By error I now have duplicate photos which I don't need. I have gone through the listing duplicate files process but all I see are photos that PSE 2020 consider as similar. This is not what I want to see. If there is a way, I would like to see only the exact duplicates. Further to this I would delete all duplicates but keep the originals. So to me there is another question - are the original photos retained without showing them as duplicates. I hope I have made this problem clear enough.
There are several ways to define 'only exact duplicates'.
Because there are different reasons to create duplicates by error and different comparison criteria. There is nothing like an 'original', at most a file which has been imported first. And when you have a set of files you consider as duplicates according to your own criteria, most of the time, you are the only one able to tell which to keep and which to discard.
Let's look at the criteria:
- same file name? Wrong, many cameras will use the same file name once the max number of files allowed in the naming scheme is reached. I have many IMG_0001.jpg files, they are NOT duplicates.
- Dates? The most relevant date type is the 'date taken' from the camera. That is not relevant for scans and even for burst mode batches.
- File size in kilobytes? That's an excellent criteria if you add the date_taken. Both criteria are used together to filter importing from the downloader.
- Visual similarity? Very important when you have different versions of the real original shot in different sizes and times. That's not used by the organizer as a criteria for filtering at import time, but to find possible such duplicates.
There are many free or low-cost duplicate finders. You can even define a number of rules to help finding sets of duplicates. The problem is to sync your choices with the catalog. If you delete the files externally, they will appear as 'disconnected' and you'll have to select them and delete them from the catalog.
The visual similarity solution offered by the organizer lets you 'stack' the possible duplicates so that you only see one when browsing. If you select which one should be the 'top of stack', then you can flatten the stack. That deletes the duplicates.
I always advise to start by searching the reason of the duplicates. Searching and stacking the dupes is the second task. Deleting the dupes comes last and keeping the stacks is often enough for me, especially for scanned old family photos coming from different sources. In that case, drive space is not the problem, it's the ability for visual searches. If I have some time working on such stacks, I choose the best (not the 'original') and flatten the stack.