Post-production
Processing * Digital * Cropping and Editing * Compressing your Photos
In the movie business, everything that's done after the cameras stop rolling is called post-production. This includes editing, soundtrack dubbing, music scoring, creation of the titles and credits, and a host of other things.
I'm borrowing the phrase for our purposes to mean things that are done after you stop shooting with the camera, up to the point where the finished pictures are ready for Web publication. This is relatively simple with a digital camera, but a little more complex with film. Let's start with film photography, shall we? (If you're shooting digital, skip down to Digital, below.)
Once you've finished shooting, the film is rolled back into its cassette and removed from the camera. Now somebody needs to process it chemically.
Usually people are shooting color print film, which can often be identified by having a name ending in -color: Kodacolor, Agfacolor, Fujicolor, and so on. (Slide films generally end in -chrome.) While it's not too terribly involved to process color slide film in a home darkroom, it's usually best to leave color print film to the professionals. So you can either drop it in a package and mail it to the lab of your choice, or you can trot on down to your local TawdryMart and give it to the high school kid running the Minilab (that's what they call those machines that do the whole processing thing).
When you pick it up (or open your mail), you'll generally find at least a set of prints and a set of negatives, cut into strips of four or five frames. If you value your pictures at all, the negatives should be saved in special negative sleeves -- these are plastic sheets, the size of a sheet of letter paper, with clear plastic sleeves for preserving film. The negatives go in the sheet, and the sheet gets labeled and inserted into a 3-ring binder. Do that right away, before something scratches the negative or your friend's 3-year-old knocks them to the floor and steps on them.
Some processors offer additional features, such as an "index print." This is what photographers call a "contact sheet," which traditionally was made in the darkroom by laying the negatives on a sheet of printing paper and exposing it to light. This produces one print sheet with little copies of all your pictures, from which you can select certain frames for enlargement.
A more recent invention -- about ten years ago -- is Kodak's PhotoCD. When you choose this, in addition to your prints, you get a CD containing VERY high-quality scans of your photos. This isn't cheap, but the scans are better than any you're likely to produce with your home scanner, and most graphic programs can read the PhotoCD format. (Note: This can't make a bad photograph better -- you might just get a really good scan of a bad photo.)
Alternatively, you may have chosen slide film. I don't generally recommend this, because it's harder to get prints or scans, but there are still some people using slide projectors. If you're one of them, there are no negatives -- slide film produces a "color positive," which has correct colors but it's transparent. In processing slides, the film strips are cut into single frames, which are then mounted in plastic holders about 2 x 2 inches in size. Slides can then be viewed in a viewer (or held up toward a white wall), projected on a screen, scanned, or printed. Like negatives, there are special plastic holders for slides, and I recommend that you use them.
So: Let's say you didn't choose PhotoCD (if you did, you can skip to Cropping and Editing below) but you have your prints or slides, and you have a scanner. (If you have slides, you'll need a special adapter to scan them.) You'll want to scan the photos at a high resolution setting -- this produces really big photo files, but we're going to edit them down anyway. (Note: It's not a bad idea to begin your cropping in the scanning process, by using a little judgement in what part of the photo you choose to scan. But don't spend a lot of time on it, because you get much greater precision in a photo editing program.)
Processing with a digital camera couldn't be much easier. Sometimes it means hooking a cable up from your camera to your computer. Usually it means taking the storage medium (SmartMedia, CompactFlash, Memory Stick, or disc, remember?) out of the camera and popping it into the appropriate device -- a reader, floppy drive, or CD drive. Then you run the download software, or in some cases you can simply copy the files from the medium to the computer with your operating system (usually Windows or Mac). If the photo is on a floppy disc or CD, you'll want to copy it to your hard drive for editing.
In most cases, you'll want to make an archival copy for backup, for which I recommend burning a CD.
And now, it's time to get out your photo editing software. You might be lucky, and have a copy of Adobe Photoshop, the perennial king of photo editing software. If you can't drop $600 on Photoshop, you can get close to the same power from JASC's Paint Shop Pro -- which is priced under $100, but is also widely available as a free, 30-day trial copy. I recommend it. You might also try Photoshop's little brother, Adobe Elements, or Macromedia's Fireworks, if you have them available.
A few notes on the editing process:
First: File formats. There's a good chance that the files came out of your camera with a .jpg extension, like this: Name.jpg. This means they're saved in the JPEG (Joint Photographic Experts' Group, pronounced "jay-peg") format, which compresses photos to take up less space. It's very effective, but it's also what's called a "lossy" format -- some data is lost each time the picture is compressed. For this reason, it's a good idea to edit them in a "non-lossy" format, such as PNG (Portable Network Graphics, pronounced "ping"), which is the native format for Macromedia's Fireworks program. Then, when you're done editing, you can save the final picture as a .jpg, producing a nice, compact file.
Next: Save your pictures with a new name at every step. If your picture is called "picture.png," make the new, cropped version "picture1.png." Name subsequent saves "picture2.png," etc. This is important, so that you don't overwrite the last step. That way you can always go back, to the beginning if need be. You can erase the extra saved files when you're done. If you overwrite your files, you might find yourself unable to undo something you've decided you don't like.
Now, on to the editing process:
While you probably framed the photo pretty carefully when you took it, it's rare that a photo can't be improved somewhat by a bit of cropping. This means cutting the sides, top, and bottom a little closer to get rid of extraneous bits, or simply to improve the composition somewhat. Sometimes you just crop a little, but don't be afraid to crop a lot. Remember: you still have the raw file, or the print, and you can always start over.
