Stories
Slash Boxes
Comments

SoylentNews is people

posted by LaminatorX on Thursday March 06 2014, @12:20PM   Printer-friendly
from the satisying-clackity-clack dept.
An anonymous coward writes "Anyone know of good affordable keyboards that are low latency (preferably backed by actual stats)? Low latency is not the same as polling rate.

I had an old keyboard that was high latency (added about 30-50ms more latency when compared to a "gaming" mouse I had!) so I bought a low end "gaming" keyboard[1] which is lower latency but the keys "stick" sometimes (e.g. the system thinks keys are still being held down even though they aren't have to press the offending keys again to unstick them). I don't want to buy an expensive keyboard and find the latency to not be really much better or even worse[2]. And yes 30-50ms can be a noticeable and significant difference in games (2-3 frames).

I've done those reaction time test stuff and I get about 150-170ms using my "fastest" mouse (I have two), 170-190 with my new keyboard and 200+ms with my old keyboard. I see many people get 200+ ( see: http://cognitivefun.net/stat/1 ). At work on my employer's macbook pro I get 220+ms. So it's likely that high latency mice/keyboards[2] and screens[3] are too common. And you can appear to have 50-80ms faster reflexes just by having better equipment.

[1] an A4Tech G800V keyboard, based on one of the few less useless responses from the Other Site when I asked a similar question. Maybe it's faulty but it's going to be hard to prove since it's intermittent. FWIW I got it for half the newegg price and the place I bought it from doesn't sell A4tech mice or keyboards anymore.

[2] http://www.blackboxtoolkit.com/responsedevices.htm l
  http://www.pstnet.com/eprimedevice.cfm

[3] http://www.displaylag.com/display-database/"
 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2, Interesting) by Anonymous Coward on Thursday March 06 2014, @01:09PM

    by Anonymous Coward on Thursday March 06 2014, @01:09PM (#11899)

    Why is there a latency at all for a normal keyboard? isn't the key down and key up events sent directly when they happen? I really don't understand why there be any "polling" involved...
    is that some way to make it cheaper to produce and save on some of the wiring (I mean the lines to keys painted on the plastic sheets in the keyboard that you connect when pressing a key) or on the little chip in the keyboard?
    I know there is some stupid keyboards that have severe limits on how many keys (and which) can be held down at the same time but that is a complete other problem.

    It could be interesting to read if someone here knows the inner workings

    Starting Score:    0  points
    Moderation   +2  
       Interesting=2, Total=2
    Extra 'Interesting' Modifier   0  

    Total Score:   2  
  • (Score: 5, Informative) by Kilo110 on Thursday March 06 2014, @01:14PM

    by Kilo110 (2853) Subscriber Badge on Thursday March 06 2014, @01:14PM (#11900)

    USB Keyboard work by polling. PS/2 Keyboards use hardware level interrupts.

    • (Score: 4, Informative) by gallondr00nk on Thursday March 06 2014, @01:45PM

      by gallondr00nk (392) on Thursday March 06 2014, @01:45PM (#11917)

      The answer then would be to find a good quality used PS/2 keyboard, perhaps a Cherry or Model M. I assume that a PS/2 mouse would be faster as well?

      It might be something as simple as buying a couple of USB -> PS/2 adapters?

      I'm not sure if it's still true, but a lot of LCD monitors displays used to add a good 50-100ms onto latency as well. It's still true of TVs. A lot of speedrun gamers using old consoles insist on using a CRT TV, because the postprocessing in an LCD TV adds lag.

      • (Score: 1) by tempest on Thursday March 06 2014, @02:21PM

        by tempest (3050) on Thursday March 06 2014, @02:21PM (#11938)

        Most "serious gaming keyboards" have PS2 connectors due to the n-key rollover issue. Personally I think "keyboard lag" is absurd, but if he's got problems with keys sticking, his whole keyboard is messed up so perhaps that's a really issue for him.

      • (Score: 4, Insightful) by mechanicjay on Thursday March 06 2014, @03:30PM

        A USB -> PS/2 adapter doesn't change the fact that you're still polling for keyboard input. Better USB -> PS/2 adapters don't have N-key rollover issues. I guess there is a enough on-chip logic to handle it. I wonder if that makes keyboard response faster or slower though.
        --
        My VMS box beat up your Windows box.
      • (Score: 0) by Anonymous Coward on Thursday March 06 2014, @09:57PM

        by Anonymous Coward on Thursday March 06 2014, @09:57PM (#12240)

        Posting AC to avoid undoing moderation...

        This is key right here. PS/2 is interrupt driven, and USB is not only polled, but shared, and hub-ed, which means the polling is tree-ed out down the hubs. The more USB things plugged in the worse it gets, even of those things are simply other USB ports with a built in HUB.

        If you know your motherboard, you can make sure your keyboard/mouse is not shearing a header with anything else (external disk drives, thumb drives, wifi, sound devices, SD card readers, etc), and this will make a big difference. Most people don't realize just how widely their USB devices fan out. Polling has to ask each device if it has data to send, and even no response takes time to check.

        I always use PS2 keyboard for game play, although I've found I can get by with a wireless USB mouse if I keep it on a separate USB header. I've tried wireless and wired USB keyboards and they are quite inferior.

    • (Score: 1) by bill_mcgonigle on Thursday March 06 2014, @02:32PM

      by bill_mcgonigle (1105) on Thursday March 06 2014, @02:32PM (#11946)

      Yes, the submitter should use any old cheap PS/2 keyboard to determine if his latency issue is really in the keyboard, or if it's in the USB stack, a hub, the USB interface itself, etc.

      Anybody can stick in some buffer logic that's too big or unnecessary.

      That said, there are some serious sampling devices that are USB-based that are used for real science, etc. There must be an established method for measuring USB bus latency from the app to the PHY on the device.

      • (Score: 2, Interesting) by len_harms on Thursday March 06 2014, @03:20PM

        by len_harms (1904) on Thursday March 06 2014, @03:20PM (#11979) Journal

        Also keep in mind the latency may not even be at the keyboard level.

        I plugged my laptop into a 55 inch lcd. Massive lag of up to a 1/4 a second. HDMI introduces a decent amount of processing. The TV itself also had a bunch of filters for smoothing. Even the length of the HDMI cable introduced a bit of lag. Switched out the laptop for a newer model (faster hdmi encoder chip). Turned off each tv smoothing mode and made the cable shorter (10ft to 3ft). The lag is mostly gone but still there.

        I can play most games on it now. My wife however is unable to finish the second board on pacman because of it. Yet she can usually get to the keys level on her laptop.

  • (Score: 1) by WizardFusion on Thursday March 06 2014, @01:25PM

    by WizardFusion (498) on Thursday March 06 2014, @01:25PM (#11903) Journal

    Not sure why parent was modded down, maybe because you were posting as AC. It does seem like a reasonable question, one I would like an answer too as well.