My OLD Canon RAW Workflow

Text and Photos Copyright Matthew K. Whitley

A New Version of this Article Is Posted HERE!

Many people that use Canon digital cameras ask the question "why use RAW format?" or seem to be intimidated by the idea of doing so. The steps involved in processing and editing images in the "digital darkroom" are confusing enough to learn without using RAW, so why bother? The RAW format provides a "digital negative" containing the untouched data as it came off of the cameras CCD or CMOS sensor. Many of the in-camera settings such as sharpening, saturation, and white balance can be chosen after the shutter is pressed by applying these settings to the RAW data on the computer. Lastly, JPEG compression used in cameras is a "lossy" format meaning that some data is lost in compressing the file. Shooting RAW allows you to bypass this compression and retain the maximum amount of data in your images. The following is a description of how I am currently processing my RAW files from my Canon EOS-1D, but will also be applicable to D30/D60/10D/300D images. Do remember that this is only the way I do things and that you should strive to develop the workflow that works best for you.

1. DOWNLOADING IMAGES
The first step in any workflow will be to get the images from the camera to the computer. My main computer is a laptop and I use a 1GB microdrive as my "digital film" so for me the first step is to use a PCMCIA card reader to download my images. I store my RAW images in a folder named RAW (original, eh?) and have subdirectories for each year. Within those year directories, I further have directories for month, and day. So for the images I shot at Jamaica Bay on 9/12/02, I would download my images to C:\RAW\2002\09\02 Jamaica Bay\. This system makes retrieving images quick and easy and facilitates backups. At this point I also often backup the new files to an external Firewire hard drive.

2. VIEWING / EDITING IMAGES
The next step is viewing the pictures and determining which are keepers and which belong in the recycle bin. For this, I use Breezebrowser ( http://www.breezebrowser.com). I open the directory where I placed my images and view the thumbnails in the breezebrowser display. At the thumbnail menu I scan the images and delete any obviously out of focus or horrible images. Then I start at the first image and double click it to bring up the larger preview with the full EXIF data and histogram. I go though the photos one by one as a traditional photographer would on a light table. I make my best effort to delete images I know I'll never use (although I tend to keep more than I should). I find this step to be an important learning process because I can correlate my good or bad images with the EXIF data and evaluate different techniques that I used in the field. Also during this step I mark outstanding images with the check mark button so that I can know which ones were the best from the previous thumbnail view.

3. BACKING UP IMAGES - UPDATE!
A major concern for photographers, film and digital alike, is protecting one's images. There are many different views on when and how to backup images and I could easily write an entire article on it, but I will cover it briefly here. After downloading and doing a brief edit of my images, I back them up to CD. This has recently been made much easier thanks to a great program called Archive Creator http://www.archivecreator.com/. This program was written by the same people that brought us YarcPlus and is currenty in a pre-release form. Archive Creator takes set of images and burns them to CD with an HTML index file complete with thumbnails of every image on the set of CD's. I use AC to create a CD backup of my images every time I return from the field. In addition to this backup to CD, I also keep a copy of my images on an external 120GB FireWire hard drive. Thus, I have 3 copies of each file: 1 on my computer, 1 on the external HD, and 1 on CD. Ideally, the CD's would be kept in a different physical location, but to be honest, I seem to keep them at my house. Recently, I ordered a DVD burner and will soon be archiving all of my images to DVD-R for more security. More on that when it arrives...

4. RAW CONVERSION
In order to manipulate images in photo editing software such as Adobe Photoshop requires that RAW files be converted to a format that these programs can recognize. This is where many people get confused/intimidated by the RAW format, although it is really quite simple. The two most popular software programs for converting RAW files are Breezebrowser and Yarc Plus (http://www.yarcplus.com), both of which are fantastic. I use YarcPlus to convert my images because I feel that it produces the highest quality images due to its proprietary Artifact Reduction Filter (ARF). After choosing which images I wish to convert using Breezebrowser, I open YarcPlus and convert the images to 16bit TIFF files (non-linear). I rarely use linear conversion if I have a borderline blown highlight and want to attempt to recover detail. In other situations, I can see no difference in the linear/non-linear files. I tend to use standard sharpening with a fine setting and color matrix 3 (although I do use matrix 4 when I want the widest color gamut).

