You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
How to open file/folder from results?
Just click twice with right mouse button - this will open parent of selected folder in default file manager
or click twice with left mouse button - selected item will open in default app to handle such extension
How to speedup search?
Some of the tools(like duplicate finder or similar images) have special cache mechanism to save to file info about checked items and also calculated data(like hash), which will speedup later checking.
How to scan removable disks without needing to calculate again hashes of files after unplugging it?
In options there is option to not delete outdated cache automatically, which will preserve hashes of checked files. Be careful, because this may increase cache size significantly which will slow down scan operations(this will be visible only when checking millions of files). Below option to disable removing cache automatically, is placed button to manually remove outdated cache(cache entries which points to non-existent files)
Why in Similar Images mode there are several files with "Original" similarity label?
When files are hashed, it may occur that 2 images(usually very similar, but may happen that are completely different) will have exactly same hash like [23u8,12u8]. When checking for similar images, comparisons are handled at hash level so result can be [23u8,12u8] is similar to [22u8,12u8] with similarity 1. If multiple images have same hash, then are threaten as originals(even are different) and files from second hash are threaten as files with very big similarity.
Why there is such big number of completely different images in Similar Images mode?
Main reason are hashes collisions.
When using smaller hash size, it is more possible, that computed hashes of different images may be similar.
Such situation is even more visible when comparing several thousands of files.
To prevent such invalid behavior decrease number of checking images or just increase hash size.
The text was updated successfully, but these errors were encountered:
Is there a way on Windows to have multiple configuration (say with different search directory lists, settings for similar image) and be able to load specific config for a task. e.g. czkawka_gui /config:project_1 czkawka_gui /config:redit
Where can I find docs to read about tools
Documentation is available here https://github.com/qarmin/czkawka/blob/master/instructions/Instruction.md
How to open file/folder from results?
Just click twice with right mouse button - this will open parent of selected folder in default file manager
or click twice with left mouse button - selected item will open in default app to handle such extension
How to speedup search?
Some of the tools(like duplicate finder or similar images) have special cache mechanism to save to file info about checked items and also calculated data(like hash), which will speedup later checking.
How to scan removable disks without needing to calculate again hashes of files after unplugging it?
In options there is option to not delete outdated cache automatically, which will preserve hashes of checked files. Be careful, because this may increase cache size significantly which will slow down scan operations(this will be visible only when checking millions of files). Below option to disable removing cache automatically, is placed button to manually remove outdated cache(cache entries which points to non-existent files)
Why in Similar Images mode there are several files with "Original" similarity label?
When files are hashed, it may occur that 2 images(usually very similar, but may happen that are completely different) will have exactly same hash like
[23u8,12u8]
. When checking for similar images, comparisons are handled at hash level so result can be[23u8,12u8]
is similar to[22u8,12u8]
with similarity 1. If multiple images have same hash, then are threaten as originals(even are different) and files from second hash are threaten as files with very big similarity.Why there is such big number of completely different images in Similar Images mode?
Main reason are hashes collisions.
When using smaller hash size, it is more possible, that computed hashes of different images may be similar.
Such situation is even more visible when comparing several thousands of files.
To prevent such invalid behavior decrease number of checking images or just increase hash size.
The text was updated successfully, but these errors were encountered: