I am conducting research to validate claims of millisecond accuracy in psychological testing software. Up front, I am not a programmer, and have a student working to write code to have a signal sent to the serial or USB port of the computer in C (of C++), Java, and Flash. The idea is to test the ability of the OS to handle a 10-millisecond pulse with increasing layers. The hypothesis we are testing is that personal microcomputers are not capable of accurate millisecond timing unless extremely complex operations are built in to the software that allow one to turn off internal operations in the computer.
I hope to do this in Windows XP as well a Mac OS X. So far, we have an .EXE that allows me to measure a pulse from the serial port in Windows. This “program” asks for the length of the trial and the type of pulse (single or multiple) and appears to work as validated by an oscilloscope.
We appear to have hit some walls accessing the serial or USB port through Java and Flash. Anyone with programming experience that can explain things in non-programming terms want to chime in?