In 1977, I went to Houston to visit Exxon Production Research at the annual presentation of their research efforts. To a young geophysicist, this was indeed a plum assignment, but I had to present what I had seen to an amphitheatre of seasoned Esso veterans when I came back. Daunting as this was, I talked for about half an hour, then I summed it up by saying that this was not the most interesting thing that I had seen. During a break, I came upon a door with a window through which a light was shining. Upon peering in, I saw a velocity analyst picking gathers on a computer screen. Once he had picked one, he could press a button and the gather was redrawn flattened. The entire set-up had to be hard-wired through a thick cable to a mainframe computer less than 20 feet away. ‘Some day,’ I said, ‘we will put our coloured pencils and our erasers away and pick our seismic data on computer screens.’ The hall erupted into laughter. That was the most outrageous thing they had ever heard!
My wild-eyed fantasy did not happen for some time and by that time I had moved to Canadian Hunter. The two main drivers for the emergence of the workstation were the advent of 3D seismic surveys and personal computers in the early 1980s. When 3D seismic arrived, we didn’t know what to do with it. There was so much data! The processors would send us booklets of inlines and crosslines, and we would pick every tenth one in both directions (again with coloured pencils) and transfer that to a map. The rest (90 percent of the data) was ignored. Into the gap sprang Landmark Graphics, with the first commercial computer workstation for seismic interpretation, on a Sun Microsystems workstation running UNIX.
These machines were quite expensive, so only the larger companies could afford them. They were shared assets, so we had to book time on them. But they could do amazing things. We could choose the data we wanted to pick from a basemap on the screen. They had autopickers which could extend your picks on good reflectors to every line and trace in the volume. Your picks appeared on the screen in colour, and you could map the results. For the first time, geophysicists could make amplitude maps and infer lithology changes from those. This revolutionized interpretation, but the era of personal workstations was still a ways off.
However, I promptly marched in to my boss, the brilliant head of Canadian Hunter, John Masters, and asked if the company could buy one for me. John passed away in 2011 and was eulogized for his uncanny ability to see trends coming before anyone else. But this was perhaps the one time he missed the boat. I was told that they were too expensive and that he liked the way I interpreted data just fine as it was. So I decided to build my own workstation. Two wonderful guys helped me. One was George Palmer of Veritas Software, who wrote the interpretation code in an application he called Vista. The other was an electronics genius and alpha geek named Paul Geis. He designed the graphics card that could display seismic data. Together, we built a workstation on a 386 PC, circa 1984. It was state of the art with a 9-track tape reader for loading SEGY data, one of the first optical drives for storing it, and a Bernoulli Box with removable 20MB platters for storing horizons.
It wasn’t perfect. It did not have the ability to work from a basemap — data was picked from a list. And it couldn’t map data, instead picks were exported to a Perkin–Elmer minicomputer for contouring. However, I picked over 500 lines with it and Hunter made several oil discoveries in Saskatchewan as a result, including the prolific Tableland and Kisbey fields.
Unfortunately Veritas Software broke up shortly afterwards, the Vista software became the property of one of the emerging companies, Seismic Image Software, and I had to say goodbye to my first, and very own, personal computer workstation.