35mm Slide Scan Metadata and Organization advice

[wbcamera]

Lightroom Version Number

Lightroom Classic 12.3

Operating System

1. Windows 10

I finally decided to digitize my fathers slide collection (40 kodak Carousels). I'm using an Epson Perfection 2400 photo with a 4 slide tray. Epson's software works on Windows XP so I created a Virtual Machine with XP got the scanner working. The slides are scanned at 2400 DPI TIFFs. I use a projector to pick out the slides I want to scan so I don't have to scan them all. The slides have a date like Jun 64 on them.

Questions:
1) What do you think of my settings of 2400 DPI TIFFS? I want to balance speed and still have a decent file.
2) What is the best way to add the date to the slides. What metadata field would you use? I looked at Capture Time to Exif not sure what the best option would be.
3) Any workflow tips? Folder organization for the slides? My current catalog has "2023-05 Shoot Name" for example.

 

[clee01l]

If you can assign a creation Date in the Scanning software then us 01mmyyyy for the slides reflecting the MMYY on the paper frame. Other wise apply this date by slide group using LrCs Capture date.
IMO, you should only organize by one of the default date schemes in Lightroom. Shoot Name is a Keyword and not a folder. You can see where "Shoot Name" is irrellevant for these images. If you can assign a correct 01mmyyyy for the scan batch, importing into one of the date named folder schemes is easy. Otherwise, after import, move the images by group to the appropriate date named folder.

 

[Paul_DS256]

Questions:
1) What do you think of my settings of 2400 DPI TIFFS? I want to balance speed and still have a decent file.
2) What is the best way to add the date to the slides. What metadata field would you use? I looked at Capture Time to Exif not sure what the best option would be.
3) Any workflow tips? Folder organization for the slides? My current catalog has "2023-05 Shoot Name" for example.

1. It depends on what you want to do with them after. Here's a link with some information scan to print. Also, not to take away from the helpful and knowledgably people here, here's a Scanning Forum. FWIW, I create thumb nail scans first. Quicker and I can then determine if the long scan is necessary.
2. I ran into several issues with dates like not knowing when the slide/negative was taken. Maybe if I was lucky, I knew the year. For this reason a lot of my create dates are dummy 1977-07-07 07:07. I ended up creating a filename convention that links the scan back to the folder/sleeve where the slide/negative is.
   
   I used the command line tool EXIFTOOL to set the following fields:
   • Camera Make/Model
   • Creation Date of image
   • Digitalized (Scan) Date
   • Film Information
   • Scanner Settings
   • Keywords on the type of film
3. Depending on the scan, I was using 'auto exposure','color restore' and 'ICE denoise' on my Epson.
   It's a lot of physical organization as well. It may be easier if to move your dad's slides to sleeves.
   As I said, I had an assignment for the sleeve and Epson's incremental numbering I wrote on the slide that was put in the sleeve. That allowed me to find the physical slide from LrC since the filename was the sleeve+slide #.
   There is a flow of prepping the next set of slides, taking the last slides off and marking with the Epson Slide number while the next set of slides is being scanned.
   

Good luck

 

[Califdan]

For scanning resolution, I start with a rough idea of what pixel dimensions I want the result to be. For example I may want the scans to be about the same as my DSLR which gives 4000 x 6000. So, if your're sanning a slide which for the sake of argument is 1" on the long side, I'd scan at 6000 PPI. However if I was scanning a print that was 10" on the long edge I'd scan at 600 ppi. In both cases my scanned image would be about 6000 pixels on the long edge.

I use an Epson flat bed scanner and using the SW that comes with it can create a set of profiles. I created a color and a monocrome profile for each long edge inch. For eacmple Color 1-2 inches, or BW 9-10 inch (only considering the long edge).

 

[dominique.gascon]

Sorry, it is bit long. I tend to be chatty.
I scanned 32 slide carousels and about 75 negative films (both color &and black & white) and here are some tips based on my experience.

To answer some of your questions:

2) What is the best way to add the date to the slides.

I use a plug-in called “Save Capture Time”. Contrary to Lightroom which adjust capture time by offsetting original values by adding a fixed value, the plugin allows you to apply a single value to selected photos or increasing it by a fixed amount.

Obviously on slides (or negatives) you don’t know when they were taken; the developed date the frame (if legible) will help to pin down the year and the month. Unless I knew the exact date (based on a specific event) I put the middle of the month and increased time from 12:00:00 by 1 second so all photos sort in proper chronological order (based on the number on the frame or on the film strip)

2b) What metadata field would you use?

I used another plug-in (Lens tagger) and added whatever data I could. I always tried to ID the film stock. It is written on the edge of the film strip, usually as a numerical code (you need to consult tables linking the codes to the films; the tables are easy to find on the web) – for instance 6031 refers to EKTACHROME 64 (Daylight). I never opened to glued cardboard frames and some films remains a mystery. The plastic frames had become in general use when I stated taking slides (1974), and I could generally ID the film type. You need gently pull the film out of the frame to read to info on the edge.

3a) Any workflow tips?

I use an Epson V850 pro which has holders for 12 transparencies or 3 film strips of 6 exposures. I scan using the Epson software at 4800dpi, which gives a 25 megapixels scan. I use Digital Ice (except for B&W for which it does not) with minimum adjustments during scan, preferring to do the adjustment in LR. I then edit the scans in photoshop to remove the dust and scratches left by DI and import in LR.

Digital Ice does not work perfectly with Kodachrome leaving artifacts on the scans. I eventually took to scan all my Kodachrome twice, with and without DI and using the scan w/o DI if DI left to many or unfixable artifacts on the scan. Since post processing take longer than scanning, it did not slow me down very much.

