If you already use a RAW workflow and are comfortable with it, this article probably isn’t for you.  However, I’ve found with the advent of Lightroom many people are using  a “RAW workflow” (because they use Lightroom) without realizing it, but don’t shoot in RAW.  Occasionally this is an informed and conscious decision, but most of the time they do not really know what a RAW file is, how it is used in a RAW workflow, what happens to it in a RAW workflow , or what the advantages are of a RAW workflow.

(It has become somewhat common practice when referring to RAW files to capitalize the word RAW.  Some do it, others don’t.  I’m not sure of the reasons, since it isn’t an acronym, but I got in the habit of it a long time ago, hence you will see me using the all caps version)

Before Lightroom, using RAW in a workflow was more like a WORKflow!  Of course, any workflow was challenging let alone adding the complexities of shooting RAW. Lightroom (and Aperture) have changed that forever, not only simplifying workflows dramatically, but also making a RAW workflow as painless as any other workflow.  If you are new at using RAW, or are considering shooting RAW, understanding what a RAW file is and how it is used in a RAW workflow may be helpful.

EVERY DIGITAL PICTURE STARTS OUT AS RAW

Yep, even that cheap point and shoot.  Even that little camera in your cell phone.  Is this important?  It may help to understand the overall concept of a RAW workflow.  So let’s start at the beginning.

A digital camera replaces film with an electronic sensor to record the image.  This sensor is composed of millions of photosites (sometimes referred to as pixels).  As an analogy for this, each photosite is like a small bucket which collects photons when they land in it.  An “exposure” is created by measuring the number of photons received by each photosite during the same interval of time.   The relative number recorded by each photosite can be used to represent the brightness of each individual photosite.

Unfortunately, the photosites themselves are “color blind” ... they can’t measure the wave length of the incoming light, only the amount of photons.  To make color images possible, nearly all digital cameras use what is referred to as a Bayer sensor, (named after it’s inventor, Dr. Bryce Bayer of Eastman Kodak).  With a Bayer sensor, each photosite is covered by a filter, either Red, Green, or Blue.  This means each photosite records only one color of light.  (If you want a more in depth explanation of how sensors work, including how the Bayer sensor reproduces color there is an excellent article complete with illustrations  at Cambridge Color, “Digital Camera Sensors”.

The information from the sensor (which is actually an analog device) is converted to a digital value, which represents  the “level” of light it received.  The number of levels that can be represented is based on the sensitivity of the sensor to distinguish differences, the more sensitive the sensor, the greater the number of levels, and this is evidenced by the bit depth the sensor records. So once we have made our capture we know two things from each photosite... the brightness level and what color of light it represents, based on the filter that covered it.  This then becomes our RAW data - not a RAW file yet, just the RAW sensor data.

To be useful we also need some additional data ... data about our data.   You’ve probably heard the term metadata used, most commonly referring to things like date and time, lens, f/stop and shutter speed. So metadata is data about the actual captured data, and some aspects of metadata are important when it comes time to create a visible file.

With these two elements, we now have the ability to process the data into a viewable image.  This is a very complex process  ... after all we don’t actually know the color that any actual photosite represents.  All we know is how much red, blue, or green light it received based on the filter that covered it.  This process is called demosaicing , where complex algorithms use the pixels surrounding each pixel to calculate the actual color a photosite should represent.  As it calculates this information it will also apply various aspects from the metadata, such as white balance.  When finished it has calculated an RGB value for each photosite which becomes of pixel of the visible image.  Immediately after an image is captured, nearly every camera will go through this process to create a visible image.

What happens at this point depends on your camera and it’s settings.  Camera phones and almost all point and shoots will take this image information and then create a standard format jpeg file.  All of the actual sensor data and some of the metadata is then discarded. 

Some high end point and shoots, and pretty much all dSLR’s allow you to create a file which contains all of the original RAW data from the sensor and all of the metadata ... this then becomes what is referred to as a “RAW” file. 

OK, SO WHY IS THIS RAW FILE BETTER?

It is important to understand how much influence metadata has on this conversion process.  Every file has at least one item of metadata that will influence the outcome of the RAW conversion, white balance.  Every camera also has additional items of metadata designed to optimize the image, affecting things like contrast or saturation. These settings can be automatic (in other words the camera makers best guess), or set by the user.  For example instead of auto white balance, you may choose to set a white balance in the camera itself.  No doubt you have seen various other settings in cameras which are designed to supply metadata instructions for the conversion process.  They often have nicknames, such as bright or vibrant and most of them affect various aspects of the conversion process.  So some parts of the metadata are like a set of instructions for the raw processor to use when converting the data into a visible image.  What is critical here is this is just metadata ... it has nothing to do with the RAW image data, and is only used when it comes time to render a visible file from that data.

