Digital Camera and Photography: Tips, Information and News | Neocamera Blog

Just after Sigma announced their 14 megapixels SD14, people complained about the way megapixels are counted, again. In fact, this is not a new complaint, people complained about the Fuji Finepix F700, the predecessor of the current Fuji Finepix S5 Pro.

The truth is that Sigma is just as wrong as every other digital camera manufacturer. A pixel is short for pixel-element and should have full-color information. On standard computer monitor, a pixel is composed of 3 sub-pixels. Each sub-pixel being one color component. Therefore, a 3 megapixels monitor has roughly 3 million pixels or 9 million sub-pixels.

When most digital camera makers measure megapixels, they count the locations on the sensor which individually accumulate photons. An accurate term for these is photosites. A standard sensor is monochrome by nature, so a Bayer filter is placed over it. This turns each photosite into a sub-pixel. Therefore a camera normally labelled as having 6 megapixels actually has 6 mega-sub-pixels.

The Sigma SD14 is different because it uses a Foveon sensor which does not use a Bayer filter. For that reason, Sigma multiplies the number of photosites by 3. After all, everyone else counts sub-pixels as pixels, and their photosites are therefore worth 3 pixels! The truth is that the SD14 has 14 million thirds-of-pixels. And a standard 10 megapixels DSLR has 10 million sub-pixels. The different is simply the spatial arangement of the sub-pixels. In the case of a standard sensor, each sub-pixel is at a different location relative to the image. In the case of a Foveon sensor, there are 3 sub-pixels at each physical location.

As for Fuji, their SuperCCD SR uses two photosites per pixel and a Bayer filter with each color covering one pixel and therefore two photosites. The two photosites are used to capture information for a single sub-pixel but with a greater dynamic range because the pair have different light sensitivities. Originally, Fuji used their first SuperCCD SR in the Finepix F700 which was labelled as a 6 megapixels digital camera. In fact, it had 6 million photosites. These photosites were combined into 3 million sub-pixels which were used to interpolate an image. Later Fuji introduced the S3 Pro with the same type of sensor and labelled it as 12 megapixels. Recently, with the announcement of the S5 Pro, Fuji refined its terminology to clarify that the S5 Pro has 12 million photosites which are combined into 6 million pixels (which are in fact sub-pixels). Its not entirely correct but its no more wrong than Sigma’s description of the SD14.

Neocamera Blog © Cybernium.

Most entry level DSLR cameras optionally come with a disposable lens. It is called the kit lens. It is generally bundled for under $100 more than the camera itself. Invariably, it is a zoom lens made of low-quality optics designed to get the new buyer started with his digital camera right away.

Notice how there are barely any new lenses which can be bought for under $150? That is because even mid-quality optics cost more. A great exception is the Canon 50mm F1.8 II lens. Since DSLR-kits are designed for entry-level users, cost is very important. To keep within those costs, camera makers generally choose poor quality optics. The exception here comes from Panasonic which uses a 14-50mm/F2.8-3.5 lens with its Lumix DMC-L1, producing one of the most expensive DSLR-kit available.

Even if the image quality of a kit-lens is acceptable, which depends on the medium and the user’s eye for detail, these are generally not versatile lenses. The minimal kit lens is an 18-55mm lens for cameras with a 1.5X or 1.6X focal-length-multiplier. A few camera makers offer an 18-70mm lens instead. Here Nikon stands out with a 18-135mm lens bundled with some of its latest DSLR cameras.

Generally, not only do these lenses have a short zoom ratio, they also have narrow maximum apertures. The Sony 18-70, for example, is quoted at 18-70 F3.5-5.6. In practice, the maximum aperture drops very fast: F4 around 20mm, F4.5 around 24 and F5.6 already at 35mm. This means that more light is quickly required as you zoom in. It also causes the camera to focus more slowly, particularly near the telephoto end of its zoom range.

The conclusion is simply that kit-lenses are disposable. Even if a kit-lens serves a temporary purpose for someone who cannot afford something better, a replacement should eventually takes its place. If the short zoom range is sufficient, a much higher-quality lenses can replace it. If the zoom range is too limited, a long-zoom ratio lens can be purchased instead. Perhaps the latter option won’t increase image quality, but it will improve versatility. There is always the option of getting several high-quality lenses to cover a wide range of focal-lengths to achieve quality and versatility. In other words, it is easy to do better, but nearly impossible to do worse.

Don’t confuse the kit-lens with higher-quality bundles offered by some manufacturers. The immediate way to notice the difference is price: those bundles cost quite a bit more than the camera itself. Also, such bundles come with one or two lenses which are readily available for purchasing separately from the DSLR.

Neocamera Blog © Cybernium.

Car advertisers often use their Formula One credentials as a justification for their cars being superior. The implied statement is that they make high-performance Formula One race cars and therefore they make high-performance street cars. The actual implication is slightly different: they know how to make high-performance race cars and therefore they know how to make some parts of high-performance street cars. The important difference lies between the words make and know. The fact that a corporation knows how to do something does not mean they do it. Why don’t they? The answer is mostly a matter of cost.

Lets make a parallel for the digital camera world: Suppose that a company, lets call it BCB for Best Camera Brand, produces a digital camera superior in every way to every other digital camera on the market. To produce that camera, the BCB-1, they put billions into a decade of research. To recover those costs and cover expensive materials, the BCB-1 retails for $500,00 USD. Now, suppose that after selling these cameras to everyone who can afford it, BCB is still at a great loss. At that point, they decide to produce the BCB-10, a lower-cost camera with a price-point under $1000 USD. How good would the BCB-10 be? Would it be superior to every other camera in the market except the BCB-1. It all depends on what corners they cut to bring the price down. The important point is that no one can tell without actually testing the camera and that the superiority of the BCB-1 does not imply the superiority of the BCB-10.

Frequently people ask which is the best brand of digital cameras. There are numerous reasons why this cannot be answered. One of which is that best is subjective and depends on the photographer, his subjects, his medium and practical reasons. Another one comes down to the Formula One Fallacy. Even if a brand had the greatest camera of all, there is no reason to believe they all their cameras are good. So, even if the question had an answer, it would not apply for anything else than buying the best camera of the best brand.

Lets pick on Nikon a little. NIkon is a strong brand with some exceptional products. Many of its cameras, like the Nikon D200 and more recently the D80, have gotten praise for their performance and feature set. Indeed, these are excellent cameras. Price-wise though, Nikon tends to price its cameras above its direct competitors. That is alright for the high-end market where photographers appreciate the subtleties of a camera’s performance. In the low-end market, price is king. To give an attractive pricing to its low-end cameras, Nikon ends up cutting more corners than most of its rivals. Presumably this is done to keep in-line with Nikon’s expected profit margins. The Coolpix range of Nikon digital cameras shows this, specially compared with A-series Powershot cameras which almost all feature full-manual controls, ISO control, manual white-balance and more. The bottom line is not to judge cameras by its brand’s flagship model.

Neocamera Blog © Cybernium.