W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > April to June 2002

Flickering

From: Lee Roberts <uce@roserockdesign.com>
Date: Mon, 15 Apr 2002 12:39:09 -0500
To: "WCAG List" <w3c-wai-gl@w3.org>
Message-ID: <NFBBJHFEOLAGEICMIMBPMEDACEAA.uce@roserockdesign.com>
Here's the information for flickering and as I
mentioned on the telecon this information is from
my training in the US Navy.

Let me start of by covering some information about
the eye.  The back half of the eye is covered by
rods and cones that are sensitive to light and
color changes.  The area directly behind the pupil
is used most of the time when we focus upon
something.  However, the rods and cones used in
the field of vision include more than those
directly behind the pupil.  It is these rods and
cones that help people see things they may not be
more accustomed to seeing.

One of the things the military does is teach
people how to look a few degrees away from an
object at night and thereby using the rods and
cones that are less accustomed to the quick
changes of light.  This lack of being accustomed
to the changes in light makes objects more visible
at night and in the dark.

To get a basic idea of how these rods and cones
come into play you can use the standard field of
vision test any doctor does to check your vision.
It is not the rods and cones directly behind the
pupil that enables one to see a wide field, but
rather the rods and cones closer to the pupil.
Again, because less light actually gets to these
rods and cones they are less accustomed to quick
changes.

I'm sure many people have seen tests like this
(http://www.exploratorium.edu/exhibits/bird_in_a_c
age/bird_in_a_cage.html).  They do an excellent
job explaining why we see things that we don't,
but let me go a step farther.  When we look at a
computer screen the rods and cones located at the
back of the eye become accustomed to the light
generated by the computer screen.  However, when
we look away, approximately just to the edge of
the monitor casing, we may see something slightly
different.  If the refresh rate is not set
correctly one may begin to see some screen
flicker.

Another test one can use is to change your screen
resolution from say 640x480 or 800x600 to 1152+.
Since your refresh rate was optimized for your
lower screen resolution, the refresh rate will not
typically be optimized for the higher screen
resolutions.  This will cause flicker and will
probably give a normal person a mild headache
within a couple of minutes.

Now, when we apply this to an image on a web page
the results may vary based upon the optimization
of the computer refresh rates used.  I would
recommend importing the image in an animation
software product like Animation Shop from JASC.
The number of frames per second is available and
the flicker rate can be calculated.  If the rate
is any where between 2 and 59 flashs per second
then the image should not be used.  Macromedia
Flash and QuickTime can produce flickering that
come up in this range as well.  If you don't want
to or can't import the image you can use a simple
calculation of flickers per second by simply
adding the number of flickers in a one second
window.  If you need a longer sampling time
remember you need to calculate the number of
seconds used before attempting to determine the
flicker rate.  For example, if I were to use a 3
second time frame and had 139 flickers in that
time, I would have about 46 flickers per second
which would put the flicker rate in the "not
allowed window".

Sincerely,
Lee Roberts
Rose Rock Design, Inc.
Building web sites accessible by EVERYONE
http://www.roserockdesign.com
Received on Monday, 15 April 2002 13:37:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:19 GMT