Once it's cropped, it's time to resize the picture. A 2-megapixel digital camera can produce pictures with a resolution of approximately 1200 x 1600 pixels, which is a lot -- about half a megabyte of data, saved as a .jpg. A scanner can usually produce high-resolution scans from 600 to 1200 lines per inch, or more, which can make humongous files. Kodak's PhotoCD files are also enormous. The reason we want to resize it is that it'll be shown on the Web, which has a limited bandwidth, and big files make slow downloads. In addition, we don't want to break the server's back with hundreds of 15-megabyte files. I'd like to see finished photos under 100 kilobytes, but don't worry -- I'll tell you how to get there.
Besides, if your computer isn't a speed demon, it can probably perform the editing functions on a small, 100kb file much faster than on a 2Mb file.
Let's start by resizing it. Most computer users these days are using a 17-inch monitor, though there are still a lot of 15-inch ones around. The 15-inch displays are often set to a resolution of 640 pixels (wide) by 400 (high), but a number of them can run at 800 x 600. The 17-inch monitors can typically show 800 x 600 or 1024 x 768. If we take 800 x 600, then, as a "standard" display setting, it would be kind to our viewers to make the pictures fit within that size. (I can resize them for the low-resolution side of our Gallery.) Since we're usually going to be using a vertical format, we want to make sure that the pictures are no more than 600 pixels tall, and somewhat shorter -- say, 500 -- allows room for the browser frame. So let's make a height of 500 pixels our goal for a vertical-format picture, and a width of about 600 the goal for horizontal-format photos.
A note on resizing: if the program you're using asks if you want to maintain the "aspect ratio," or proportions, of the photo as it's being resized, choose to do so. Then, when you enter the height you want, it will calculate the width to keep it in proportion. If this option isn't offered, you may have to calculate the width yourself to avoid distortion. To do this, divide the old height by the new height, then take the number you get and divide the old width by that. For example: A photo 1000 pixels tall by 700 wide is being reduced to 400 pixels tall. So, 1000/400=2.5. Now divide the original width -- 700 -- by 2.5, and we see that the final width should be 280 pixels. Simple, right?
Now we look carefully at the picture. A lot of times cropping and resizing are all a photo needs to be acceptable. Other times we may need to adjust other things, such as colors or contrast. If you want a little more light, look for a control called Gamma Correction. This can work wonders with a poorly exposed picture. You can also improve slightly soft focus by using the Sharpness control, but be careful -- Sharpness works by increasing the contrast between adjacent pixels. Too much, and it starts to look scary.
Familiarize yourself with these controls, and other editing tools you might have at your disposal. Remember, though, that we're looking for good photos documenting your costumes -- not underwater effects or oil-paint simulations. If a certain effect or filter doesn't improve the photo, use your Undo tool, and try something else. It won't take long before you're adept at improving photos with a light touch.
Let's recap a typical process:
Those five steps will improve most pictures. If your subject has redeye from a flash, add a step at 3 -- change the red pupils to black.
Finally, we need to save the final version as a compact file. If you've been saving as a .png or .bmp, the files will be pretty big, even after resizing -- very likely anywhere from 200 to 600 kilobytes. When you're done messing with the picture -- and only when you're done -- save it again, first in the format you've been using, then as a .jpg file. The .png or .bmp file is your final working copy, for editing purposes. The .jpg version will look pretty much the same, but some data will be lost in the compression process. You might not see it, but if you try editing it, you'll have less to work with -- that's why you saved the .png file.
Still, the 600kb .png file might very well come out as a .jpg of about 80-150 kb. If it's 80, that's fine. If it's much over 100kb, I'd like to see it reduced. How is this done?
This is a feature of the .jpg format: it uses variable compression ratios. That might sound like it belongs in an internal combustion engine, but it doesn't. It means that you can use the .jpg format to compress a picture to 90%, or 35%, or 100%, or whatever you choose. These percentages mentioned refer to percentages of the original picture, not the percentage it's compressed. At 100% it's not compressed at all. At 90% you might not see a difference, unless you enlarge it a lot. At 60% the differences will be subtle, but visible. At 20%, it won't look very good.
There are a number of programs that let you specify the compression on .jpg files, and there's a fair chance one came with your scanner or digital camera. MGI's PhotoSuite, which came with my scanner, lets you set the compression on .jpgs but doesn't let you compare the results. More useful are Macromedia's Fireworks or a program from Plenio Software Solutions called Web Graphics Optimizer. Both of these allow you to see your original, uncompressed file in one window, and the compressed version in another, giving you both the compression percentage and the size, in kilobytes, of the compressed file.
You may be amazed at the compressed sizes. Understand that compressing a picture 50% doesn't mean that the finished file will be exactly 50% of the original size -- more likely, it'll be about 30%. A 120kb photo, compressed to 70%, might easily come out at 65kb -- or it might be 40kb. The reason it varies is that .jpg compression saves each pixel with reference to its neighbors -- for instance, a large area of white has very little change, and can compress very neatly. An area with lots of detail won't compress so well, and at high compressions will lose most of that detail. So no two photos compress in exactly the same way -- you will need to use careful judgement to decide how much is enough.
But most photos can be comfortably compressed to under 100kb and still look very good. That's the sort of file I'd be very happy to see coming in for use in our Gallery.