I used to have references to forensic papers claiming one pass is enough and forensic papers claiming one pass is not enough. Today it seems that pretty much everyone agrees that one pass is enough, and there have been forensic papers that demonstrate this to be true. So I am inclined to agree that one pass is enough. The issue with not using ATA secure erase is that even with 35 passes, there could be 'magnetic residue' on the edge of tracks. Normal wipe programs bring the drive head down the track down the center and don't overwrite data on the edge of the track consistently. ATA secure erase puts the head off track by a few degrees during the second pass so it erases this data. It is technically one pass in both cases though, just the second pass is done with the head at a different position. There is a forensic paper showing that data can be pulled off a drive with multiple pass overwrites that don't use ATA secure erase, but the issue isn't due to a single pass not being enough. csrc.nist.gov/publications/nistpubs/800-88/NISTSP800-88_rev1.pdfcsrc.nist.gov/publications/nistpubs/800-88/NISTSP800-88_rev1.pdf