Digital Camera File Formats

When you take a photograph using a film camera the image is recorded onto film, and the film used also becomes the storage medium, the repository, where the image is permanently kept. It might be slide film or negative film, but whichever it is there is usually only the one image recorded and saved for each shot taken. So keeping this safe is vitally important. In the case of print film a print is usually made, and this is what most people look at and tend to keep safe, stored in print albums. Sometimes the two are kept together, but often not, and it is commonly the case that sooner or later the negatives are discarded or lost. Yet the negative is the more important since it contains the information needed to make another print. Copies of prints are never as good as those made from the original negative. Keeping slides safe is even more important since there is no print that exists as a secondary backup.

In contrast to film capture when you take an image with a digital camera the image recorded is saved to a file, an image file, that can be read by equipment that is capable of doing so, and displaying the information it reads as an image. This information is in binary code. Viewing a digital image file without equipment that can read and decipher the binary information is not possible. And electrical power of some kind is needed, either battery or mains. Unlike film it can't be read - viewed - just by sight with the aid of light. And the equipment must be capable of reading the particular image file format used, because there are a number of different ones that are used, for different purposes. This is where, in the opinion of many, the biggest dangers with digital imaging lay.

But it also has it's advantages. The image file can be duplicated as many times as is needed without the slightest degradation of the information it contains. Each copy being as good as another. So copies can be made and stored safely away in various locations on different types of media.

We'll take a look at digital camera image file formats and discuss what the pro's and con's with each are. You might also like to read our page Storing Photo's  on preserving image files for long term storage, access, and future proof readability

File 'bit' rates.

Although we think of digital cameras as digital devices, that is only really applicable to their output. Their sensors are actually analogue devices. They send out a voltage or current which is proportional to the intensity of the light that falls on them. This analogue signal is converted into a digital signal by an analogue to digital converter, known as A/D, and the method is known as quantisation. The signal that the A/D outputs is digital 'bits', binary code.

In computer terms, a "bit" (binary digit) is the smallest piece of information and has a value of either "0" or "1" which actually corresponds to one of the millions of transistors (switches) inside the computer being 'off' or 'on'.

There are 1024 bits in a 'byte'. 1024 bytes in a megabyte - Mb. And 1024 Mb in a Gigabyte - Gb. There is also a Terabyte - Tb, which is 1024Gb. And it won't be too long before computer hard drive capacities reach this volume the way things are progressing. We are referring here to single hard drive capacities, which are already up to the 400Gb mark.  (November 2007 - 1Tb hard drives announced)

In a 1 bit image file the binary value "0" is assigned to black and "1" to white.  So just two tones, black and white.

A 2 bit image can have 2^2 = 4 tones: 00(black), 01(grey), 10(grey), and 11(white). 

A 8 bit image can have 2^8 = 256 tones ranging from 00000000 (0) to 11111111 (255). 

A 12bit image can have 2^12 = 4096 tones.

A 16bit image can have 2^16 = 65536 tones.

The most common file bit rate is 8bit. Because this bit value only refers to a single channel, and colour images have three channels, Red, Green and Blue, 8bit files are often referred to as 24 bit images as the information is 3x8bit.

Jpeg file format images are often referred to as 24bit images because they can store up to 8bits in each of the 3 color channels and therefore allow for 256 x 256 x 256 = 16.7 million colors. This is known in computer terms as 'true colour'. You'll probably recognize this from your computer's monitor graphics settings. 

When discussing 8bit tonal values along the 0-255 scale, a particular tone is often expressed as 0/255 - pure black, 255/255 - pure white, 128/255 - the mid-point tone. This is much less confusing, and much more practical way of reference, than trying to refer to them in binary code terms, which few could probably understand and work out in a reasonable time scale if at all. We're not sure we could.

File bit rates don't have a theoretical limit, but in practical terms they do, for the bigger the bit rate the more information there is, and the larger the file size. So the longer it therefore takes to open, manipulate and process such files, and the bigger storage space they require.

