Digitizing film with a digital camera

[philipus]

Advertisement (gone after registration)

Apologies for quoting myself but just to explain what I meant. In this old post is the test I made using my Coolscan V ED and my Canon 5D2 with the 100L macro. In this test the camera pulled more from the negative than the scanner. That can be a good or a bad thing. Typically for my workflow I aim to get as lifeless, flat and boring a scan as I am able to in order to preserve flexibility for the post-processing (this is why I wrote somewhat sloppily below that the camera was 'close' to the V ED). In my experience a sharper file may not withstand post-processing so well as I prefer to sharpen in at least two later stages, sharpening before finalising an image for computer/internet viewing and final sharpening before printing. But for a straight out of the camera file the digitized result was, I think, very good.

There is a great convenience in having digital ICE even though it does soften images a bit. I am no longer using the V ED or my Coolscan 9000 but that function saved me lots of time. Since a while back I have switched to a Flextight X1, which doesn't have digital ICE. Flexcolor's Flextouch function doesn't work well enough for me but with proper cleaning of negs (Ilford's Antistaticum is a must) dust-spotting is usually not very time-consuming for a 2000-3000 dpi file. An X1 scan at full resolution will however reveal literally every bloody little thing on the neg which is in equal measure extremely impressive and extremely frustrating so I only scan at full resolution when absolutely necessary.

I'm sure that in the future I will switch to digitising using a suitable camera rig. My main concern at present is that it would require stitching anything larger than 35mm film and I am not very keen on that. So for now I stick with my scanners even though I have the 5D2 and the macro lens.

I think it is a very positive sign that just in the last year or two the number of sites and forum threads for instance here and at RFF showing various ways to digitize has increased very much. So there are a lot of people out there who try this out and come up with very creative solutions, both in terms of equipment, rigs and ways to post-process (including "getting rid" of the orange mask on C41). On this note just a friendly piece of advice to the OP. Since there are quite a lot of us in this forum in particular who are testing and trying out things just like you don't discount the forum as such only because views contrary to your own are expressed. 

br

Philip

 

On 4/30/2019 at 12:39 PM, philipus said:

I love using my film scanners but I also like digitizing and have tested it with 135 film obtaining results which are close to my Coolscan V ED. I'm under no illusion that at some point old scanners will not be the basis for a viable workflow so anything that tests new ways to stick with film is good imho.

Just a few suggestions, if I may - to better show quality differences you might consider using less contrasty images, as well as adding a few colour images. It is a bit difficult as the blog post stands at present for the reader to form an opinion.

And, why are the Epson scans much more dusty than the digitized frames? That also detracts a bit from the comparison.

Br

Philip

 

[dau]

I think it's nearly always worthwhile to reflect on the tools and techniques being used. Even a comparison between the good old scanner and the even more ancient way of photographing your negatives or slides might bring surprising and encouraging results.

In doing so, one should be aware of the possibilities of the tools involved. For instance, the true optical resolution of a Leica SL is close to 2000 dpi, while the scanners mentioned in the article boast optical resolutions of 6900 and 8200 dpi, respectively.

However, the material being used for demonstration in the article points to other areas which might yield much greater improvements in work flow, i.e. by better controlling the exposure and the development leading to acceptably dense negatives with less curl. Indipendently of the imaging method chosen, dirt and dust management seem to be issues well worth exploring.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Jossie]

Quote

There is a great convenience in having digital ICE even though it does soften images a bit.

This may be true for the ICE of Nikon, which I know only from the tests published. It is not true for SilverFast's iSRD, if set up correctly, which affects only the area damaged by dust/scratch. Of course, in these parts all information is lost. Parts of the image without dust / scratches are not touched.

Hermann-Josef

[philipus]

Thanks Hermann-Josef, I didn't know that's how Silverfast works. Just so I understand what do you mean by "all information is lost"? Perhaps you could show an example? 

Br

Philip

13 hours ago, Jossie said:

This may be true for the ICE of Nikon, which I know only from the tests published. It is not true for SilverFast's iSRD, if set up correctly, which affects only the area damaged by dust/scratch. Of course, in these parts all information is lost. Parts of the image without dust / scratches are not touched.

Hermann-Josef

[Jossie]

Hi Philip,

well, this is more or less trivial. What I meant was that the information in pixels, that are affected by dust / scratch, is lost. It cannot be recovered by ICE or whatever it is called. An example of iSRD working on a slide is given above. I carefully checked what iSRD does by subtracting a cleaned image from the original and iSRD only touches the pixels affected by dust, as I had said. So there is not loss in resolution in the untouched part. However, the parameters for iSRD have to be carefully chosen, especially for Kodachrome slides.

Hermann-Josef