A better benchmark will be using iozone from www.iozone.org.
It simulates many read/write scenarios, not only with different file size, but different record-size per read/write, for sequential, random, reverse, buffered (fread, fwrite), create and re-read/re-write.
It also take into account the test machine main and cache memory size.
With this, you can see the effect of using a drive with 512 bytes sector and that with 4K sector.
Definitely a lot more objective.