How many megapixels do you need – really?

When the first DSLR cameras were released at the end of the 20th century, photographers were more obsessed with the megapixel counts of cameras than virtually any other imaging parameter. Back in 1999, the Nikon D1 had an effective resolution of three megapixels, which was commonly seen as ‘professional quality’.

It took a long time for sensor resolution to ramp up and even by September 2001, the highest resolution offered by Canon’s top DSLR was four megapixels, while the Nikon D1X provided five megapixels. Incidentally, both these cameras had APS-C sized sensor chips, another factor that has changed over time, along with other imaging capabilities (notably sensitivity ranges, video recording and continuous shooting).

The prices asked haven’t changed all that much ““ but buyers get a lot more for their bucks!

The arrival of DSLRs with ‘full frame’ (36 x 24mm) sensors in 2002 should have pleased plenty of potential buyers, particularly as sensor resolution also increased to 11 megapixels (or a little higher). By late 2014, resolutions in excess of 20 megapixels were considered normal, with the Nikon D810 offering 36.3 megapixels in a semi-professional camera body.

With the launch of its first Micro Four Thirds (M4/3) system camera in 2009 an Olympus executive apparently claimed photographers only required 12 megapixels. But by 2012, when the (recently discontinued) OM-D E-M5 was launched, Olympus had joined Panasonic, which introduced a 16-megapixel sensor in the previous year. Both companies have standardised on 16 megapixels, at least for the time being for the 17.3 x 13.0 mm sensor size.

Other CSC manufacturers have gone higher, with Sony offering 24 megapixels in its APS-C chips and 36 megapixels in ‘full frame’, Samsung providing 20 megapixels and Canon at 18 megapixels for the APS-C sensor size. Even Nikon is offering 18 megapixels with its Nikon 1 system based on a 1-inch type (12.8 x 9.6 mm) sensor.

Panasonic’s Lumix GM5 camera provides the advantage of an electronic viewfinder in a pocketable camera body. Its 16-megapixel resolution is enough for making A3-sized prints.

Practical Considerations

With such a range of sensor sizes and megapixel counts to choose from, how do you know what you really need? Essentially, it depends upon what you do with your pictures. And that will be influenced by the type of photographer you are.

Casual snapshooters who only use their photos for sharing on the internet should be much less concerned about the megapixel counts of their cameras than serious photo enthusiasts, provided they got the shot they were after. This is one reason why an increasing number of snapshooters have turned to their phones for most of the pictures they take, which has seen the decimation of the point-and-shoot camera market in recent years.

Very few photographers in this category would even be able to see any differences between JPEG images recorded with a 20+ megapixel camera and 8-megapixel shots from a smart-phone. It takes a trained eye to know what to look for, particularly when images are viewed online. This has been repeatedly confirmed by tests and surveys.

It also underpins the rationale for JPEG compression and the sRGB colour space. Both are based upon providing the minimum amount of information needed for an image that looks acceptable to most people across a wide range of products and viewing platforms.

Even if they print their best shots, snapshooters will generally opt for snapshot sized   (15 x 10 cm) output. At a pinch, such images could also be incorporated into smaller photo books, particularly if production is outsourced to a capable supplier with good resizing facilities.

Nikon has standardised upon 24-megapixel resolution for its entry-level DSLRs, which is probably much more than most potential users require.

Serious photographers fall into a different category and are more likely to be able to discern differences between shots from a DSLR or high-resolution CS and shots from a smart-phone. More experienced shooters might also be able to see differences between shots from a 16 megapixel M4/3 camera and a full frame camera with 20 megapixels or more.

When shooting in low light levels at ISO 6400 or higher, more pixels won’t be as useful as ‘better pixels’ delivered by larger photosites. The same is true for dynamic range. Nikon’s 36.3 megapixel D810 can’t encompass as wide a dynamic range as the 16.2 megapixel D4S, even though both cameras have ‘full frame’ sensors.

Assuming you have appropriate lenses, understand how to use your camera’s exposure controls and capture and process raw files effectively, a 16 megapixel M4/3 sensor will enable you to print your photos at 300 ppi (pixels/inch) on an A3 sheet of paper without resampling or on an A3+ sheet with minimal resampling. And if your resizing techniques are first-rate, you could even go up to A2 size (42.0 x 59.4cm). In each case, the image would be surrounded by a white border of suitable width.

What would you get from more pixels? Larger print sizes and maybe greater edge sharpness at the same output size. You would also have greater scope for cropping images, although that should never be a motivator for a serious photographer.

Heavy cropping is a sign of a rank amateur who is too lazy to compose shots in the viewfinder. These shooters often use wider-than-optimal focal lengths and then attempt to correct compositions by cropping away parts of the subject they don’t want. But reducing a 4608 x 3456 pixel image to 2400 x 1800 pixels actually discards three quarters of the pixels, which reduces image quality.

Quality-conscious photographers will have quite different priorities. Most will capture raw files and place much more importance on having capable equipment, including appropriate lenses for the types of photos they like to shoot. They will only resort to cropping when they need to change the aspect ratio of the picture or when they can’t be in the right position with the right lens for the shot. And it will be done judiciously.

Landscape photographers shooting for larger output sizes than A3+ with lower resolution cameras can always stitch several images together or use image stacking to encompass subjects with a wide dynamic range that can’t be covered by shooting in the raw file format. Either option has more potential for producing optimal image quality than just having more pixels on the existing sensor, regardless of the sensor format.

With an effective resolution of 36.4 megapixels, Sony’s ?7R is best suited to experienced photographers who have a genuine need for high-resolution files. Its high price tag puts it squarely into the ‘professional’ category.

Large Sensor Benefits

Although cameras with M4/3 and smaller sensors can offer many of the capabilities required by serious shooters, they have some noteworthy limitations. Smaller sensors tend to limit the size of the individual photodiodes used for capturing light and creating image pixels. For example, on a 16 megapixel M4/3 sensor each photosite is a little less than four microns square.

By comparison, on a 16 megapixel ‘full frame sensor, photosites are almost eight four microns square, which is four times the area of the photosites on the M4/3 sensor. (For interest, photosites on a 16-megapixel APS-C sensor are just over five microns square, while those on a 1-inch type chip are less than three microns square.)

Larger photosites are more able to capture photons of light and this has quantifiable benefits. Image noise is reduced, often noticeably, as sensor size increases. The dynamic range the sensor can record may also increase. It can be easier to develop efficient autofocusing systems for larger sensors, particularly when phase-detection is available.

Many professional photographers are using the 16-megapixel Olympus OM-D E-M1 when they need a capable, lightweight camera system with a wide choice of high-performance lenses