In 1986, I was just starting up the ARRI Lighting Control business in London. There was me, Peter Ed, and an engineer called Mike Hayward, and we were working in the kitchen in one of ARRI's buildings near Heathrow Airport. I remember our first product was a range of multiplex and demultiplex devices we called Connexion. Dimmers were still mostly analog then, and these were hot products.

I was keen to jump on board with the new open protocol that I knew was being worked on in the States. But there was disagreement within the committee as to the eventual baud rate. As I recall, Steve Terry was pushing the 256kB that was eventually adopted, but there was a strong holdout for 192kB from a manufacturer who didn't want to change.

I remember clearly talking to Steve one day that summer, when we were right at our go-no decision point for the Connexion product. I was strongly against the 192kB camp, but I was nervous about committing to 256kB while there was a possibility that I might be out of line when the standard was eventually published. Steve and I agreed that I would go ahead, and that he would use our commitment to 256kB as a fait accompli to swing the committee to the faster baud rate. It worked, and I believe the ARRI Connexion products may have been the very first DMX devices to hit the market.