After I process a file for the web I save it in three separate places. I don't do anything extra to the files between each save, but each additional save procedure adds approx. 5-10 KB to the file size. It's not a big deal, but I was just wondering why a file would grow bigger when you're not doing any additional work on it.
Hmmm. Never noticed, but I don't save to multiple places. I save it to one, then work within the Mac desktop routine to drag and drop it to the other destinations, often as part of a set within a folder. I'd always heard (and verified) that resaving jpegs adds to the artifacts. Is that part of what's going on?
Carol and Hank - What you are referring to and linking to results when you continually close a JPG and re-open and re-save it in Photoshop. If you have a file open, and you just keep saving it over and over it won't ever change, since PS isn't uncompressing and recompressing the file each time.
Bill - To get back to your question, I've done something similar and haven't noticed a difference in sizes (but haven't looked closely). Is it possible you're saving to different sized drives, possibly with different file systems? Or is it possible that Windows/MacOS is reporting different drives differently? I've found OS file size reporting isn't necessarily consistent, although on the same drive I would expect it to be the same.
Other than that, I'd just mess around with it to try to find out what's going on - so Carol's suggestion to try "save as" or "save for web" etc is a good idea (it can't hurt after all).
Bill, maybe I'm thinking too hard about this, but if I understand the dilema correctly (you have an image open and save that same image 3 times - not save, re-save, save, re-save) my guess is the places you are saving your image to have different file allocation tables, and use different memory 'clusters'. In english, files can be stored as 'clusters' eg clusters of data of a certain size rather than all the data written contiguously. So if you save a file to a certain disk that has a larger cluster size, then that file may be a bit larger compared to a disk with smaller cluster sizes. The contents of the file are the same, but there is a little extra space in the larger based upon how the system utilizes memory. This is something you may notice with smaller files whose size is relative to that of the cluster size (eg save for web files). You can see if this is the case by save a small text file containing only a few letters, and save them to the same locations you save your images to...if their size differs as well, then that is the most likely answer.
Thanks all for your comments, but none really seem to hit the nail on the head (I think). It's not a dilema or anything, just wondering. I'm saving to the same HD, same system, same everything, it doesn't seem to matter about what degree of compression I use. As I mentioned, It's not a big difference in file size, just 5-10 KB. I always save to my NPN folder first because otherwise the size difference might throw me over the 200 KB limit.
Greg - you may be on to the answer perhaps, but I don't understand everything you said! I'll try your suggested experiment.
Like I said, this isn't a big problem or anything, just wondering.
Bill....my file sizes stay the same when saved in PS , but I have noticed when attaching an image file to an email or uploading to NPN, for some reason, the file is larger than what it was saved at, by anywhere from 10-30 KB more.
I did an experiment with saving JPGs and PNGs of the same small image (a less complex graphic than a photo) some time ago, and the same thing happened with the JPGs. Artifacts also increased. What Carol and Hank refer to is, I think, what's happening if you're simply "saving" as opposed to "saving as." The first saved image is compressed the way you want, and the compression algorithm simplifies the data (especially the data used to display detail-less and OOF areas). If you save that image again, the compression of the original JPG (which is, as far as we're concerned, the most nearly perfect) creates some artifacts. They appear mostly along borders between areas of contrasting light values, and since what was there in the first JPG was as smooth as you could get it, the weird little pixels require more data to be included in the file in order to display them. Then if you save the already-saved image (the 2nd version of the JPG that you made to post), it has a few more artifacts that require a few more bits to record where they are and how to display them.
You can test this by cropping a photo to be a little piece of where the sky meets the earth and has a couple of little tree-shapes there, too. Save it as a JPG with a compression of, say, 10. Then open the cropped image and save it again (not "save as," just "save"), and close the image. And open again and save again. Image file size will increase and image quality will decrease. My 'experiment,' with pictures and numbers is here.
The conventional wisdom is to "Save" an image as a JPG once, then if you open it again to "Save As" each time. Doing this is a pain. But I really hate JPG artifacts. To make matters worse, in Paint Shop Pro (which I use), if you merely "Save" instead of "Save As," the Exif data is stripped from the file. When that happens, I get really mad at Corel. Having told you more than you wanted to know (an occupational hazard even though I retired), I'm off to post the plea for help with an almost-Photoshop-specific technique that I came to this forum to post. But then I found this fascinating topic.
