USB or Keyboard Polling Rate

Hi,

The OS X kernel is designed to be preemptable, so it is possible to possible to schedule a low-latency real-time thread that can run in a quasi-realtime deterministic manner.

I have a feeling you're going to have to do some of the coding yourself to do what you want to do. This article, though old, gives some hints in sections 5.4 and 6: http://www.usenix.org/events/bsdcon/fullpapers/gerbarg/gerbarghtml/

You might want to take a look at OS X programming relating to MIDI I/O since that requires accurate timing.

Other things to keep in mind if you are measuring response times of someone pressing a key in response to a stimulus - what is your display's refresh rate and latency. Also, how deterministic is the keyboard scanning inside the keyboard? If your doing a psych experiment, using MIDI to trigger a stimulus presentation and read a button press may be the best way to get the timing accuracy you want.

(Apologies for the re-post. My cat stepped on the keyboard and posted the previous article before I could make grammar corrections -Bill)

Hi,

The OS X kernel is designed to be preemptable, so it is possible to schedule a low-latency real-time thread that can run in a quasi-realtime deterministic manner.

I have a feeling you're going to have to do some of the coding yourself to do what you want to do. This article, though old, gives some hints in sections 5.4 and 6: http://www.usenix.org/events/bsdcon/fullpapers/gerbarg/gerbarghtml/

You might want to take a look at OS X programming relating to MIDI I/O since that requires accurate timing.

Keep in mind when measuring response times of someone pressing a key in response to a stimulus: 1. What is your display's refresh rate and latency. 2. How deterministic is the keyboard scanning inside the keyboard? If your doing a psych experiment, using MIDI to trigger a stimulus presentation and read a button press may be an easy way to get the timing accuracy you want.

-Bill