Some file formats can be saved as 8bit or 16bit. There is much opinion given as to the image quality superiority of 16bit files, which are capable of producing not 256 tonal levels but 2^16 - 65536 levels. And thus 65536 x 65536 x - 65536 colours - 281 trillion colours. 

This is a subject we will deal with further on.

File formats 

Although there are a number of graphic file formats used, there are just three digital camera image file formats that will be encountered at present when using a digital camera. JPEG, TIFF, and RAW. However whilst the first two are common file formats the last is not. Or should we say it does not refer to a particular file format, but rather a generic type. A native or propriety file format. 

JPEG

Jpeg is the universal file format used to store and exchange image files in a space efficient way, by using compression. It was devised by the Joint Photographic Experts Group, from whence the acronym comes, and it's currently the main file format used by all digital cameras to save the images that are taken. The amount of compression used can be varied, but the larger the compression used, giving smaller file sizes, the larger the loss of image quality. This usually forms the basis of digital camera image quality settings. Images can be taken in a variety of sizes up to the maximum sensor pixel rating, and at a range of compression settings, often good, better, best, or a star rating */**/***

Jpeg is what is know as a 'Lossy' format. It compresses image files to reduce their size, so they take up less storage space, and it does so by using advanced algorithms to examine and discard information that it considers is not needed. Duplicate information. This is done when a file is first 'saved' as a Jpeg, in this case in-camera. When you open a Jpeg file it is un-compressed, and when you close it after viewing it is re-compressed. These actions do not alter the file information which does not change. No matter how many times you do this. Nor does this alter if you move the file or make a copy, duplicate or print it. But it does if you open an image, alter it, and then save the changes made. And this has a cumulative effect. So continually opening Jpeg files, making changes to the image, and then saving these changes is not wise, as it eventually leads to a file with poor image quality.

Jpeg is an 8bit file format. Saving files in the lowest compression and highest quality has been proved to provide image files that are equally as good as those from Tiff files, with few artifacts or aberrations. Given the universal use of Jpeg files, they can be read by virtually any digital file device, from the ordinary computer to the cheapest DVD players and memory card reading image display devices, they remain the first choice for most users.

TIFF

Tiff files can be 8bit or 16bit. Tiff is the acronym for Tagged Image File Format. This is an image format known as bitmap. Although mainly referred to a high quality non-compression format, the file size remaining large and un-compressed, Tiff files can actually be compressed should the need arise. The most well know compression formats used are ZIP and LZW. However these are really only best used when large areas of plain colour are involved, so it is best not to compress Tiff photographic files. Because no compression takes place these files can be altered and changes saved as often as you wish, no degradation of the image takes place, save those that might occur as a result of the particular changes made, depending on what they are.

Because of their reputation for preserving high image quality this file format was once quite common in advanced digital cameras. However the size of the files, and the time and power requirements needed to process them has led to these disappearing from most cameras, in favour of the RAW formats. A considerable number of digital image users archive their image files in this format, some using the 16bit format. This requires a lot of space.

RAW

RAW does not describe a file format, but rather a file type, so it's is not a acronym, but refers to the information provided straight from the sensor allegedly in it's native form, raw and un-processed, although this is now accepted as not being quite true. These files are known as native or propriety file formats as they are only used by the hardware or software that generates them. As such, each camera make has it's own RAW format, and no two are alike, either in the basic information contained, the amount of compression applied, or whether how much of it is encrypted, a little 'bonus' some camera makers are including in an attempt to prevent any software but their own from decoding it. However, whilst files can be saved in the Raw file formats, once they are processed they have to be saved in another file format, Tiff or Jpeg being the most common. Those wishing to achieve the highest possible image quality take images in the RAW file format and then save them as 16bit Tiff's.

