exposures

At some point nearly all owners of digital SLR cameras (DSLRs) will have to face up to the question of how to organise their “workflow”, amateurs and professionals alike.

By workflow, I mean a systematic, planned and repeatable pathway for transitioning the image data captured in the camera into usable images and safeguarding them in organised, secure long term storage.

That may sound a little overelaborate for a mere enthusiast, but it follows naturally from the decisions you have to make about what you will do with your images once they are out of the camera. Are you going to work with RAW image files or JPEGs? What tools are you going to use to make adjustments, crop, retouch? How are you going to organise your photos on your computer so you always know where to find images, and know which ones are finalised, unprocessed or rejected? Which formats will you use for long term storage of your final images? What are you going to do about backing up your images?

Even if you only address this at a simplistic level it leads to a form of workflow. There has to be some pattern to the way you organise what you do with your photos or you will end up with utter chaos.

The term Workflow tends to be associated with the way a professional photographer would organise their work. Obviously, when it’s your living you have to be able to get from raw images to prints in the client’s hands both quickly and efficiently. You must be able to find and identify images instantly and reliably, and you can’t afford to let anything get lost. The issue for amateurs (thinking particularly of DSLR-toting enthusiasts) is that most of the reasons why professionals establish workflows are just as applicable to them.

It’s more important to be organised now than in the days of film for two reasons:

1. You are (or should be) taking lots more pictures than you ever did. The cost of the medium (film) is no longer a constraint.
2. The “digital darkroom” is quicker, easier to use and more readily available than the chemical darkroom of yore. I’m thinking here of “digital development” applications such as Adobe Lightroom and Photoshop. For most people that means the norm will be significantly more processing steps twixt shutter-click and print.

One of the biggest decisions to make is to settle on a file format for permanent storage of processed images, particularly if you work with RAW images as I imagine many DSLR owners will do.

In an ideal world you’d want your permanent record of your final images to be:

(1) LOSSLESS

That is, your final image is preserved with no loss of quality.

(2) USABLE

If you want a duplicate print or belatedly put a photo on the web, you want to be able to do this quickly without too much work. So if you decided only to keep your RAW files, say, you’d have some work to do to recreate a print and might not be able to reproduce your exposure, white balance and other adjustments exactly unless you kept a separate record somewhere.

(3) COMPACT

You will in general be storing your images on your computer hard disk and disk space is not limitless. Have you noticed the size of a TIFF file created from a RAW image, particularly if your DSLR has a resolution of 12MP or 14MP? If you back up your photos on optical media or in the “cloud” it will be costing you that much more if your file sizes are large.

(4) FUTURE PROOF

If the file format becomes obsolete your old photos may become unusable, or you could find yourself having to convert large numbers of old photos to newer formats, time and time again.

And the nominees are …

(a) JPEG

The trusty ubiquitous JPEG. Universally recognised and the mainstay of photos on the web. Doesn’t score all that well on LOSSLESS. It is a lossy compressed format, given to compression artifacts. It only records colour information to 8 bits whereas most DSLRs output colour data to 12 bits of resolution so there is always some loss of colour precision and colour gradation. Therefore, if you wanted to go back later and make a very large print, you would no longer be able to get the same quality as if you had kept the RAW file.

Obviously the JPEG is eminently USABLE. You can display it directly on the web or print using any number of applications which support it. It also does well in the COMPACT stakes, thanks to the compression. A file size of 2-3Mb can be expected starting with a 10MP image. It is also as FUTURE PROOF as anything else. I cannot imagine JPEGs losing browser support for decades to come.

The same observations apply to similar formats such as PNGs, which I am lumping in with JPEGs for the purposes of this analysis.

(b) TIFF

TIFFs can hold 16 bits of colour resolution and compression (if used) is lossless so you can use them to store your final processed photos with no loss of quality, satisfying the LOSSLESS objective.

Not quite as USABLE as JPEGs, because TIFFs are not displayable in browsers, but there are plenty of applications which can print TIFFs directly and it is not hard to convert a TIFF to a JPEG or PNG as needed for web display.

The biggest problem is that TIFFS are decidedly not COMPACT. A TIFF will be considerably larger than the corresponding RAW file because it records the separate red, green and blue components for each pixel emerging from the “demosaicing” process whereas RAW files hold the bare Bayer array data which only records data for one colour per pixel. That inflates the file size by a factor of three, and the fact that it records colours to 16 bits not 12 inflates the file size by a further one-third. Such compression as available is lossless and might or might not have a significant impact on the final file size, depending on the characteristics of the image. Uncompressed 16 bit TIFFs made from high resolution RAW files are massive beasts indeed, commonly 60Mb or larger. They are though probably as FUTURE PROOF as JPEGs.

(c) RAW

The RAW file is by definition LOSSLESS. The biggest drawback is that it is not directly USABLE. If you only kept the RAW file then you would need to repeat your digital darkroom work before you could obtain a print or recreate a JPEG for web use. If duplicating a print it might not come out exactly the same the second time unless you kept a very detailed separate record of exposure, colour balance, other retouching. In most cases impractical. Keeping the RAW plus corresponding JPEG of the final image would help with usability.

RAW files are reasonably COMPACT at around 10Mb for a 10MP image. Obviously a lot bigger than JPEGs but still manageable by today’s standards, and considerably smaller than TIFFs.

Not necessarily FUTURE PROOF. RAW formats are proprietary, varying from manufacturer to manufacturer, and manufacturers bring out new ones from time to time as the capabilities of their camera models change. You cannot assume that the version of Photoshop released in 10 years’ time will recognise a RAW format no longer being used by a given camera manufacturer for current models.

(d) DNG

DNG (Digital NeGative) is Adobe’s attempt to establish a universal RAW format, to tackle the concerns about future proofing. A very few cameras do use DNG as their actual RAW format and you can convert proprietary RAW files to DNG using say Adobe Photoshop with no loss of quality.

The problem is that the market has not bought into DNG wholesale, at least not yet. DNG is becoming more widely accepted but very much at a snail’s pace, while manufacturers continue to bring out one new proprietary RAW format after another. Unless DNG can become properly established then it is at risk of falling by the wayside so more hindrance than help in the future-proofing stakes.

I think this is because DNG is Adobe’s which makes it just as proprietary as Canon’s or Nikon’s formats. Manufacturers might be more accepting of a standard introduced by an industry group, like JPEG.

(e) RAW + XMP

The XMP file is the “sidecar” file created automatically by Adobe applications such as Lightroom and Adobe Camera Raw to record the sequence of adjustments you make to a RAW image. Lightroom, for example, makes no changes to the RAW file as you adjust exposure, sharpness, colour balance, cropping etc. It just saves a text record of what you do in an accompanying XMP file. Whenever you reopen Lightroom and select an image the changes you made previously are retrieved from the XMP and reapplied to the intact RAW image file, and the processed image can readily be printed or exported to TIFF/JPEG/PNG, etc.

This is a great solution in terms of LOSSLESSness, COMPACTness and USABILITY. There is a big question mark though over FUTURE PROOFing, both because RAWs are not future proof and because you have introduced a dependency on Adobe software.

I am currently going with option (e) despite the future-proofing issue, accepting there is no perfect solution at the moment.