Subject Re: Character Sets
Author peter_jacobi.rm
Hi Uwe,

I agree with Henrik, that ISO8859_1 would
be good choice. Unicode support has it downsides,
just scroll some days back in the ML archives
to find a discussion on this.

--- In firebird-support@yahoogroups.com, Uwe Oeder <uweo@c...> wrote:
> Would that not mean that if I have a Field of VARCHAR with 100 length
> then the actual bytes stored would be amount of multiple bytes *
length of
> VARCHAR field ?

Yes CHAR(10) charset UNICODE_FSS
allocates 30 bytes. It is compressed on disk,
but for several issues it surfaces, e.g.
Indices.

Regards,
Peter Jacobi