Writing SCSI 9-Track Mag Tapes from Windows/DOS

On 10/02/2015 11:28 PM, Marc Verdiell wrote: Thanks for the kind words. I didn't do the reverse, as my interest lies primarily in retrieving data from old media, not creating more data on old media. Generally, I make the SIMH file from a tape and have a bunch of utilities to process the file for conversion to modern formats (code translation, deblocking, un-archiving, etc.). That being said, there are routines in that SCSI package for writing tape. They were borrowed from an old forensics package that had to pretty much work with any SCSI device, so I'm pretty confident that they'll work. They were replaced about 8 years of field use with a C++ version (much easier to handle the minutiae of tape when encapsulated as a C++ object). I'm sure that John's st() utilities are equally valuable, but I never could get them to work with my setup. The general problem with UNIX/BSD/Linux is that the philosophy there is that everything is basically a disk file, however physically embodied. Tapes are not disks. In general, I don't like most SCSI tape drives for data retrieval, as the SCSI interface seems to require that implied read-ahead be used to keep the tape moving. There's nothing quite like watching a SCSI tape drive going through a "scrubbing" cycle trying to read a tape block that you never asked for. --Chuck