PTB-3: flickering black & white opposite images not result solid gray

1
I am attaching a short program called not_solid_gray.m which flickers,
frame for frame, a small black & white checkerboard with its black &
white inverse image.

Theoretically, the resulting presentation should be a solid gray
square. But in reality, using MATLAB 2007a and PTB-3 on an
Intel-based Mac min and also a Windows XP PC, artifacts clearly define
the borders between the black and white squares within the
checkerboard during flickering. (Though the bulk of the area is the
expected average gray.)

Just wondering if anyone can explain why the border artifact appear?

I did notice that the coloring of those border artifacts can be
changed depending on which video buffers (i.e., buffer 1 or buffer 2)
the flickered textures are drawn. These comments are repeated within
the code itself and possibly more clear in context.

Thanks for any insight.

Daniel Shima
Vanderbilt Vision Research Center
2
Daniel,

I believe that at least some of the problem here is due to two fairly well
known issues related to display hardware. The first is that the frequency
response of the gun circuitry is not high enough to follow the square wave
at the white-dark boundary. The second is that within patches of maximum
intensity the power supply can not drive the guns at a constant level across
the full patch.

My memory is that Dennis Pelli has written an article/not about this. These
are also issues that George Sperling has thought a lot about.

Ted Wright

On Tue, 10 Jul 2007, Daniel Shima wrote:

> I am attaching a short program called not_solid_gray.m which flickers,
> frame for frame, a small black & white checkerboard with its black &
> white inverse image.
>
> Theoretically, the resulting presentation should be a solid gray
> square. But in reality, using MATLAB 2007a and PTB-3 on an
> Intel-based Mac min and also a Windows XP PC, artifacts clearly define
> the borders between the black and white squares within the
> checkerboard during flickering. (Though the bulk of the area is the
> expected average gray.)
>
> Just wondering if anyone can explain why the border artifact appear?
>
> I did notice that the coloring of those border artifacts can be
> changed depending on which video buffers (i.e., buffer 1 or buffer 2)
> the flickered textures are drawn. These comments are repeated within
> the code itself and possibly more clear in context.
>
> Thanks for any insight.
>
> Daniel Shima
> Vanderbilt Vision Research Center
>
>

--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.