EternalNY1

I don't get it. I just don't see how it could ever end up with a larger file. Because it's a compression algorithm, if it determines that the end result of the algorithm results in a file that is even 1 byte larger, it should discard the results of the algorithm and just return the original bytes. Because no information is added, I can't see why it would ever result in a larger file. If the JPEG algorithm in this case can't do its job (reducing file size while attempting to preserve image quality), then give up and just stick with the original. I can see no justification for me cropping 2000 pixels from an image and ending up with a LARGER file. That's broken.

I see what you're saying, but JPEG is a compression algorithm. If the original image has larger dimensions than the one that's being saved, I don't see any logic in the resulting file size being LARGER than the original. If I reduce a 10 x 10 pixel image to 8 x 8, under what circumstances can an INCREASE in file size be justified? The JPEG algorithm can't add additional information to the image, so the only logical outcome would be a smaller filesize, even at "100%". I'm clearly missing something here.

Maybe I'm missing something obvious ... but since an image can never be "improved" by software once the bytes are gone due to lossy compression, how can there ever be justification for a LARGER filesize after reducing the height/width? Even if I selected 100%, I'd expect a smaller size because for the same compression and a smaller image, less bytes are needed. The fact that I am making the image dimensions smaller AND reducing the compression ratio below 100%, and ending up with a larger file, makes no sense. I've been programming for years ... something doesn't seem right here.

I'm sure this has been asked before since it's such a basic question, but I searched and couldn't find. Why does Paint.Net save a 200 KB JPG file at 90% and end up with a larger file (by as much as 50k in this example), after I remove information (i.e. I will crop out a portion of the image, resulting in a smaller image in terms of height/width). I would assume that a 100% ratio would end up with a file of equal size (all bytes preserved), and anything less than that would be a smaller file (due to more compression and information loss). What am I missing?