Owing to the huge growth in disk sizes & the chance of an unrecoverable read error during a single parity rebuild, I think it would be wise to default to DDR if the unit is initialized with large drives e.g. >=2TB.
Playing with the RAID5 rebuild calculator here will enable anyone interested to see why. With consumer grade drives which typically have a URE rate specified in the 1 in 10^14 class beyond 2TB the chance of a successful rebuild drops below 50%.
While that calculator is pessimistic since it uses URE rate “Equal to” & manufacturers specify URE rates of “Less than” it’s getting into really sketchy territory rebuilding 5 drive single parity arrays with modern big drives.
Even if Beyond RAID handles a rebuild more intelligently than a cheap hardware card & only rebuilds the data & parity for the one failed drive (making rebuilds equivalent to a 3 drive array with drives of the size of the failed one) AND only rebuilds for occupied space rather than free space as well which the faster rebuilds on emptier units would indicate is the case, your chances of a successful rebuild still drop below 50% once a consumer grade failed drive contains more than about 4TB of data.
A warning in the dashboard UI that single redundancy is unwise with large disks if it’s subsequently selected would also be useful.