[ATM] zone one profiles

["Koehler, Steve" <Steve_Koehler@securecomputing.com>]

Here's another attempt to make sure everyone sees the odd brightness profile I
see in my zone one example.  I have posted two more graphs, one for zone 1 and
one for zone 5.  These graphs differ from the one I previously posted, because I
did not flip the right side profile over the left side--the profiles are as you
see them in the test.

    http://www.visi.com/~mkoehler/zoneone/zone1_lines2.png
    http://www.visi.com/~mkoehler/zoneone/zone5_lines2.png

Zone 5 is pretty well behaved.  The profiles all slant down from left to right,
and stay roughly parallel as they change in brightness from top to bottom.  For
this zone, J is clearly the best null.

Zone 1 is different.  The lines are roughly parallel for the first six frames,
but then the red (right) profile starts to blip up in the center.  By the time
the lines meet in I and J, the blip up is already significant.  It continues to
go up more and more by the last frame.

Here you can see that the best null, based on total brightness, is between I and
J, somewhat closer to J.  This matches the brightness ratio annotations to the
right of each frame.

Here are some of the things I think may contribute to reading difficulties in
the presence of this asymmetry:

1. The spacing between the lines at the center of the zone increases faster as
you move out from null than in from null.  Compare frames D and O to see this.
This may throw off a bracketing technique.

2. If you compare only the brightness of the zone center, you would
find the null between images H and I (too far in).  This is the mistake I
thought most people would make.

3. If you compare only the brightness at the zone edge, you would find the
null around K or L (too far out).

-- Steve Koehler
   steve_koehler@securecomputing.com


_______________________________________________
ATM mailing list http://www.atmlist.net/