Hi Leo,

Leo Weppelman wrote:
[...]
> > How come that the 1K-sector is so much slower on output, but about
> > the same on input???
> 
> Yes, weird isn't it... I have run multiple tests because I didn't believe
> it at first, but they were pretty consistent. For completeness, I think
> these tests should be repeated on more than 1 type of disk. It might have
> been a problem with the maxtor firmware. When I started playing with bigger
> sector sizes, I had the idea that 1K blocks should be slightly faster...

Maybe this is due to the write cache being smaller? Maybe the cache
is not optimized for larger blocks?

Would be interesting to see tests with other drives, maybe combined with
the
information on how the cache is set up on these drives...

  Guenther