^What he said, mostly. RAID-1 mirrors data on a pair of drives. If one dies, you don't lose data. Of course, data loss caused by file corruption (e.g. viruses or accidental deletion) are not prevented by RAID1, so backups should still be done. In RAID1, writes are slower because all the data must be written to both drives. Reads CAN be faster, since there will be times when each drive in the pair can be servicing a different request at the same time. The capacity is that of the smallest of the two drives. If the drives are equal size (the usual arrangement), capacity is equal to one drive.
RAID0 stripes data across multiple drives. Each drive handles less of the data, so reads and writes are both faster, BUT if any drive in the array fails, all data is lost. Total capacity is the size of the smallest drive times the number of drives. If all drives are the same size (the usual arrangement), that means capacity is equal to the sum of all the drives in the array (which can be more than two).
RAID0 also works most reliably with enterprise-class drives that have "TLER," which is Time Limited Error Recovery. If a drive takes too long to respond to a request, the controller (whether hardware or software) could believe the drive has failed, which kills the array. RAID1 is a lot less sensitive to this, since the drives do not have to respond precisely together on all requests like they do in RAID0. I run RAID1 pairs on both of my rigs, in one case using "LP" (same as "green") drives. I would not dream of using such drives in a RAID0, but this RAID1 has be flawless for almost two years now.