Going from memory, so there could be minor errors: Fastnet was a replacement for serial-based transfers. The VAX used a DR11-W to communicate with an Ungermann- Bass controller. The UB was a big box with several Z80- based boards. The UB ran CP/M loaded from 8" floppies over HPIB. The Z80 boards were all the same. Each board had a daughter board with different types of interfaces. The master board had a serial port, parallel port, and HPIB interface. The board that communicated with the DR11-W had four parallel ports. There were also boards with six serial ports and a board with an Ethernet port. Each Z80 board was booted from the master board with a customized CP/M system. The remote UB boxes had the same board types as in the master UB box. The remote UB boxes downloaded their software over Ethernet. The ATE systems connected to the remote UB box through one of the Z80 boards with a parallel daughterboard. Each remote box could connect up to three testers (requiring three separate Z80 boards with parallel daughterboards). The remote boxes could also have one of the Z80 boards with a serial daughterboard to connect to the serial terminal stations on the testers, and software allowed access that connected to the remote UB box looked like a serial board to the tester (and was really just a serial board without the UART, so the data lines went direct to the remote UB; that's where the speed increase came from). The UB network used an old version of the XNS protocol. We had Sentry 20 and 21 testers. The processor in the testers had a ~900KHz clock. The operating system was M3 and the test programs were written in FACTOR. -- Eric Dittman dittman(a)dittman.net