Binary+Fastload (uses a huge file to speedup loading database into ARB; whether there really is some speedup depends on your system and machine, try it out)
Ascii (human-readable; extra HUGE)
Select extra compression:
None (compatible with arb 6.0.x and older)
otherwise ARB uses the selected command line compressor (like gzip, bzip2, xz) to compress the database during save. When loading a database, arb detects if a database is extra-compressed and automatically selects the correct decompressor.
Note: you may rename an extra-compressed .arb file into .arb.gz or .arb.bz2 before
giving it away to a pre-arb-6.0 user. He will be able to open it using
an archive manager and extract the contained .arb manually.
You may also save only the changes done to the database, using 'Quicksave changes' or 'Quicksave changes as' (accessible via the DB name button in the ARB_NT top area).
'Quicksave changes' saves all changes to the database 'data.arb' to a file 'data.aXX' where XX is a numbered from 01 to 99 and 00 (means 100).
This is normally much faster than saving the entire database and results in much less disc consumption.
Each time you save your changes, XX is increased by one. The last 10 data.aXX do not get deleted instantly, allowing you to step back to an older state.
All data.aXX are independent and you may delete old changes files without losing data (as long as the latest is unchanged).
When you start arb you can choose among the different files:
enter 'arb filename.arb' to search and load the latest changes,
enter 'arb filename.aXX' to load a user-defined version.
Loading a specific quick save is also possible from the ARB INTRO window. To do so, enter '*' as suffix.
Some database changes would cause huge-size quick-save files. Therefore ARB disables the ability to quick-save the database after several operations (e.g. when changing the order of database entries). If you try to quick-save, you will get an error message.