If you discard the RAW information by not saving a RAW file when shooting, you are stuck with whatever the camera creates in the jpeg file.  The camera usually does a pretty good job at rendering a viewable image, but not always.  If you choose to save the RAW file, all of this metadata is editable, and your computer (which is much more powerful than the camera) can then render the image using the new metadata settings.

As an example, lets suppose the picture is “over exposed” ... the image created by the camera is a little light and washed out.  On the computer we can decrease the exposure, so the resulting image file will be normalized.  Since the RAW data is strictly linear, moving the exposure like this does not create any problems with color shifts, as long as we didn’t actually clip any pixels (expose them to so much light we surpassed their maximum recording capability so we can no longer distinguish differences in brightness levels). Additionally, this decrease in the exposure is additional metadata for use when the file is rendered ... it doesn’t change any of the actual image data captured by the camera.  Another very common example is white balance.  Perhaps the camera was fooled and has assigned standard daylight as the white balance.  However  you know the scene was taken in overcast conditions.  You can override the default white balance with a new white balance setting, and apply this new setting when demosaicing the file.  In fact, you can manually tweak the white balance, or use a dropper tool to select an object that you know to be neutral in color.  As long as you have saved the RAW file, the white balance set by the camera is really irrelevant, because you can always set or adjust it to a correct white balance.

So if we have a RAW file on the computer we can use a RAW processing program to edit the metadata and then render out a new version of the file - in fact that is the very purpose of a RAW processing program, or RAW converter as they are typically called. In the RAW converter we can even add metadata to further improve the file.  This metadata editing is extensive ... we can modify exposure, contrast, white balance, brightness, black levels, white levels, saturation.  We can create curves, remove noise, and sharpen the image.  We can indicate cropping and rotation adjustments to be applied when the viewable image is rendered. With Lightroom and Adobe Camera Raw,  we can even do many of these things to parts of the images, called local adjustments.  All of this information can be used as part of the rendering process, resulting in a higher quality image file when we decide to render it. At this point we can also render out a full 16/bit Tiff file, so our image file can be furthered modified in a pixel editing program such as Photoshop. Because of the high quality and amount of data, this pixel editing will allow us a much higher quality final image file than if we were to do the same steps with an jpeg file made in the camera.

BUT RAW FILES ARE SO BIG, AND ALL OF THAT RAW PROCESSING IS SO SLOW ..

These are the two main complaints normally leveled against shooting in RAW.  A RAW file is substantially larger than the compressed jpeg version the camera renders from it.  In the past, nearly every RAW file ended up with an equivalent tiff file so you could print it or edit it further, and these are even larger than a RAW file. The necessary storage was expensive.  All of that work to render out those RAW files seemed excessive.  Things have changed however.

First, the economics of storage has changed dramatically.  Multi-gig compact flash cards can be had for under $15 ... I just bought 2 4gig cards at Costco for $25.  They’re not the fastest out there, but they are plenty fast enough for even my p65+ back.  I know many photographers take enough cards when traveling to never have to reuse one as sort of a backup system ... I have 16 4gig cards for my Italy trip for use just on my 5D Mark2.  Terabyte hard drives are now common place and cheap, and even laptops can be upgraded to 500 gigabyte drives for under $200.

Using Lightroom the workflow is exactly the same as when you work with jpeg files or any other image format.  The entire RAW process is really transparent to the workflow. For many images you can do all of the modifications in Lightroom with metadata edits, and then output the results to a printer or a jpeg file without ever having to render a tiff file. Most photographers using Lightroom estimate they use Photoshop as much as 90% less than they used to. 

BUT ISN’T THIS LIKE USING PHOTOSHOP TO ARTIFICIALLY ENHANCE THE IMAGE?

Many photographers are concerned about the realism of digital photography, and try very hard to not let the abilities of a digital workflow influence their images to an extreme degree.  I’ve heard a few express this is why they do not shoot in RAW ... they are going to get it in the camera.

The confusion here is getting it in the camera is sort of like saying “getting it on film”.  But then I would ask what type of film, and what are you going to do in the darkroom?

The reality is the camera  does not see the world like a human (true with both digital and film cameras).  There is no “interpretive” process like our brain, and a RAW file without any metadata will deliver a pretty horrible image.  If you leave it up to the camera, you aren’t eliminating metadata or the alterations it will make to try and make the image look like something we would expect to see.  Rather you are letting a “computer” and the person who programmed it to make educated guesses via the camera’s firmware decide for you.  Nothing special or more natural about that. In fact, it’s probably completely opposite of what you are trying to accomplish.  If indeed this is your personal philosophy you are much better off taking control of it yourself.

