We are pleased to announce a new utility you may be interested in checking out:

No Sleep Till...


Keep your Mac Awake
A simple program residing in the top menu bar used to control how and when your Mac is permitted to sleep. Currently beta testing. Please give it a try and contact us if you have any issues. Our beta testers also have the opportunity to get a free license code.


What are Hash Methods?

So I finally am finding time to get back to my library cleanup. I LOVE version 3 of ADD! Nice work.

The reject tagging is working great for me, with the propagation of ratings and comments to the 'original' file keeping critical information with one file.

I need more clarity on the different Hash Methods, as I am relying heavily on the reject tagging more than I'd prefer, but I have 250,000 images to get through and I dont want to do too much hand visual inspection on what is a dupe, original, unique, etc...

So.... because I'm counting on ADD to determine 'original' and 'dupe'..

What is Hash Method Automatic?

What is the most critical, MD5 or SHA1?

What is Direct Compare.

A lot of my "duplicates" may turn out to be a jpg and a raw file of the same photo. It appears on "Automatic" it found these as "duplicates". Should it?

The explanation of these would be good in preference settings pane.

Thanks,
Paul Clark

Hash Methods

Nice to hear from you again Paul.

Let me deal with Direct Compare first. This method does a bit by bit comparison of the underlying image files that are potential duplicates. If anything at all differers between the files, then they are not deemed to be "Hash Duplicates"

MD5 vs SHA1: using either will provide equivalent results.

With the Automatic hash method, ADD determines the way in which images are compared. The methods are either the Direct Compare, or by computing the SHA1 signatures of the files. If ADD finds only two potential duplicates (two images with the same file size) it will use the Direct Comparison. Otherwise (there are three or more images with the same file size) it will use SHA1. This was done for the sake of speed efficiency.

Automatic mode (in fact any of the Hash methods) should not be declaring the jpg and raw as duplicates. If you have any examples of this happening I'd like to get my hands on the files to see what is going on. ImageIQ on the other hand will see these as the same image and tag them appropriately. It may be that the the images in question were marked as "Date Duplicates" and ImageIQ determined them to be similar. Even if that was the case, ADD would not declare one as an Original and the other rejected.

I have a new release of ADD coming out shortly which improves on the ImageIQ processing and allows it to detect images that have been rotated in 90 degree increments as being similar. At the same time, I'll be improving the in-application help, and will include more information about the Hash Methods.