Medium Format Forum

Register a free account now!

If you are registered, you get access to the members only section, can participate in the buy & sell second hand forum and last but not least you can reserve your preferred username before someone else takes it.

How I Make Scans

Deon

Member
I make my scans on the cheap. I totally understand that there are lots of much better scanners out there, but I simply can’t afford any of them. I am also aware of film copy systems that use a DSLR to copy slides and negatives. I need higher quality and larger file sizes than those tools can create. Don’t get me wrong, they are simply amazing tools, especially once you see how quickly you can create a pretty good quality digital copy.

First off, calibrate your monitor, otherwise, why bother! I’m a Mac user but, that really doesn’t matter, you simply need to be comfortable/competent with what ever you have. Next I’m using an Epson V700 PHOTO flat bed scanner. These scanners DO NOT focus on the film plane very well when you select scan from holder. So, I’m using a BetterScan scan bed, that allows me to focus the bed to the scanner (takes a while to set up, once done, Look how sharp the grain is!). The software that comes with the V700/V750 is pretty crude, kinda like trying to adjust your camera with a hammer. I’m using Silverfast Ai9 Studio software. I always make my scans as big as the scanner will let me WITHOUT using the interpolation software in the scanner (if you need it bigger, the software in photoshop is much better). I also make the scanner do multiple passes across the film using ProPhoto or adobe1998 color space (sRGB stands for small or stupid and should only be used for web). A 2 1/4 transparency can take more than 20 minutes to scan. The Silverfast software has so many ways to adjust my scan, taking the scan closer to what I want/envision before I open it in photoshop. Also, the software has so many profiles for so many brands and kinds of negative films (B+W & Color), making for a better scan to start with, sometimes using the wrong profile for better results.

A long time ago, I had a job making color separations using a LogE 40x40 copy camera. This learning opportunity helped me understand how separating the colors into CMYK for offset printing or RGB for digital printing is essential in making a good color image no matter how you are going to use it. After I open my scan in photoshop (CC 2024), I straiten, crop and size it and go over the entire image at 100 percent or better looking for dust spots or anomalies to clone out.

Once that is done, I create a new layer for levels and adjust if needed. I usually don’t need to make an adjustment or much at all, if I made my scan right.

Next, I create a layer for curves. This is where I open each color adjusting the red (cyan), green (magenta) and blue (yellow) independently. Usually I open the curves overall just to see while making my adjustments (depending on exposure of course). First I start with the red curve grab at the center and close down the red curve slowly watching how that affects the other colors in the image. Repeat for the Green and Blue. After one round I usually go through the process a time or two with smaller/micro adjustments. Once you do this a couple of times you will begin to see the effect on the overall color cast of the image. Also, you will find different color palates in the images you’re scanning (ie: landscape, portrait, studio/also color negative or transparency) that you will become used to. With experience I have found on occasions I will anchor a colors curve at a specific spot and adjust the curve with two or more points to accomplish my overall color goal.

Third, I create a Hue/Saturation layer. I leave the overall saturation alone for the moment. Starting at the top and go through each color independently, slightly over saturating that color (so you can see what you are doing) and adjust the hue to taste then saturation for that color, repeat to the bottom of the list (skipping any color that doesn’t exist in the image). Then I come back to the overall color saturation and adjust to taste. This step starts to find the more subtle colors in the scan.

The last layer I add is Selective Color. With this layer I can find specific colors that I feel needs to be adjusted. At this point I pick only the colors I want to change and open the color and finding all the colors you can then add or subtract from that color to tune that color more precisely. This really helps with subtle colors.

Lastly I go over the image to scrutinize the color further making smaller and smaller adjustments here and there using these layers. I save the full size image with all layers as a photoshop document (average size 2Gb) Catalog, archive and back up three times.

This is how I make scans. That doesn’t mean it’s the right way, just the way I do it. Something I’ve been aware of for years is that in Photoshop there are so many ways of making things happen, there is no right or wrong way to do it, it’s just different. So, if you have short cuts, ideas or improvements on this idea, I’m all ears!

Hasselblad 500C/M w/80mm f=2.8 C lens (my first Hasselblad). Kodak Ektachrome 64 film (vintage 1984 image, newly scanned)
MeToo.jpg
 
Just out of curiosity, what is the scan size you typically end up with using this process (pixel size)? I’ve talked with many that like the Epson V700 although you’ve clearly modified it.

