A couple weeks ago I got myself a Yashica X3.0 half frame 35mm camera. A half frame camera creates portrait oriented negatives which are ~18mm wide and ~24mm high. Turned sideways, this is very close in size to the APS sized sensors in many DSLR cameras. Read the following Wikipedia article for a deeper dive into the half frame format :
Half-frame camera - Wikipedia, the free encyclopedia
Without adjusting my macro setup from its original configuration for scanning traditional 35mm landscape oriented film, the half frame negative projection does not fill up the entire sensor. Here is what the negative looked like at 1:1.5. I rotated the image in Aperture to make Wonder Woman stand straight. Otherwise, she would be sideways.
Cropping the image down to only include the half frame negative results in an ~8MP resolution image. In the example above I am cropping down a bit more because the Yashica couldn't focus closer. I removed some of the spacers between my lens and my negative holder to get closer to 1:1.
Now I am closer to a full 16MP shot. Again, I'm cropping down for compositional purposes.
I wanted to compare the 1:1.5 and almost 1:1 shots to see if there is an increased perception of the grain. A larger grainier picture does not have much worth unless you are interested in the grain.
The color processing between the shots is a bit different despite lifting all the adjustments from one image and stamping them in the next. I think bringing the negative in closer changed the color and the intensity of the projected image onto the sensor. The 1:1 shot was 1/25th and the 1:1.5 shot was 1/30th. I'm also wondering now if the focusing and the lens performance was different at the two macro distances. Otherwise, in terms of grain, I think the 1:1 shot does show more of the texture of the film but it doesn't seem distracting. I can't see any difference when viewing the images scaled down to fit on my 21" monitor. My gut feeling says that a 4"x6" print won't show any difference either since that kind of detail gets averaged out during the downscaling process.
I have more to investigate but I wanted to post my initial results.