From Scott Lurndal@3:633/10 to All on Thu Oct 16 21:07:41 2025
Assumption corruption.
----------------------
Fresh out of school, I was tasked with developing a new
system initialization sequence for a new version of the
Burroughs Medium systems MCP. This included a utility
that copied the installation media from a 9-track tape
to disk.
I implemented a double-buffer mechanism to overlap the
write to disk with the next read from the tape device.
I had assumed that all disk devices were faster than
all tape devices, so I didn't bother to wait for the
disk I/O to complete before starting the next tape
read.
The load worked great, and was much faster than the
prior utility. Until it was tested on an older
disk drive (207) that was actually slower than the
200IPS STC 6250BPI tape drive (1.25 MByte/sec).