0

I plugged in my old PS/2 keyboard while getting my USB keyboard clean and dry, and I think I feel it has lower latency, and I want to test the latencies of the keyboards now.

According to the test at Keyboard Latency Test, the "shortest key press" of my PS/2 keyboard took 26 ms and the "estimated scan rate" of my PS/2 keyboard is 38.4615 Hz though the vast majority of the key presses took more than a 100 ms and the longest key press was Ctrl with 1204 ms followed by divide with 946 ms on the PS/2 keyboard.

I wonder if these numbers are accurate and whether the test will be accurate with my USB keyboard when it's ready.

Is there a better test that I can use instead?

I'm a Windows user in case it matters.

Improve this question
4
  • 2
    To my mind, all that test can tell is how quickly you can lift your finger after pressing + network lag. If I bounce a pen off a key, it's far shorter than & can do manually. I don't see it as vaguely useful. There's no way for it to tell how long between you physically closing a contact & it registering on a) your system or b) a server half a planet away.
    – Tetsujin
    Commented Jul 5, 2023 at 9:30
  • @Tetsujin Where does network lag come from if it's a client side JavaScript test?
    – grawity_u1686
    Commented Jul 5, 2023 at 10:47
  • @u1686_grawity - wasn't aware of that. It still can't tell you the time between you hitting a key & it receiving the data; only the time it takes you to let go again. I could get 30ms with my finger, 20ms bouncing a pen off it.
    – Tetsujin
    Commented Jul 5, 2023 at 10:53
  • While the test can measure the polling rate, it cannot tell if there is a delay in the pipeline, i.e., due to a fixed delay between the time contact is made and the signal reaches the PC. You'd need a mechanical device to press a key, and measure the time between physical press and OS getting message. What counts, ultimately, is your perception of keyboard responsiveness in actual use.
    – DrMoishe Pippik
    Commented Jul 5, 2023 at 15:14

1 Answer 1

Reset to default
3

While the test can measure the polling rate, it cannot tell if there is a delay in the pipeline, i.e., due to a fixed delay between the time contact is made and the signal reaches the PC. You'd need a mechanical device to press a key, and measure the latency between physical press and OS getting message. What counts, ultimately, is your perception of keyboard responsiveness in actual use.

Consider how an organist plays: there are various delays from the time the key is pressed until sound is perceived, anywhere from 30 to 150 ms for mechanical and pneumatic organs! The organist apparently learns to ignore aural feedback, relying on their internalized timing. So if you use a particular keyboard, and you've learned its idiosyncrasies, that might feel best to you, until you accommodate another.

That said, wireless keyboards add an extra layer of transmission, reception and decoding, which might slow things somewhat.

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .