Or 1/3 as you can run a Raid 5 with 3 disks minimum.

I do agree that Raid 5 is a much better solution to data lose prevention, as in Raid 1 you are just getting a basic mirror of one HD to the other so you have an identical mirrored backup of the data at all times, which means that if sone of the disks failed you have to rebuild the while raid set.

Whereas with Raid 5 the disks use block-level striping with parity data distributed across all member disks so when a disk fails it still goes on working, reconstructing the lost HDs data from the other drives seemlessly and when you swap out the dead drive for a new one it just rebuilds the data.

Raid 5 does however have two flaws. First it doesn't handle small files well. When you make a write to the Raid 5 array it has to read and write a whole parity block (stripe) from the set, so if a parity block is larger than the file being written it has to write the whole block each time, not just the file size. So in your case, if you are copying a lot of very small TOSEC set files it will really hit the performance as it will need to write a whole parity block for each file.

The other flaw is actually data loss. In normal operation the seemless reconstruction of data is brilliant, but where there is a single parity block per stripe (the extra disk), the failure of a second drive in a set results in total data loss. Obviously the likelihood of the failure of two drives at once is very low unless the Raid 5 set contains a large number of disks in the set.