5. PHOTOSHOP
Next, I take the TIFF files and open the first one in Adobe Photoshop 7. Because the file is 16bit, some options are not available without converting to 8bit mode. After opening the file, I first choose the levels option (Image-->Adjustments-->Levels). The histogram represents the relative number of pixels at each brightness level...on the left side are the darkest pixels and on the right side are the highlights. Spikes on either end of the graph may represent blown highlights or detail lost in the shadows. I tend to bring both the white and dark slider towards the middle of the graph just a bit until the arrows line up with the beginning of edge of the graph. Often, this increases contrast in the image and can make a dramatic difference. I do find that a well exposed picture will require the least manipulation at this step. After the levels adjustment, I rarely tweak the levels by using the curves command (Image-->Adjustments-->Curves). At this point, I sometimes increase color saturation to mimic Velvia film by using Fred Miranda's Digital Velvia (http://www.fredmiranda.com).

The last step in my workflow is to sharpen the image. Many different opinions exist on the best way to perform sharpening. Some people use the select or lasso tools to select the subject and apply selective sharpening, which reduces the noise that is generated by sharpening. Others prefer Fred Miranda's actions such as CS Pro. In the past, I have sharpened my images with the standard unsharp mask tool, but I have recently begun using Fred Miranda's 1D CS Pro V2 Photoshop Actions to do my sharpening. These are automated little programs that function within photoshop. I did not like Fred's previous actions, mainly because I thought the sharpening was too aggressive and created artifacts. This new action, however, is superb and provides much better options for sharpening. With the 1D files, I find that I need to do very little sharpening so I typically use CS Pro set on Very Low Intensity or Low Intensity. The beauty of this action is that it using a series of steps to intelligently identify the main subject and selectively sharpen that portion of the image. The advantage of this approach is that it avoids the increased noise in the background areas and haloing that can occur with the usual Unsharp Mask. For $15, it's a steal!

6. SAVING
At this point, I save my final image. If I know or think that I may want to print the image, I save the file as a TIFF file. If I don't think I will print it, I save an image in JPEG format (level 10) under a file folder structure identical to the one above. It is important to remember to convert to 8bit TIFF for sending files to most printers or saving as JPEG (actually, PS won't let you save as JPEG in 16bit format).

7. PRINTING - UPDATE!
Only recently have I begun printing a majority of my images. Many people prefer to print on one of the many high end ink jet printers now available. I have made the decision, at least for now to send my images out for printing. The vast majority of my images are printed on a Fuji Frontier machine on professional quality photographic paper by epixel.com, but I also print some images on a Lightjet printer from pictopia.com. Both produce superb output, indistinguishable from traditional photographic prints to my eye. I should note here that the following "print preparation steps" are performed prior to applying any sharpening. I always perform sharpening as the last step before saving the file, regardless of whether it is destined for the web or the printed page. To prepare my images for printing, I first crop into the appropriate aspect ratio. Most digital SLR's produce images with an aspect ratio of 3:2 (that of 35mm film). Most of what I print is in the 8x10 format, so I select the cropping tool in photoshop and use the option for a fixed aspect ration (which I set to 8 and 10). I then select the area which I want to print and crop the image to that size.

At this point, I go into Image-->Image Size and make sure that the checkbox for resample image is unchecked. I enter the appropriate image size (in this case 8 x 10 inches) into the Document Size boxes and click ok. I then open the same menu and look at the resolution. I prefer to have the resolution at 240 pixels/inch for printing. If this number is below 240, I then check the Resample Image checkbox and make sure it is set to Bicubic. I then enter 240 into the resolution box and press ok. This "upsamples" the images and increases the file size. If the value for resolution is significantly less than 240 when I open the Image Size box, I will do the previous steps to increase the resolution, but instead of entering 240 right away, I do so in 50 pixel increments. In other words, in the original value was 100, I would set it to 150 and click ok. Then I would open the Image Size box again and enter 200, repeating this until the target resolution of 240 is achieved. The reason for this is a s follows. When you increase the resolution of an image as I just described, you are telling photoshop to make the image larger and thus create data where it didn't exist before. This causes Photoshop to make "guesses" as to what the best color value is for the new data it generates. It does this by looking at adjacent pixels, measuring their values, and making a guess as to what the new pixel should be. Some have suggested that Photoshop's guesses are more accurate if you do this upsampling in small increments (i.e 50 pixels at a time). After getting the image into the appropriate aspect ratio and resolution, I then save the file as a TIFF (generally uncompressed) to send to the printer. In the future, I hope to delve into ICC profiles and color management, but right now I find that what I see on my screen matches what I get in print very closely.

That's it! Once you get the hang of it, working with RAW format really provides an enormous amount of flexibility in processing images. I hope that this has helped and please feel free to email me at matthew@mkwphotography.com with any tips or questions!