 22, 23, 24,25, 26, 33, 77, 78, 79, 80, 88, and 89 as big endian. Therefore, if the input unit to AJAX is any of those units, the correct format is automatically selected. By the same token, if you use any of those units as output unit, the output file will also be big endian. All SCALE sequences use the units in such a way that the data libraries are read on one of the above units. If your AJAX is not using one of these units, the output file will be little endian and will cause problems downstream in SCALE.
Please let me know if you’d like additional clarification.
Best Regards,
Brad Rearden and Doro Wiarda (AMPX Developer)