[PATCH weston v3 3/3] Introduce wl_relative_pointer interface

[Jonas Ådahl]

On Thu, Oct 08, 2015 at 12:15:10PM -0500, Derek Foreman wrote:
> 
> But if it's a 1000hz gaming mouse the compositor will generate 1000
> events per second for the client to discard?
> 
> Why not just allow the pointer to be bound as either/both serial and
> absolute, and have each be a fully functional interface?

Forgot this part. I don't think we should have any duplications of
protocol like this. We'd make the API twice as big for no significant
wins.

If one doesn't want higher frequence relative motions, then one doesn't
need to get the relative pointer; and if one doesn't care for absolute
motions, then one can just ignore them. They will be fewer of them
anyway, since wl_fixed_t can't represent all the events that can be
represented via the double fixed type.


Jonas