Keyboard Latency

It's an ok read but...

>Also, the idea that it's fine that keys have a bunch of useless travel because you can pre-depress the key before really pressing the key is just absurd.

It isn't useless travel. If travel was useless we'd all use touchscreen-like or mice-button-like keyboards. The Apple keyboard keys are the worst possible keys for gaming. Lots of competitive gamers switch out switches in their mice because they get misclicks just by resting a finger on a button. Travel might be too far on some keyboards but it is way way worse using something build like the Apple keyboard. Many (all I know) competitive gamers also add rubber rings to their keyboard switches which adds to the feel of how much (little) keys needs to travel.

IMO it is wrong to measure latency including key travel because of the above and because we don't move our fingers equally fast or precise. This means no-one can get the same result every time and never across different people (repeatable and reproducible results). It basically make the tests in the article useless unless it is the only test or the only tested keyboards, IMO. There's a good reason input latency is not normally tested including human movements.

This is a good-but-old post from Dan Luu. There's no date on the article, but an entry in his Atom feed from 2017 references this post. I doubt much as changed since then, so his findings are likely still relevant.