Film from Digital (Part 5 )

In this update I attempt to print a couple of the images from the composite negative I made with Gammatech. I will discuss the pitfalls and potential issues with this approach. 

Eight 65:24 Images on one sheet of 4x5 film

In the original post I made a composite 4x5 negative from 8 different 65:24 aspect ratio digital images from my Fuji GFXs ii camera. Each of these has the Fuji Acros film simulation applied as well as a 'strong small' grain simulation. Each image is virtually full resolution, being scaled from the native 8256 pixels width on the camera to 8192 pixels to fit two across on the film at the Gammatech full resolution (16,384 pixels). Each image is 58mm wide so could fit (just) in a 6x6 enlarger.

Challenges with Multiple Images on Same Negative

This many images on one negative while very efficient in terms of cost does present some challenges in the enlarger. There is not room in the 4x5 negative holder to center the desired image. This means I am printing off center from on the enlarger and probably outside the sweet spot in terms of sharpness and vignetting. This leads to the temptation to cut the images apart.

The other consequence is that I of course needed my 80mm lens to enlarge to the same size as the other versions at 19"x7" image size. This is about 11x enlargement (compared to 5x in the prior prints). The result, as you would expect, is the image is slightly softer due to the greater magnification. 

This negative seems to me to have less contrast than the one from Bayeaux and took some work to get the print into decent shape. I ended up with all #5 (hard) filter and even then, I opted to halt development at 2 minutes short of the usual 3 minutes. 

I use Moersch ECO-4812 developer which at 1:10 dilution offers the option of stopping the development at 2 minutes compared to the nominal 3 minutes to enhance the contrast. This helped get the contrast for this image into the right range. ECO-4812 is indeed a great developer and this option is one of its many benefits. 

You can see this when developing a print as the highlight grays come up in only final minute of 3 minutes of development.  Below is a test strip I used to determine how much this would help. Note the highlights on the right are brighter though the shadows are just as dark as the longer development time version. 

Test strip for development time #5 32 sec
left: 3 minutes in developer, right: 2 minutes in developer.

scan045_stitch fx #5 45 sec 2 min develop

I decided to take my own advice and carefully cut the 4x5 negative in half along the long axis. This made it quite easy to position in the 6x6 negative holder. The ends were slightly cropped as the negative is slightly wider than the negative holder aperture. I then made a 9 1/2" x 3 1/2" print on 8x10 RC paper I cut in half. This came out very nicely and was pretty easy to print as I only needed to burn the grass in the foreground 1 stop. 

scan048 f16 #5 26 sec #00 8 sec burn grass #00 8 sec

I think this session gave me the most to learn about film negatives from digital images. Unless I want a 4x5 image to start with I am probably better off using the medium format options from Gammatech. I can fit two 6x6 images on the 4x5 film, but the cost is the same but with the added effort of separating the two images. For these wide aspect ratio images, I can fit two on a 6x7 or 6x6 negative and get full camera resolution.  Full frame images I think would best fit on a 6x6 negative. 6x6 gives me the ability to print on any enlarger I have. The more film you use the less grain becomes an issue. Thus, it is tempting to use 4x5 film but at twice ($20 vs $10) the cost. 

For this project, this sheet of film cost $20 for 8 images. On 6x6/6x7 with two images per negative it would have cost $40 for all 8 images. With two per sheet of 4x5 then $80 for almost double the film area of the 6x7 option. 

Again, in this evaluation I did not see any evidence of pixels. Each image was 8192 pixels long then printed to 19 inches. This gives 431 dpi so not surprising they are not visible. I also suspect the pixels are not like the ones we are used to seeing on the screen from a digital image in that they are not small sharp squares or rectangles but rather more amorphous which would hide nicely in an analog process.

The other area to explore is the manipulation of the image to make it easier to print.