IMPORTANT POINT TO REMEMBER ...

   

When you are editing your image in a program like Lightroom or other RAW processor, it is important to understand you are only editing the metadata, not the actual RAW data from the sensor.  Yes, you see all of the changes on the screen ... because every time you make a change the program will take the RAW data and the new metadata and render a new preview of the image.  But no matter what you do, you never actually modify the image data itself, and you can always revert the image to the original state that came from the camera.  When in Lightroom, everything you do to an image is non-destructive, because you are not editing the image itself.

As I mentioned, a few years ago a RAW workflow was pretty time consuming.  Many photographers got very good at using the camera settings to manipulate the metadata to get what they wanted when they took the pictures.  However, now using a Lightroom workflow, it is actually easier to use a RAW workflow for most applications.

So Lightroom is the nearly perfect marriage of a DAM (digital asset management) program, and a RAW converter.  You can keyword, rate, purge and adjust files.  You can create collections so related files from various locations can be seen as a group together. Most of the time you find groups of files requiring the same changes  and after adjusting one, you can sync any or all of the new settings to a group of files.  If you find nearly every image you import needs the same basic changes, you can actually create a preset and apply that when importing.  It isn’t long until you fly through the adjustments to your captures after a shoot.

This isn’t really a tutorial about using Lightroom ... my main goal is getting photographers to try Lightroom and a RAW workflow. If you are interested in learning how to use Lightroom more efficiently, there are quite a few good resources.  First, you may want to check out the video training available from Luminous Landscape.  Two of these have to do with Lightroom ... Luminous Landscape Guide to Lightroom 2 and Luminous Landscape Guide to Asset Management.

Perhaps the best book about using Lightroom is Martin Evening’s book, The Adobe Photoshop Lightroom 2 Book.  Scott Kelby’s book The Adobe Photoshop Lightroom Book for Digital Photographers is also a good resource.

There are several sources for video tutorials all over the web.   You might want to start with some from Adobe.  Lynda.com has several, although they are not free.  And if you are a member of NAPP, there are plenty of resources there including a podcast.

At it’s very basic level, Lightroom is not intimidating at all.  If you haven’t tried it, download the free 30 day trial.  Load it up, and play with it.  Some important points ...

1. 1. Lightroom doesn’t organize your image files.   You do that within the framework of your own operating systems file structure.  What it does do is make it easy to find where they are on your hard drive(s).
   

2. 2.To use your images in Lightroom, you import them.  This doesn’t move them or make new copies of them, it simply creates database records so Lightroom knows where they are on your hard drive and what type of files they are.
   

3. 3.As you do things to your images remember all you are doing is modifying metatada .. you are providing new instructions for use when rendering a viewable image from your RAW data.  You can’t change or damage your original RAW data.
   

4. 4.If you are working with native RAW files, (such as .CR2 files or .NEF files) the file itself contains the metadata created by the camera at capture.  This metadata cannot be modified ... changes you make to the metadata are stored by Lightroom or saved to what is known as a .xmp sidecar file.  So the changes you make can override the original metadata, but the file itself can always be completely reverted to it’s original state as captured by the camera.  However, other file types, such as .jpg and .DNG files store the metadata in the header of the file, and standard metadata changes are saved there.
   

5. 5.Even though the heart and soul of Lightroom is a powerful RAW processing engine, you can manage all of your image files with it such as jpegs.
   

6. 6.Presets are one of Lightrooms most powerful features ... you can create a preset for nearly everything you do so you can apply the same action to other files later on.
   

7. 7.The history pallet is amazing ... unlike Photoshop, the history is part of Lightrooms database so even when restarting the program you can see every modification you have made to a file in the past, and move backwards to any point you would like.
   

8. 8.Last, don’t overlook Lightrooms other functions.  It’s printing module is really cool, and makes printing things like packages very easy.  As an example of the web module, this gallery page was created using only Lightroom and a 3rd party web module plugin  ... and Lightroom uploaded it directly to the web server using FTP ... all I did was pick the pictures, make sure my metadata contains titles and descriptions, setup a some options, and click upload.
   

Lightroom is an outstanding program.  You will probably find you need Adobe Photoshop far less often ... most say their time in Photoshop drops as much as 50-90% once they start using Lightroom.  Since it is designed to maximize a RAW workflow, if you aren’t shooting RAW but wanting to use Lightroom, I recommend you switch to shooting RAW.