Quality of digitized slides

Submitted by Klaus on Thu, 11/12/2015 - 17:02

This topic was moved from http://gwulo.com/atom/22944 to the forum section. I think digitizing of slides is of general interest.

I'm happy that my most of 1980's images are so clear and sharp. There might be two reasons for that: First one is that Kodachrome 25 colour slide films were really good - only 25 ASA, but with very high resolution. This in conjunction with the 50mm lens of my Asahi Pentax ES2 - unfortunately broken long time ago.

The second reason is the way to digitize the slides. I tried it myself with a usual desktop digitizer. The result was reasonable but not really good. Main reason for that might be the paper frames that Kodak used at that time. When these slides are in the light beam of the projector, after a couple of seconds,  they heat up and "move foreward with a plop". The image on the screen is out of focus then, this normally is correct by the autofocus device. When you digitize the slide the same thing happens, and the result is a double image and an unfocussed image in your file. Via internet I found a small company that uses a high resolution camera with a (what they call) full-size sensor. They take photos of the silde with three different exposures and calculate the optimum result. To give an idea I show the photographed (above) and the (self-)digitized slide (below):

On a first glance there seems to be no big difference. However, when you go into details it's easy to see:

Left hand side is the photographed, right hand side the (self-)digitized one. On both you can see a clock, but reading the time is only possible on the left one.

The price is reasonable, it's more expensive than direct digitizing, but not much.  

Regards, Klaus

Hi Klaus

I found that the best way to digitise a thousand or so of my old slides (usually ASA 25, but even ASA 10) was to project them onto a white wall in a darkened room and then photograph them with my small Canon digital camera mounted on a tripod.  This produced much better results than I'd achieved previously using scanners.  

Best wishes Andrew