Low Latency Keyboards

[Andrei Chiffa]

A friend of mine, doing perception research in neuroscience, pointed out a keyboard they use in experiments with below 1 ms of latency:  http://www.empirisoft.com/directinkb.aspx. It's not only claimed by the producer - they validated it, and the same tests showed around 8-16 ms latency for almost any other keyboard due to USB polling rate.

 

I was wondering if anyone had experience with gaming with ultrafast keyboards in games along the lines of CS?

 

Conventional wisdom says that it should make no difference, the same used to be said about monitors going above 60 Hz and then above 120 Hz and I am curious of hearing about the experience of people who actually tested them, preferably paired with high refresh rate monitors.

[Oshino Shinobu]

Haven't tested it, but considering the difference between 16ms and 1ms is 0.015 seconds, it's really not something a human can perceive. 

[Andrei Chiffa]

24 minutes ago, Oshino Shinobu said:

Haven't tested it, but considering the difference between 16ms and 1ms is 0.015 seconds, it's really not something a human can perceive. 

yeah, I would guess the same thing, but on the other hand, a 60Hz monitor vs 240Hz monitor is actually 16ms vs 4ms of lag and is still perceptible/useful. Moreover, there are some research suggesting humans can perceive latencies as low as 2ms on simple tasks, for instance here and here.

[Guest]

I did pickup on this when doing a study last year in uni for a psychology paper when using reaction times, and time measurements for processing mathematics.

[Andrei Chiffa]

1 hour ago, RorzNZ said:

I did pickup on this when doing a study last year in uni for a psychology paper when using reaction times, and time measurements for processing mathematics.

Yeah - my friend is also in psy/neuro, but more in the delay between successive keystrokes. Did you have the opportunity to actually game with it? 

[Hiya!]

1 hour ago, Andrei Chiffa said:

yeah, I would guess the same thing, but on the other hand, a 60Hz monitor vs 240Hz monitor is actually 16ms vs 4ms of lag and is still perceptible/useful. Moreover, there are some research suggesting humans can perceive latencies as low as 2ms on simple tasks, for instance here and here.

I think refresh rate is more noticeable in this case cause it gives u more motion clarity and it allows for the game engine to run at it full speed without showing a really noticeable tearing, 16ms vs 4ms? thats not lag that is just how long frames are staying in your display. Input lag is the delay between your action and what happen on the screen.

So why high refresh rate monitor often make our action more responsive?

Thats because u are seeing every micro movement u make.

 

and my opinion on your question is Its just a gimmick that manufacture has to come up with in each of their new product.

[Andrei Chiffa]

On 2/7/2018 at 11:50 AM, Hiya! said:

and my opinion on your question is Its just a gimmick that manufacture has to come up with in each of their new product.

2

And LTT is just a gimmick by Nvidia to sell more cards. Especially seen the recent mining videos. <\sarcasm>.

For the rest, from the two links I've posted above, the more immediate feedback seems to be playing a role as well. And once again, I am expecting most of the lag time spent in the rendering pipelines of OS/application, as well as the mouse latency to have more of an effect than the keyboard. But the question has came up before (for instance here or here), had been used to sell peripherals (here) and I am yet to found an account of an actual testing yet (except for this guy working with editors, who claims to hit 1.5ms latency on GVim).

 

Edit 1:

Besides Linus touches on the activation times of keyboards in competitive gaming as limiting factor in his techquickie on PS/2.

Edit 2:

It looks that someone actually tried to measure keyboard latencies with an ultrafast camera here : they are realistically in the 20-60 ms range, mostly limited by the press travel time before activation.

[Guest]

6 hours ago, Andrei Chiffa said:

Yeah - my friend is also in psy/neuro, but more in the delay between successive keystrokes. Did you have the opportunity to actually game with it? 

I mean to say it was a confounding issue in the study to type out answers and in another the delay in keystrokes when identifying certian emotions on screen.

[Apoc79]

Sorry guys I know I'm necro posting AF but had to comment on this topic regarding latency perception. As FPV racer I can confirm that in most cases guys can visually see and notice a massive difference if you switch to a camera that has 2ms more latency and even so with the control interface part which is in most cases either 11ms (Spectrum / SBUS) and older 22ms controllers. This affects flight performance, reaction time etc etc. Once you start only using lowest possible latency equipment, flight almost becomes intuitive. Now in regards to gaming, I've been using a wireless keyboard for gaming with everything else as good as it gets, 144mhz screen, corsair mouse, and I can honestly say what a massive difference it made just by switch to the crappiest of cheapy wired keyboards. So depending on what the difference in ms is between a decent mechanical keyboard and an ultra low latency one, if its any more than 2ms, I'm sure it will be perceivable, by some of us at least  

[MakaDesu]

actually isnt just polling rate that matter ins delay

https://blog.wooting.nl/what-influences-keyboard-speed/

i wish i have more content about microcontrollers scan rate in keyboards project, im trying making a 2keys board with smallest delay as possible ( cause osu! , actually going from -10ms / +10ms instability to -7ms / +7ms in hits, would make i gain tons of ranks ) 

you can see it happening too in LTT video when they test the mouses delay with a  1000fps camera , and they be like "whooo, theres so many difference in delay, theyre not giving 1000hz poll as they say  " but actully it was at 1000hz poll, just different microcontrollers and coding