The connector on the bottom of the controller is a normal female header. The kind that you would solder onto a PCB, not kind you would scavenge from a ribbon cable. I used a Dremel to cut the hole in the bottom and glued the header in place. I used a 2x15 header, but in hindsight realize that I should have used a 2x16 instead. Since everything is in fours, it would have made for much neater wiring matrix (4 buttons, 4 trigger/bumpers, 4 dpad, 4 analog pot wipers, etc). That would have left a couple more connections available for stuff like status LEDs in the actual controller (i.e. when a script is running, etc), or additional tact buttons for additional scripts/actions.
I am going to be opening up the controller to add some tact buttons to the front of it. I will take some photos then.
This is in preparation of a much 'smaller' v2 shield that won't have a display or the large dpdt switches. I am thinking about using a couple shift registers to reduce the number of Arduino pins needed significantly, and possibly go with a Nano/Mini/Ardweeny and 2 small stacked shields (1 for the Nano, the shift registers and the AD5204/6, and the other for the optocouplers/CD4066B?). I hadn't used shift registers until this project, where I didn't have enough pins for the 2 digit display. I suspect that they would be plenty fast to handle the controller 'buttons'. And, with a few 74HC595 chained, one would really only need 3 pins for all of the buttons (would want to control OE so as to avoid the random button presses when powered up). Plus another 4 or 5 for the digital potentiometers. And, a couple digital input pins for the tact buttons. Nothing a Nano/Mini/Ardweeny wouldn't be able to handle.
But, I am not going to start on that until the digital potentiometers arrive and I have had time to tinker with them. Having full analog control of the analog sticks would be worth the v2 'upgrade'. Otherwise, the project is working fine, so why tinker with it. 