While we were assembling the hardware and software, Scott’s dad, Aleck, had tracked down a source of SeaMark side-scan sonar data (the University of Washington Marine Survey Group) and had obtained three minutes of data on about fifty 5 1/2″ floppies. Scott had copied this analog data to a WORM cartridge, and wrote a program to read it into the the computer and process it.
One morning he called to invite me over to his folks’ house in North Seattle. I was introduced to his dad and mom and then set down to see what he had done. He explained the process as he booted up the computer and loaded the WORM cartridge, and then he typed a command at the C: prompt and pressed Enter to start his application. What happened next was visually stunning: What he called a “waterfall display” rippled down from the top of the screen at the same speed that sonar data was originally sent from the towfish to the boat. Each horizontal line on the monitor represented one sonar sweep. Each line was divided in the middle, with the returns from the port side of the boat on the left side of the monitor and those from the starboard side of the boat on the right side of the monitor. Each pixel to the left and right of center represented one returned sonar “ping.” The color of each pixel was determined (in Scott’s code) by the strength of the return: Scott explained that harder objects returned a stronger signal, so rocks and metal man-made objects produced the strongest returns. By mapping these strong returns to bright colors, and weak returns, for example from sand or mud, to light colors, the more interesting objects could be made to stand out on the display.
We replayed the three minutes of data over and over, comparing this to the standard sonar display we had both seen, and talking about how much easier it would be to identify objects on the bottom if you had this system.