That truck again,...again!

[P3D Dylan The Hippy Wabbit <spacey@xxxxxxxxxxxxx>]

Jonathan Gross (oops, sorry about the spelling last time) said of JPEG
compression for stereo pairs:-

> The problem is that with stereoscopic images pairs, the subtle high
>frequency variations in color between the left and right fields may be
>all that defines horizontal boundary differences (parallax) for a
>particular feature.  As has been noted many times before on this list,
>the visual system is amazingly sensitive to minute differences in images
>presented to the left and right eyes. This discrimination is so acute in
>the horizontal direction that you can throw away every other (vertical)
>raster of an image pair, as long as you maintain the resolution in the
>horizontal axis.

Larry Berlin argued otherwise in more detail than I could, at least as it
applies to stereo.

However, in the three days since I posted it neither of these good people
(or anyone else for that matter) have responded to my challenge to provide
evidence.  Come on guys!  We can state and restate opinions till we all
wanna hurl at the mention of the name, but only facts will settle the
point.

So how about it?  I'll organise the pictures and crunch the numbers.  Any
one interested can email me for the files, sort them in order of quality,
and tell me the order they arrived at.

I expect that anyone can tell the high compression files from the low
compression ones, but at the top of the scale there should be little or no
relationship.  Just how far up the scale though?

Anyway, if no one expresses an interest this time I'll shut up, OK?

Dave Spacey

______________________________________________________________________________

When you break rules break 'em good and hard.   - Terry Pratchett


------------------------------