I have thousands of 35mm transparencies filed away and normally never looked at. Recently I had to dig them out as I was doing a 50 year family scrapbook (via Blurb) for my younger family members as a keepsake. I scanned most of the images using my Epson V750Pro and this provided adequate results for this book. Whilst going through the slides I found a few that I wanted to print to a larger size and arranged for them to be professionally drum scanned by Metro Imaging in London. This was done a few years ago and it was mostly 40 year old transparencies.
I recently bought a Nikon ES-2 slide copying kit and I was curious to see how this would compare with the V750 and the drum scans. I was actually surprised just how well they stood up to the professionally done shots and how much better than on the V750Pro.
I was using a D850 plus 60mm AFS micro. Lighting was just daylight through a window although I did try using a remote speedlight but found it to harsh. It is not the greatest shot in the world and the framing is slightly different on each shot due to the constraints of the mounts.
Considering that the drum scans cost about £25 each and the ES-2 retails for around £130 in the Uk I think that this is a viable and easy to use method. I have attached images of all three methods for anyone interested in using this technique.
Hi David,
Much as I would expect, you are doing very well digitizing the slides with the D850. I would expect very good performance. I am not clear on the lens used - if I'm not mistaken it would be the AF-S Micro NIKKOR 60mm f/2.8G ED macro lens, which is well regarded.
Turning to the comparison you posted, it is of course difficult to come to definitive conclusions about which approach produces superior results based on one photo vastly compressed for the limits of posting on an internet forum, but from what I could see, I'd have the following tentative observations (and I'm assuming here that none of these samples have been post-processed apart from resizing them for posting - is that right? - otherwise all bets are off in respect of comparisons.
One needs to evaluate such comparisons for a number of aspects of image quality: resolution and sharpness, tonal reproduction, colour reproduction.
For resolution and sharpness, the drum scan looks best, followed by the Nikon, followed by the V750. You can see it best in detail of the window curtains and distinctiveness of the tree branches. Assuming that the drum scan soaked out most of the resolution and sharpness of the original media, my experience with an Epson V750 indicates that this scanner can do better than what your image shows, raising a question about whether the feet of the slide holders are optimally adjusted, assuming you used Epson slide holders. I would also be interested to know what software you used with the V750 and what it's sharpen settings were. The drum scan is the only version that is complete on both sides, so I can't compare sharpness on the right side of the image. All that said, I suggest you select some slides that have good detail in the corners and see the difference between the V750 and the Nikon, once you've checked the V750 frame height. I suspect the Nikon may lead. To further investigate this aspect, as magnification plays a big role in perceived resolution and edge sharpness, it would be important to make the comparisons at the magnifications relevant to the print or viewing sizes of interest.
On tonal reproduction, I would again give first place to the drum scan, evidence of which is the shadow detail on the left side of the photo being much better revealed there than in the Nikon version. Unfortunately the V750 scan is incomplete on the left side, so it can't be properly compared.
On colour reproduction, one doesn't know what the real colours and contrast (which influences colour appearance) were in the scene being photographed, but to my mind, the Nikon result looks the best, followed by the V750, then the drum scan. The V750 appears to have a yellowish cast and the drum scan a cyan cast, but I'd be less concerned about these effects because they are easily fixed in post-capture editing. In fact, if the compression of shadow detail in the Nikon version can be opened-up in post-capture editing, it's a win win.
Given that tone and colour are much easier to reinterpret in post-processing than is resolution/sharpness, I think your Nikon solution looks viable, but a final impression should depend on being able to capture and evaluate the whole of the slide, the optimization of focus in the V750, what can be achieved in post-processing and finally how it all looks at the viewing magnification whether in print or on screen.
Bottom line - so far it looks to me that your Nikon approach seems pretty good.