Your editing process is interesting and very similar to what I do in Capture One, but I end up with smaller sizes because of the non-destructive edit. But, I use layers for each of the major adjustments and start with opacity of 50% so I can easily go back later and increase an edit (whereas starting at 100% means I can only decrease an edit).

Whilst I would love the results from a good drum scanner, that is a little above space / cost for my work area. I used a Nikon Coolscan for about 20 years for 35mm black and white ending up with really good results, but have struggled with 120. My current workflow has been to use a dupe stand with a 99CRI backlight and a 63mp Leica SL3 holding an older Leica R APO 100 f/2.8 Macro lens with the Elpro attached. While older that lens has 40mm line pairs that wave between 75 and 80% contrast with zero distortion, so I end up focusing on the grain.

My only issue in duping this way is that the resulting file is a negative and scans end up as positives, which makes the editing more intuitive (instead of thinking backwards).
 
Just out of curiosity, what is the scan size you typically end up with using this process (pixel size)? I’ve talked with many that like the Epson V700 although you’ve clearly modified it.

Your editing process is interesting and very similar to what I do in Capture One, but I end up with smaller sizes because of the non-destructive edit. But, I use layers for each of the major adjustments and start with opacity of 50% so I can easily go back later and increase an edit (whereas starting at 100% means I can only decrease an edit).

Whilst I would love the results from a good drum scanner, that is a little above space / cost for my work area. I used a Nikon Coolscan for about 20 years for 35mm black and white ending up with really good results, but have struggled with 120. My current workflow has been to use a dupe stand with a 99CRI backlight and a 63mp Leica SL3 holding an older Leica R APO 100 f/2.8 Macro lens with the Elpro attached. While older that lens has 40mm line pairs that wave between 75 and 80% contrast with zero distortion, so I end up focusing on the grain.

My only issue in duping this way is that the resulting file is a negative and scans end up as positives, which makes the editing more intuitive (instead of thinking backwards).
I scan just outside the rebate edge of the film (usually 14400 pix), so when I crop as close to the inside of the frame as possible I end up with 12000 pixels across (40 inch @300 dpi). Images created with the 350mm f=5.6 CF lens I only scan to 9000 (30 inch @300dpi) pixels, because 350mm Hasselblad lenses are not that sharp and have some nasty chromatic aberrations that show up big time if I scan too big.

Working in layers (as long as you save the layers) makes everything nondestructive, you can always revert to the original.

I crop in camera when shooting, never in the darkroom or scanner. I think it makes life easier to get the image as close as possible to your vision in camera, long before the darkroom or photoshop.

Hasselblad 500C/M w/ 100mm f=3.5 C T* lens (Panatomic-X film). This is an old image of my wife Trish with her SWC/M (gift from Hasselblad). Hiking the Washington Coast at Shishi Beach.
053184c#10.jpg
 
I’m actually going to the coast in the early am with the 100 f/3.5 - one of my favorite lenses in film and digital. Likely shoot FPan at ISO 25. This is a great shot! Thanks for the details on the scanning…
 
Right after I wrote about my scanner, it died! I'm not all that upset since I figure it has easily made over 35,000 scans, not to mention the platen glass was getting pretty scratched up. I found a new in box old v700 on FeeBay, so I don't need to update Silverfast software (scanner specific).

Something I neglected to tell you about regarding scanning. CALIBRATE your scanner! I forgot all about this very important aspect of a quality scan, until I was setting up the my new scanner... I use an IT8 calibration target with Silverfast. Buy the target that is made by the manufacture of the film you are using the most. I have three IT8 targets for the films I use (2 color transparency and 1 color print). Calibrating your scanner will make subtle colors come out and show off how wonderful they are, instead of boldly seeing only RGB colors.

I'm back home after a nearly month long road trip to central Oregon delivering prints for a two person show I'm in through October 31st 2024. (Art Adventure Gallery - Madras, Oregon)

Scappoose, Oregon - Hasselblad Flexbody w 100mm CFi f=3.5 lens - Fuji Velvia film
OR-2128.jpg
 
Back
Top