Yeah, that's a good start.

It wouldn't be too hard to code something from scratch either for someone with coding experience. For a naive method, record the start time and check the system clock until it's N seconds later. Meanwhile append the key presses to a string. At the end, report string length divided by time elapsed.

One other thing I have to worry about is the up and down signals and Keyclone's latency. It's slow enough that key mashing isn't always a great idea. Maybe I'll get a hardware mux eventually.

I'm a little curious about this now. Maybe I'll whip up a quick VI...really the reason I didn't yet is that it only runs on my laptop, not the nice mechanical keyboard I use to play on.

Counter-edit edit:

Cool. Sounds more believable than the 20+ he was claiming. It sounds in line with the research paper I found - I think they were using a telegraph key or something like that, so the shorter stroke length of a keyboard would speed it up a little.

If you really, really need it to go faster, you could always get a mouse with a loose wheel...still a technically legal method, although it might give you a high enough rate to seem suspicious.

Another edit:

Acoustic analysis might be the most effective. And you could do it with a decent microphone and free sound editing software (although you'd want more for a less simple analysis).


Also, lest there be any further confusion about my initial claim, let me clarify a point of semantics:

couple, n. 4: Informal. A few; several: a couple of days.