Reaction time, was Re: Portability versus the "Gee Whiz" factor

HaB JacKaL wrote:

That’s ridiculous. Human eye can’t recognize more than 25fps. If your
game works at 45fps on your base target,
well you can lower your base target. :slight_smile:

Perhaps the eye can’t recognize the difference, but you can definately
tell a difference in the responsiveness of the controls. Try aiming
accurately in Quake 3 at 30fps, then try again at 65fps. YOu will
notice a HUGE difference in the way everything feels at 60fps.

The US military learned from the very first flight simulators that the
time from when a control input is entered into the simulator to the time
the reaction is seen must not be more than 100 milliseconds or the
simulator actually reduce the pilots skills. This rule applies to action
games too.

In a game you are usually viewing one frame, while another frame is
being drawn, amd your input will becomes visible in the frame after
that. In other words there is a 2 to 3 frame delay between the time you
move a mouse and the time you see the result. At 30 fps (33.33…
milliseconds per frame) you are right at the 100 millisecond limit. At
60 fps (16.66… milliseconds per frame) you see the effect of your
input with a 50 millisecond delay. (This all assumes double buffering, a
triple buffered system adds one more frame of delay.)

Thanks for point this out, I had forgotten about this effect. This
implies that for action games 30 fps is the MINIMUM you should aim for
with 60+ being the goal.

	Bob Pendleton> 

-HaB


Join the world’s largest e-mail service with MSN Hotmail.
http://www.hotmail.com


SDL mailing list
SDL at libsdl.org
http://www.libsdl.org/mailman/listinfo/sdl


±-----------------------------------------+

  • Bob Pendleton, an experienced C/C++/Java +
  • UNIX/Linux programmer, researcher, and +
  • system architect, is seeking full time, +
  • consulting, or contract employment. +
  • Resume: http://www.jump.net/~bobp +
  • Email: @Bob_Pendleton +
    ±-----------------------------------------+

The US military learned from the very first flight simulators that the
time from when a control input is entered into the simulator to the time
the reaction is seen must not be more than 100 milliseconds or the
simulator actually reduce the pilots skills. This rule applies to action
games too.

In a game you are usually viewing one frame, while another frame is
being drawn, amd your input will becomes visible in the frame after
that. In other words there is a 2 to 3 frame delay between the time you
move a mouse and the time you see the result. At 30 fps (33.33…
milliseconds per frame) you are right at the 100 millisecond limit. At
60 fps (16.66… milliseconds per frame) you see the effect of your
input with a 50 millisecond delay. (This all assumes double buffering, a
triple buffered system adds one more frame of delay.)

Thanks for point this out, I had forgotten about this effect. This
implies that for action games 30 fps is the MINIMUM you should aim for
with 60+ being the goal.

I’ve played many action games (Deus Ex, for example, that is quite heavy) with my
crappy Matrox Mystique G200 8Mb (it was slow even when it came out) and a P2 400.
Sincerely, even if frame rate hasn’t ever passed 25fps at 800x600x16, I didn’t
ever notice any disease, until fps dropped to less than 12 in some points.

So, my suggestion is to indicate as MINIMUM harware requests the ones necessary to
run the game at 25-30fps (even 20, if it’s a strategical one - I’m playing
Black&White in low detail at 800x600 and I’m quite -even if not totally- happy
with my piece of junk) and the RECOMMENDED ones to play it at 60fps. Also remember
to put in some scalable detail level.
You may also want to indicate some YOU-ARE-AN-IDIOT-THAT-VANISHED-10-SALARIES-TO-
BUY-HIS-COMPUTER requests to play your funny game at 1600x1200x32 at 120fps, while
PC is running in background 5 dedicated game servers, two MPEG4 realtime encoders
and it’s preparing dinner meanwhile. =;-)