>The EyeBall works on direct viewing of the LEDs, and I think this method
>makes sense.  I don't see what you gain with a diffuser.  It would fuzz the
>pixels and decrease brightness.
>
A piece of paper looks really nice, it indeed decreases brightness, but 
the large advantage of the diffuser is that you can see the star at any 
place on the sphere and I think that's different then the goal of the 
EyeBall.

>  Also the cool thing people seem to like
>about the EyeBall is that the image appears to hang in space.  You don't see
>the rotating arm, just the pixels.
>  
>
>>BTW, how many LEDs did you have to control in the EyeBall ?
>>    
>>
>
>The first EyeBall had 16 LEDs taking up about 3 1/2 inches on a 12 inch
>diameter sphere.  It was designed to display text with a single built in
>font.  The second EyeBall is 18 inches in diameter and has 96 RGB LEDs
>creating a vertical display height of about a foot.  That's 188 total LEDs
>that can each be modulated to 2 bits of intensity per pixel.  That is
>sufficient to display recognizable individual faces, although that's mostly
>a bonus.  They originally cared most about arbitrary downloadable fonts and
>simple low-color logos.  They didn't complain though when we added an extra
>bit per color per pixel.
>
>The total display is 512 x 96 pixels.  There is a dsPIC handling the data
>from the rendering processor (which is another dsPIC) and managing a CPLD
>that clocks the data to the LEDs.  If you work it out, 512 x 96 pixels x
>30Hz x 8 bits/pixel comes out to a significant data rate.
>
>
>  
>
Wow, that's a whole lot,
I didn't realize that it was so much bandwidth.
(And I wanted it to realize with just 1 16F877, still hoping  :-(

cheers,
Stef

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist