The phototransistor method is similar to the work that I've done with the light dependent resistor, but with hopefully more accuracy. The theory is much the same, in the the amount of current that flows will depend on the amount of light the transistor is exposed to. However, the response times for the phototransistor should be in the order of microseconds.
The setup is similar to most other tests, an Arduino Giga with bespoke shield controlling the input devices (Brook UFB Fusions, again with custom PCBs attached to make controlling easier.) The touch screen allows for various different factors to be controlled.
The phototransistors are held against the screen with a custom rig, combining 8mm and 6mm threaded rods with 3D printed joins. This allows for precise positioning. Further specificity is attained via blinkers that allows for a specific area of the screen to be focused on (3mm size in the instance of Tekken 8 command history).
Tests are then conducted, with 1000 tests taking ~6 minutes. The advantage of this system is that the data is all ready for presentation as soon as it is done, either through a serial connection or written to USB.
The disadvantage is that there isn't any real review of data possible, there is no video file produced that myself or someone else could go back to review later.