Because each camera uses different RAW formats, and the files have to be processed before they can be used, most digital cameras that produce RAW files come with RAW converter software. The complexity and quality of this software varies greatly, as does the image quality that is produced, and this rather defeats the object. As a result, many independent RAW conversion software applications have emerged. Some of these work very well, but it has also become clear that whilst one may produce high quality images with the RAW files from one camera make, it may not do the same with those from another, since no two are the same. Adobe, famous for the Photoshop image editing software, and much more besides, has developed it's own RAW file called DNG, which it is promoting as an 'open' format and trying to persuade all to use. As well as this it has it's own RAW converter called Adobe Camera RAW - ACR - which is proving to be perhaps the best all round RAW conversion software having equal results with any RAW file format. This converter is provided free as a plug-in for Photoshop and Elements, and regularly updated for the latest cameras, as the camera makers are continually altering their native RAW file formats.

The appeal the RAW file formats have for many is that it enables shots that turn out poor to stand a greater chance of being 'rescued' by altering the processing applied. Adjusting the white balance, the sensitivity, the contrast, all those parts that would normally be processed in-camera at default settings. The drawback with RAW files is that because of this they cannot be used as they stand. Printed straight out from a memory card, read by any ordinary software program, or reduced in size for e-mailing. They must be processed and converted first, and this requires the use of a computer. And although they are often compressed in size compared to Jpeg's they still take up a lot of room on a memory card. And of course then you end up with two files for every shot taken, the raw file, and it's processed counterpart.

Raw files are generally saved at the bit rate that their A/D converter uses, normally anywhere between 10/14 bits A/D.

Files formats - image editing and printing.

Whilst there are many that save their images in 16bit file formats, the number of applications that can open and manipulate them is small but growing. Most can only work with 8bit files due to the size and amount of information contained. Some dedicated image editing applications can, and you can carry out basic tasks such as printing and rotating. But generally for more complicated manipulation they have to be re-sampled down to 8bit size, just as most other applications have to do just to open them. This rather defeats the point of 16bit files, as image information is then lost, and the files are no better than straight 8bit files. Another problem is the computing power needed. Even if an application can open 16bit files it needs a very powerful machine to be able to work with the images in a reasonable timeframe.

However there is another catch that most users are unaware of. Most normal commercial and consumer printers are unable to deal with anything other than 8bit files. What happens when you send a larger file bit size to print is that it is down-sampled by the printer software and the O/s to 8bit size. There is no indication this is going on, it happens in the background. The only outward sign is that a 16bit file will take considerably longer to print, from the time it is sent to print to the finished print appearing, in comparison to a 8bit file, because of the extra processing taking place. And so improved image quality does not result. The only time that using 16bit files has an advantage is with the use of outside agencies. Stock photo, or magazine printing, where the file size may make a difference. Other than that there is no good reason for most to use 16bit files.

It is important to realize that the size of an image file, both when stored and when opened relies not only on the size of the sensor pixel count of the camera used to take it, but the file type it is taken and stored in subsequently. To give an example a 6mp file is the same size when opened whether it is 8bit Jpeg or 8bit TIFF, only when the file is closed does the size alter, the Jpeg requiring much less storage space, being compressed, than the TIFF. But there is a large size difference between an 8bit TIFF, and a 16bit TIFF, which is twice the size. It doesn't result in a larger file in the sense that it can be printed larger, the number of pixels remains constant, there is just more tonal and colour information.

Future accessibility

The biggest danger with regard to using any file format is not really short term use, important though this is, but long term accessibility. There is no guarantee that the software that can currently read and open these file formats will be around and compatible with future computer Operating systems. Or that future software will be able to read them. This is already a problem with some older computer file formats used in the past, and not obscure and little used one's either but those that in their heyday were common and heavily used.

Although at present it would seem highly unlikely that this would happen with the Jpeg or TIFF formats since they are continuing in use and not being superceded by others this is not the same for the RAW formats, which are constantly being revised and upgraded. This is another factor to consider when choosing which format to use.