Re: Human tIming perception (was: RT patch)

From: Lee Revell (rlrevell_at_joe-job.com)
Date: 05/31/05

  • Next message: Russell King: "Re: Will __pa(vmalloc()) ever work?"
    To: Steve Finney <saf76@earthlink.net>
    Date:	Tue, 31 May 2005 16:03:55 -0400
    
    

    On Tue, 2005-05-31 at 12:59 -0400, Steve Finney wrote:
    > It takes (IIRC) about a 10 ms or
    > so difference in the sounded sequence
    > for someone to be able to report that there's been a change, but
    > a cnange in the timing of the person's finger movements occurs
    > (_immediately_) at perturbations smaller than 10 ms. That is, there
    > appears to be some dissociation between conscious perception and
    > perceptual/motor behavior.

    Any decent guitar player who has used their computer as an effects unit
    could tell you this. I can easily perceive the difference between 1.3
    and 2.6, and 2.6 and 5ms latencies. And there's at least one person
    (also a guitarist, who I have added to the cc:) who swears he cam
    perceive the difference between 0.6 and 1.3ms. Soundcard ADCs typically
    add 1.5ms latency in each direction, so the actual floor seems to be
    around 3-5ms.

    Lee

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/


  • Next message: Russell King: "Re: Will __pa(vmalloc()) ever work?"

    Relevant Pages

    • Re: [patch] voluntary-preempt-2.6.9-rc1-bk4-Q5
      ... considering my relatively underpowered system. ... fast UP system you would not see any latencies worse than 100 usecs. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [RCU problem] was VFS: file-max limit 50044 reached
      ... > cpu is able to queue more than 10.000.000 items per second in a list. ... and probably OOM avoidance has a higher priority than latencies in DOS ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [patch] voluntary-preempt-2.6.8.1-P0
      ... logscale y in gnuplot). ... If these latencies have a relationship to the ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: RT patch acceptance
      ... if you drop just a few unimportant IRQs ... > latencies, so they aren't going to notice any changes in ... Copyright 2005 by Maurice Eugene Heskett, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)