Thank's Rosemairie. I think you may have hit upon the answer; that PS is adding additional compression (or re-compressing) on each addition save, which may add a few extra KB each time. i has never thought of that. I always use "Save as" to save my files so I can slightly change my title depending upon which folder I'm saving it to.
Bill, this is a result of the compression algorithm. In fact if you use "Save As" and look at the file size you will sometimes see that quality level 7 or 8 may produce a smaller file than quality level 6 or similar counter-intuitive result. Compression introduces artifacts, which then become part of the image and look to the next compression cycle as additional image detail so file sizes may grow with multiple saves.
This whole issue and its associated confusion can be completely avoided by religously following one rule: Never, ever modify a jpg file.
If you are going to work on an original (straight out of the camera) jpg file ALWAYS save it to a lossless compression format (TIFF, PSD, Pspimage, etc.) before making a single change. From then on, always use the new file as your master file. Any work you do on the "master" should be saved as a new full size lossless format file with a different name. When preparing an image for the web, make your changes to the lossless file (downsizing & sharpening) then save-as a new file in the same lossless format before converting it to a jpg. I always use the same name except for adding the new image size. For example; "mountain.tif" becomes "mountain 720x480.tif".
Once the downsized file is saved in the lossless format re-save it as a jpg for upload.
One advantage of the method is that you can quickly go back to the lossless downsized file and re-save it as a jpeg with different compression if you don't like how the first attempt looked on the web site. You don't have go back through the downsizing and sharpening process.
Anytime you want to rework an image always go back to the "master" or lossless version of it. Once again, following the rule: Never, ever modify a jpg file!
Discussion
15 repliesAfter I process a file for the web I save it in three separate places. I don't do anything extra to the files between each save, but each additional save procedure adds approx. 5-10 KB to the file size. It's not a big deal, but I was just wondering why a file would grow bigger when you're not doing any additional work on it.
Thanks.
Guy
This article may help a bit. Evidently as we save and resave jpeg images, they become more complex.
Have you tried doing a save as option?
Carol
Bill - To get back to your question, I've done something similar and haven't noticed a difference in sizes (but haven't looked closely). Is it possible you're saving to different sized drives, possibly with different file systems? Or is it possible that Windows/MacOS is reporting different drives differently? I've found OS file size reporting isn't necessarily consistent, although on the same drive I would expect it to be the same.
Other than that, I'd just mess around with it to try to find out what's going on - so Carol's suggestion to try "save as" or "save for web" etc is a good idea (it can't hurt after all).
Greg - you may be on to the answer perhaps, but I don't understand everything you said! I'll try your suggested experiment.
Like I said, this isn't a big problem or anything, just wondering.
You can test this by cropping a photo to be a little piece of where the sky meets the earth and has a couple of little tree-shapes there, too. Save it as a JPG with a compression of, say, 10. Then open the cropped image and save it again (not "save as," just "save"), and close the image. And open again and save again. Image file size will increase and image quality will decrease. My 'experiment,' with pictures and numbers is here.
The conventional wisdom is to "Save" an image as a JPG once, then if you open it again to "Save As" each time. Doing this is a pain. But I really hate JPG artifacts. To make matters worse, in Paint Shop Pro (which I use), if you merely "Save" instead of "Save As," the Exif data is stripped from the file. When that happens, I get really mad at Corel. Having told you more than you wanted to know (an occupational hazard
Greg - IMO everything having to do with computers is WAY more complicated than they catually are!
Guy
If you are going to work on an original (straight out of the camera) jpg file ALWAYS save it to a lossless compression format (TIFF, PSD, Pspimage, etc.) before making a single change. From then on, always use the new file as your master file. Any work you do on the "master" should be saved as a new full size lossless format file with a different name. When preparing an image for the web, make your changes to the lossless file (downsizing & sharpening) then save-as a new file in the same lossless format before converting it to a jpg. I always use the same name except for adding the new image size. For example; "mountain.tif" becomes "mountain 720x480.tif".
Once the downsized file is saved in the lossless format re-save it as a jpg for upload.
One advantage of the method is that you can quickly go back to the lossless downsized file and re-save it as a jpeg with different compression if you don't like how the first attempt looked on the web site. You don't have go back through the downsizing and sharpening process.
Anytime you want to rework an image always go back to the "master" or lossless version of it. Once again, following the rule: Never, ever modify a jpg file!