The scanner has an automatic mode where it detects the photos and a manual mode where you draw a frame around the photos. I realised late that the scanner trimmed the edge of the photos (rounded corner on the frames, edge of the frame, and holder or slides not perfectly parallel to the edge of the scanner). I had lots of cut feet or heads in tightly framed photos and I had to rescan in manual mode. I am not sure if it is worthwhile to scan systematically in manual mode and rotate and trim manually afterwards the photos. It would take longer.

3b) Folder organization for the slides?

By year (the LR catalog is a much better relational database is than Widows Explorer). I could have dumped everything in one folder, but small chunks are more manageable. I named my files CAR_XXX_NNN where XXX is the carousel no. and NNN is the slot number in the carousel. Ditto for film strips with sleeve number and frame number read of the film stock. The scans can be easily linked the physical photo if need be.



Have fun. It is a big job. It took me 3 years, on and off to do it. I have still 10 carousels from my dad, but it will be more difficult to catalog them; I also need to master colour restoration as many of his slides from the 50’s are badly faded.

 

[Paul_DS256]

I always tried to ID the film stock.

Here's a cross reference site of Kodak film numbers to stock.

Digital Ice does not work perfectly with Kodachrome leaving artifacts on the scans.

You may want to consider a 3rd party scanning tool rather than the Epson software if you have Kodachrome. See this article.

 

[Conrad Chavez]

Epson's software works on Windows XP so I created a Virtual Machine with XP got the scanner working.

If the virtual machine solution works for you, keep doing that. But if in the future you want to simplify your setup, you should know that you don’t have to use the 20-year-old Epson software. Many people who own older scanners use applications such as VueScan (what I use) or SilverFast to drive them on current operating systems. Those developers keep their applications compatible with both the latest OS changes and with a very long list of older scanners, and offer enough options and quality that you don’t really miss the OEM software. Sometimes they even add new features that were not available or possible in the original software from 20 years ago.

1) What do you think of my settings of 2400 DPI TIFFS? I want to balance speed and still have a decent file.

It looks like 2400 dpi is the maximum optical scanning resolution of the Epson Perfection 2400, right? If so, that’s a good choice to stay with, at least for scanning film. (Don’t bother setting it to any interpolated resolutions, only optical.)

I scan a lot of film at once, so as the price of mass storage dropped, I stopped trying to decide what resolution to scan at, because having to make the decision just slowed down the whole process. Now I just scan all film at the scanner’s maximum optical resolution, and know that it will be enough for enlargements. I do use TIFF with ZIP compression to lower the file size a little more.

If I have time some day, I will go through in Lightroom Classic and filter for all film scans that are keepers but not favorites (the ones still useful for family research or low-res online sharing, but might never be printed at a large size ). I might convert those all from TIFF to Lossy DNG in bulk, to recover a lot of storage space. But there’s still a lot of space on my photo archive volume, so doing that is not a high priority yet.

 

[wbcamera]

A lot of great info. I didn't know that 3rd party software could work. Rather not spend the money on it as it's a one time project. Any tutorials on using EXIF Tool that people recommend.

 

[Conrad Chavez]

If you aren’t comfortable working with command line tools, there are some utilities out there that provide a much easier point-and-click interface for Exiftool. One tool I know of for Windows is Exiftool GUI. There may be better ones than that, but I’m not familiar with them because I am using a Mac with other, similar tools.

 

[jmj2001]

I have been doing this off and on for a number of years with my father's slide collection (similar in size to your's) and my own from pre-digital days. I gradually evolved a workflow which saves quite a lot of time. I won't write it up here in detail but just mention a few points as suggestions. Of course many people have evolved their own ideas ...

1. I started with an Epson scanner(1200 XP I think it was ) with the slide holders etc but scanners are so slow that I realised I would never finish that way.
2. So I switched to DSLR scanning using an Opteka slide duplicator. DSLR scans have the huge advantage that you get a high-resolution, high bit-depth, RAW file in seconds and you can easily scan a whole carousel in less than an hour. The resolution is enough to resolve the film grain. The scans done this way are much better than I ever achieved with the (old) EPSON although I would not claim they are the best possible. Of course, you don't get Digital ICE.
3. I usually use aperture priority f8, auto-exposure and focus (sometimes with bracketing). I used to use a cloudy sky as light source but now I use a Kaiser Slimline Plano light panel for consistency.
4. I have another large cheap light box on which I lay out the slides to show any hand-written "metadata" - dates, places, etc. I take a photo of this with my smartphone usually to preserve the information linked to a thumbnail of each photo.
5. I import the photos and the metadata image to a collection in Lightroom Classic (LRC) . Generally I make a collection for each "film container" which might be a carousel, a box of slides, pack of old prints (scanned otherwise), etc. These collections go in a big collection set called "Film Containers".
6. Initially all photos get the Blue colour label in LRC to indicate that they need to be dated. This can be the most time-consuming part of the process so is often left until later.
7. Later I edit the Capture TIme metadata in LRC according to the best information I have. If I can only pin down the month, I set the 1st day at midnight or noon. Then I remove the Blue label.
8. Then I cull, rate, crop, adjust, etc. the RAW files. This can be done efficiently because, in many cases, each photo in a film container will require similar adjustments to start with.
9. After that you can do all the usual things like put the photos into collections according to themes, events and so on. They remain in their original Film Container collections as well.

I recently bought another slide (and negative) duplicator without a close-up lens inside so I can fully exploit my high-quality macro lens but I have not had time to try it yet.