[opendtv] Re: (No Date: Tue, 28 Sep 2004 10:30:37 -0400

  • From: Craig Birkmaier <craig@xxxxxxxxx>
  • To: opendtv@xxxxxxxxxxxxx
  • Date: Tue, 2 Nov 2004 09:06:24 -0400

At 12:13 PM -0500 11/1/04, Manfredi, Albert E wrote:
>Sorry, Craig. You can't weasel out of it so easily.
>The site describes the shape of the makula and the
>eye tracking movements. Both favor horizontal views.
>And heads also pan horizontally more easily than
>vertically.

This has NOTHING to do with the field of vision. Keep in mind that 
there are two parallel processes going on in Human Vision:

1. Wide area detection of stimuli - this is CRITICAL to this 
discussion. Only a very tiny area of the retina - the fovea - 
responsible for high resolution vision. It is central in location and 
has a very high density of Cones that are responsible for both color 
and high detail image perception.  Outside of the Fovea the 
distribution of Rods is generally quite uniform, providing two 
overlapping - NEARLY Hemispheric fields that are combined to create 
the total field of view - i.e. the ability to provide a stimulus that 
directs a complex matrix of actions to track the sitmulus, providing 
a highly detailed view of the stimulus. Here are two references that 
should help to clear up this argument:

http://www.egt.bme.hu/vilagit/angol/1_5a.htm

This page provides a graphic illustrating the actual field of vision, 
which is essentially a hemisphere within which the fovea can be 
directed to acquire a high resolution image.

The following reference is much more specific:

http://www.engr.panam.edu/~caharlow/EE4333/notes/ch2_humv.pdf

Field-of-view. The field-of-view is 90 degrees to each side, 60 
degrees up and 75 degrees
down. Both eyes yield 75% of a hemisphere. There is a stereo capability.

I went back through the list archived to try to find other 
references, since we have been down this path before. Back in 2000 
you argued that the vertical field of view was only 90 degrees. I 
argued that it is greater than 130 degrees. I know there was a 
specific reference that confirmed my position, but I cannot find it 
on the spur of the moment. But the link above confirms what iIstated 
four years ago.

Now the embarrasing part Bert.based on the information linked above, 
the human field of view is very close to a 4:3 aspect ratio...CLEARLY 
it is not anywhere near as wide as you contend.

That being said, there is NO meaningful relationship between the 
legacy 4:3 television format and our field of view.The reason for 
this is simple. NTSC was designed to cover only a small portion of 
our field of view - about 11 degrees horizontally. This more closely 
matches the area of our foveal vision than our total field of view. 
It also means that to watch the typical NTSC TV your eye does not 
need to move at all. Any eye-tracking that needs to take place can be 
accommodated by the rapid saccade movements of the Fovea, which are 
required to acquire a high resolution image.

The major difference with HDTV is that to stimulate the induction 
effect, the display must cover a significantly larger portion of the 
human field of view. NHK say it should be at least 30 degrees 
horizontally. Clearly even this size screen does not come close to 
covering the total field of view, but it is sufficient to stimulate 
eye tracking of motion within the image, which does not occur with 
NTSC.

2. The human visual system acquires high resolution images through 
the central foveal image receptors. This involves directed eye 
movement to track motion and a process of sampling that involves 
rapid saccade - i.e. rapid small movements that allow the fovea to 
take a series of overlapping high resolution images, which are 
processed in the Brain to create a high resolution image. This is 
further enhanced by small vibration like movements that cause the 
photons to be distributed across multiple cones.

Your original source has some good coverage of this process:

http://www.4colorvision.com/reading/foveola.htm

What all this means is that it is nearly impossible fully stimulate 
the total field of vision with a projected image, but that really 
does not matter. What is important is that the image cover a large 
enough portion of our field of view to induce directed eye movement, 
in order to acquire high resolution images.

>Your vertical peripheral vision is obscured by your
>eyebrows on top and by your cheak bones below.
>Again, unless you happen to be severely bug-eyed.

Which accounts for the reduced vertical field of view I  reported 
above.  But no where near the reduction that you believe to be the 
case.

>
>The result is that peripheral vision horizontally
>is good for just about 180 degree coverage, where
>vertically it's not close to that.

135 degrees. The slight advantage in width (180 degrees) is due 
primarily to the binocular nature of our vision, which overlaps two 
hemispheric views.

>
>Even if the fovea is circular, your peripheral
>vision, as well as head and eye moviements favor a
>wide apsect ratio vision.

Don't get confused by the facts Bert. We have two types of vision 
that overlap. The most important take away in this discussion is that 
the field of high resolution image acquisition is confined to only a 
tiny portion of our total field of view. Obviously IMAX works with a 
1:1 aspect ratio, and Cinerama works with a 2.35:1 or wider aspect 
ratio.

So why are you so adamant about a particular aspect ratio, when you 
are not even motivated to go buy a set with a big wide screen?

Regards
Craig
 
 
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at 
FreeLists.org 

- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts:

  1. Home
  2. opendtv
  3. Archive
  4. 11-2004