Sigh......... I'm surprised so many of you are so close..... yet so far away.
The simple explanation:They only need to save the coordinate numbers and "size" information of all your creature's pieces... and letting the computer generate the rest when you play the game.
This information stored is rather small and can be stored in very small files.
For anyone interested in the
more precise explanation:(Now, I'm not totally familiar with the PNG format, so some of this information might be slightly different)
(Otherwise, it is accurate for most image formats)
No, the data is NOT hidden in metadata.
The PNG's contain 3 sections of Image data, but that has nothing to do with the creature data.
Using 3 sections of image data allows the file's parts to be processed individually.... it's just a method used to allow slower computers to display parts of images before the entire image is downloaded.
As for the 'hidden data', it is stored with not only each pixel, but within each color of each pixel.
Computers interpret and display colors as red, blue, and green.
To store each color, takes 8 bits. To store each pixel, it takes 24 bits.
Now those 8 bits for each color, represent the numbers 0-255... which in turn represent hues (intensities) of that color.
Now, if you take real-life matters into hand, you are left with 256 intensities of a single color... the human eye can only distinguish between a fraction of those hues...
Which means, if you take the last bit (known as the "Least Significant Bit" or LSB) and wipe all of them out to 0.... you are left with an image, that according to the human eye, is exactly the same as the previous image.
(The trick here, is that you are only differientiating the colors by 1/256th of a hue... something that the human eye can not tell the difference between even when compared side by side)
Now, taking those LSB's, you can store 3 bits of information for each pixel stored.
To read the data back, you simply read only 8th bit within each pixel.
Not only that, but I've read that the creature files are utilizing "alpha channels" which in laymen's terms, are simply Transparent layers... Which again, obviously means absolutely nothing to the human eye.
Within these layers, you can again store more information within each byte (Byte = 8 bits) of the alpha channel.
I'm not completely familiar with image schematics, so i'm not sure how much data you can store within an alpha channel, whether it is the full byte, or again using only 1 of the 8 bits.
And that ends your lesson in image steganography.
Enjoy what little time you have left of the day.
I'm sorry to post so much, but I thought some of you guys might want to get your stories straight!
