There is a reason that “spot the difference” images have been a mainstay in many magazines for a long time now. Simply put: Spotting small differences in images is really hard.
The intention with this little tool is to make spotting differences in images quick and painless.
But who would need something like this? Well... besides the obvious case of the hardcore “spot the difference” player looking to up his game, there are many cases where you need a tool like this: A web-developer comparing layouts in two screenshots, a graphics-designer with badly
named files looking to find the difference between “121a.jpg” and “121b.jpg”, someone comparing two letters for any differences, and many more.
It's a bit surprising that there aren't more online tools for comparing or diffing images right now. There is a number of great libraries capable of making very quick and easy comparisions, but they are a bit hard to use for most people – ImageMagick for example is an excellent
tool,
but a bit hard to install and requires the user to be familiar with the command line.
The “Fuzz” parameter describes how exact the comparison between the images is. A fuzz of 0 means that the comparison is extremely strict and shows even smallest differences. A high fuzz value means that only very strong differences in the image show up.
The “Highlight Color” is simply the color in which the differences will be highlighted. It has no effect on the actual comparison – it's just a convenience for the user.
The comparision itself is done through ImageMagick.
Currently we allow images with the following 3 Mime-Types:
The maximum file size for the images to compare is 16MB.
If the images are of different size, the bigger image is resized to the size of the smaller image.
During the resizing the bigger width/height is squashed to the smaller width/height. So the resulting image has the shortest width as width, and the shortest height as height.
The library used for the comparison can handle different Mime-Types, but there might be some inaccuracy.
Yes – temporarily. We regularly delete old uploads, but they are stored temporarily in order to be processed and served.
Technically yes, if they can guess the random strin g we use as a name for the file. However, that is very unlikely.
If there is some sort of problem, you can feel free to to contact us through the contact form and we'll see if there is anything we can do.
Maybe. :) Just write us a message through the contact-form and we'll see if there is anything we can do.