min() vs limit optimization

[Paul Medynski]

So which of these is faster:

select min(StuffId) from Stuff where StuffId > X;

select StuffId from Stuff where StuffId > X limit 1;

where StuffId is the primary key of the Stuff table, and X is some variable. In my tests using an InnoDB table, both of these queries return the same result.

I read in the 5.1 Manual (Section 7.4.5) that min() is optimized when its argument has a unique key used in the where clause, but it only showed an example using equals (=). I need to know if that optimization holds true when using greater-than (>). In Sybase, min() is brain-dead, and using the limit is extremely fast.

-Paul