Just something to keep in mind when you create a database using an existing FMT of a UTF-8 database. You might want to do that to copy a database or to re-create a corrupted database.
When the original database uses UTF-8 and you make a DMP-file to transfer the records, the resultant file will contain data in UTF-8.
But when this DMP-file is read into a newly-created database, even when that database is based on (a copy of) the original FMT, all data will be converted back to the Windows character set. So, for instance, any accented characters that do not occur in the Windows character set will be stripped bare.
Obviously the fact that a database has been converted to UTF-8 is not stored in the FMT and Cardbox does not warn the user that reading a UTF-8 DMP-file into a Windows database will result in loss of data.
Always check that a new database is UTF-8, and convert it when necessary, even when you think it's already UTF-8 because you've converted it before.