The age old choice between image quality and file size, thanks to our friend JPG.
One from the inbox this morning. I suppose it couldn’t hurt to share my response here:
Let’s say you’ve a JPEG file, a photo straight from your camera-phone, and its filesize is 32k. You open that file into Photoshop, don’t make any changes to it, and then want to save it again (don’t ask why, you just do). You’re asked to choose a quality/compression level. You select one, and save your new image over the old one.
Now, my question is: how do you save the file so that it’s not loosing any more detail to the JPEG compression algorithm, but at the same time you’re not tripling the file size?
Short answer: you can’t.
Because JPG is a lossy compression, no matter what you do with the settings, if you open up a file and re-save it as a JPG, you will lose image quality (whether you perceive it or not).
When you open a JPG, it’s converted to the internal bitmapping model of the editor (we’ll again assume Photoshop here) and converted to a full, uncompressed 24-bit image. From that point on it’s lossless — aside from any quality degradation from the camera’s original save — until the point when you save it again.
The JPG algorithm isn’t reversible so nothing from the original save influences the new save, which is why you won’t be able to achieve an identical file size or quality level. Instead, Photoshop recalculates the compression as if it’s working on an uncompressed image, which it technically is. Therefore the results you get when saving are entirely dependent on Photoshop’s JPG saving options.
The choice between a huge file size or better image quality is simply a consequence of using a lossy format. The only way to maintain the integrity of the image after the save is to choose a lossless format like PNG or TIF, but then your file sizes suffer. The only way to keep file size down is to drop the quality settings, but then your image suffers.
But hey, we’ve been dealing with that choice on the web for years now anyway.