Realist format slide scanner?

I currently use a Canoscan 9000F. I do not remove the slides from the mounts, but lay the cleaned, mounted pair right on the platen and scan both chips at once. I scan at 4800 dpi setting, although the Canoscan is purported to scan at 9600 dpi (optical).

There is a workflow required to produce good results, and StereoPhotoMaker is part of it...

See the results here (at lower resolution, and in crosseye format...

https://www.flickr.com/photos/turbguy/sets/72157622321846829/#

All of these shots were taken with either an f3.5 or f2.8 Stereo Realist Camera made by David White Co.

In the 1970's, my neighbor's eyesight began to fail, so she sold me her f3.5 Stereo Realist so it could continue to see use. She used it VERY heavily for at least 20 years. I used it just as much for over 20 MORE years, until I upgraded to an f2.8 Stereo Realist. It still worked well when I sold it! The Realist is a very sturdy, simple camera. In a pinch, it also can be used as a weapon!

I couldn't afford slide film in those early years, so I fed it re-spooled B&W film, processed it myself, and made Holmes views in my "darkroom", seen here...

<a href="http://www.flickr.com/photos/turbguy/4634280265/">www.flickr.com/photos/turbguy/4634280265/</a>

To produce a single antique Holmes (3.5&quot; x 7&quot;) card took about an hour (not including drying time). How things have changed!

I had produced HUNDREDS of negatives that I've never seen printed or made positive. A modern film scanner is changing that rapidly....

Slides and Negatives presented here were scanned with a CanoScan 8400F and 9000F scanner, cropped and re-aligned with StereoPhotoMaker, then tweaked (levels, curves and color balanced) and sharpened with Photoshop. Recently, I have discovered that even very dense negatives can be successfully scanned by fusing three scans at different exposures with Photomatix software.

The scanner can recognize dust and scratches on E6 films and &quot;removes&quot; them. Sometimes this produces small spot &quot;retinal rivalries&quot; as the software &quot;fills-in&quot; or clones the obscured image with extrapolated data from surrounding areas. Spots/scratches on black and white negatives cannot be detected and these remain in the displayed image.

The Stereo Realist can have some &quot;Quirks&quot;. Some users have called these quirks &quot;features&quot;! It was common to experience a small light leak on the right film chip using modern high speed (100 ISO) films due to an inadequated film door seal. These remain in some of these scans. Another is a slight difference in exposure between pairs. While the shutter is fully sychronized, the irises are not so precise. The scanner will pick up these slight exposure differences and some of this shows as &quot;retinal rivalries&quot; in the highlights (typically in clouds). The solution was to always set the lens aperture from one direction to compensate for very slight &quot;slack&quot; in the iris adjustment mechanism. An example of both &quot;features&quot; is here:

<a href="http://www.flickr.com/photos/turbguy/3949565494/in/set-72157622321846829/">www.flickr.com/photos/turbguy/3949565494/in/set-721576223...</a>

The f3.5 Realist lenses will vignette heavily when used at f16 or f22. Here's an example:

<a href="http://www.flickr.com/photos/turbguy/4468636702/in/set-72157622321846829/">www.flickr.com/photos/turbguy/4468636702/in/set-721576223...</a>

It's best to go no smaller than f11 to avoid vignetting. Some of this vignetting can be seen in the scans of early slides and negatives. The f2.8 lenses do NOT vignette at small apertures, and will produce superior photographic results overall.

For a great resource about the Realist, go to DrT's page here:

<a href="http://www.3drealist.com/" rel="nofollow">www.3drealist.com/</a>

All the images here are arranged for &quot;crosseye&quot; viewing.

You can use :

<a href="http://www.schillr.com/index.php5?n=1&amp;outputmode=parallel&amp;query=&amp;user=turbguy&amp;color=&amp;sort=" rel="nofollow">www.schillr.com/index.php5?n=1&amp;outputmode=parallel&amp;am...</a>

To see them in Parallel mode!