The first in a short series on getting the most out of digital images
I went to a photographers meeting a couple of weeks ago where pictures taken by around 20 different photographers (including myself) were all projected. Virtually all of them were good pictures – and most of the photographers were managing to make a living from photography, but technically I felt a number were letting the people who took them down in various fairly simple ways.
Back in the darkroom age, most photographers either learnt to make at least halfway decent black and whites or, more often once they started to make some money, had their work printed for them by a decent printer. With colour, for publication we largely shot on slide film and accepted the results from the lab on our Ektachrome or whatever, handing the trannies in for publication.
Now, most publications will no longer accept prints or slides and expect digital, but there are still quite a few photographers who haven’t yet really learnt how to get the most of of what they shoot on digital.
In most respects, digital has had the result of increasing technical standards compared to film. Certainly in low light we can produce pictures with a technical quality that is way better than film. But where we used to rely on labs to get things right, we now have to do it ourselves. So here is the first part in a series with some simple advice for photographers.
Monitor
Probably you’ve stood at some time in an electrical shop looking at the same programme displayed on a whole range of different TV screens and noticed the differences in colour, brightness and contrast between them. All of them are getting the same input signal, but may produce very different pictures. Computer screens are not a lot different (especially as we move to digital TV) and your image file may display very differently on your own screen to than when viewed by an editor or projected onto a screen. Those pictures with a nasty yellow cast may have looked perfect on someone’s notebook screen.
If you work with images it is worth buying a good screen to display your pictures. Unfortunately few if any notebooks come with particularly good screens – and a really decent screen for editing your pictures can easily cost as much as a notebook.
At home I use an Eizo monitor. It cost about as much as the large box by my feet that sends it a signal, but was worth every penny. Away from home I use an Acer notebook, and the images don’t quite look the same.
Hardware Calibration
Although the notebook screen hasn’t got anything like the picture quality of the expensive Eizo (or the remarkably good ancient but fairly massive Iiyama cathode ray screen I still also use) my pictures display in a reasonably similar fashion on all three screens. And I can be confident that any editor etc who knows what they are doing will also see something pretty similar. My confidence comes from regular calibration with a suitable hardware device, in my case a Pantone (GretagMacbeth) eye-one (i1), now replaced by an improved model.
You can get by without such a device – just as you can get by on a laptop screen – and your results may well be ok, but they will not be optimum. Particularly if you want to make your own prints, the better monitor will make it easier to get predictable results.
If you can’t afford a hardware calibration device, there are some web sites that have useful pages to make a rough check on some aspects of your monitor, and you may well be able to set up your monitor roughly using these and the controls on your monitor. Usually you should set the colour temperature to 6500K and use Gamma 2.2 (both on Mac and PC.)
Camera
If you are serious about photography then you will normally be shooting on a DSLR camera with at least a “4/3” sensor, more probably one roughly half the size of a normal 35mm film frame or roughly the same size. Compact cameras (except the latest Sigma DP1) have smaller sensors, usually smaller than a fingernail. Although some of these perform near miracles, they still cannot compare for resolution and image quality with the larger sensors, and at higher ISO the noise becomes excessive.
RAW vs Jpeg
As well as a decent camera – and of course a decent lens – you also need to learn how to use it. Test after test has shown that every camera that has both RAW and Jpeg modes can produce better images when shooting RAW. Almost always there is greater dynamic range but even more important is the flexibility to alter and enhance images that RAW processing provides. The difference is in some ways similar to the difference between using colour neg and tranny, where the printing process gives you a degree of control over how your images look, while the slide you just accept and put on the light box. You can do some post-processing with jpeg, but it is sometimes at the expense of visible image degradation, while RAW allows much great control without any quality sacrifice.
Sometimes of course speed is essential, and the ability to use jpeg files straight out of the camera is vital. If possible it is a good idea to shoot both RAW – to get the best out of when you have time – and jpeg for immediate use.
Colour Depth
Digital cameras transform the minute analogue electrical signals recorded by sensor cells into digital signals. Most cameras work using 12 or 14 bits for each cell to store the value, enabling them to distinguish either roughly 4,000 or 16,000 levels. In most cameras, each cell is behind a colour filter and these are levels of either red, green or blue light, depending on the filter.
The jpeg format only allows 8 bits for each colour – 256 levels, and when your camera writes a jpeg file it has to decide how to scale the data down to fit into this smaller number of levels. Exactly how it does this will depend on various camera settings for the colour space, contrast, colour temperature etc. Once this process has been carried out, and data discarded it cannot be retrieved. The 8 bits in each of the red, green and blue channels of the image make this ’24 bit colour’, also sometimes called ‘Truecolor’
RAW files – or at least most of them – actually contain a jpeg file. In the Nikon NEF format it is a jpeg using ‘basic’ compression, and some software will quickly extract these for you. The jpg is used for the image you see on the screen on the back of the camera. The file also contains the full 12 or 14 bits of data corresponding the the sensor cells, along with information about the camera settings, including the white balance etc. One difference between different software for processing RAW files is sometimes in how much of this information is used automatically, with the camera maker’s own software sometimes understanding more than other software.
You may also have the option of outputting 16 bit TIFF files, particularly from RAW processing software. These files can essentially contain all of the image data from the raw file, but are padded out to occupy greater space. However unlike RAW files they are understood as image files by a wide range of display software.
Colour Spaces
Cameras can only record a fairly limited part of the visual range, with many colours being out of gamut. Different sensors have their own characteristics, producing different electrical values from the cells from the same scene.
The values for pixels need to be connected in some way to actual colours and two ways of doing this are in common use for camera images, sRGB and AdobeRGB (1998) ICC colour profiles, each covering a different range of the visible colours.
AdobeRGB covers a wider range of colours, and is thus usually the better choice for camera images, but the sRGB range is actually a better match to the colours that most monitors can display, and is thus the normal choice for compact cameras.
· sRGB for web use and on-screen presentations
· AdobeRGB for reproduction
It is generally more sensible to set AdobeRGB as your camera profile and to make this the basis of your workflow, converting files to sRGB if you need these for screen use. The Adobe RGB space has a wider colour range.
When supplying files for any purpose always ensure they are tagged appropriately. Unfortunately much software – including almost all web browsers – doesn’t understand ICC colour profiles and will display Adobe RGB files wrongly.
Print services for amateur use always expect sRGB. For professional printing you should consult the lab, which should also be able to supply you a printer profile you can use for ‘soft-proofing’ your images on screen in Photoshop.
Continued in ‘Better Imaging 2″