Twenty-One-Mega-Pixels!!!

Canon's new baby is a "Mark III" sporting twenty-one mega-pixels. The "1D" series of cameras has been a very successful "flagship" for them and they have done a great job with their marketing. I know numerous pros that just "gotta have one" whenever they release a new model.

It is hard for me to get enthusiastic about an endless parade of pixels when the "newest" and "best" camera a company produces requires noise filtering to suppress the result of cramming smaller pixels in the same space that was used previously just to raise the count and make photographers foam at the mouth. I really question the wisdom in continually adding pixels when you then need to refine the image filtering technology used to smear the data, hiding the noise those pixels create.

This is a lot different than adding smaller "dots" in an image to make finer gradations in a print, like the RET (resolution enhancement technology) used in many printers. Adding more pixels to camera sensors is counter-productive if they create so much noise that engineers have to use image softening technology to hide it. So, I have to pose the question: Wouldn't it be wise to produce more efficient photo-sites and keep fewer of them on your CMOS sensors rather than raise the count then refine your blurring technology to hide the noise you are creating by increasing your pixel count?

Nikon once said that they would never use full-frame sensors, but they were finally forced to follow Canon's lead... because some professional photographers were thinking about jumping ship and buying Canon, just to get a full-frame sensor camera, even if it meant that they would have to buy all new lenses. That shows just how successful Canon's marketing team was in selling the idea that a real professional needs full-frame sensors to capture the quality of images necessary to stay competitive.

You know, the saddest thing about all of this is that the pixel-count madness has influenced consumers even more. Little point and shoot digi-cams are being produced NOW with as many as twelve-mega-pixels on a sensor the size of your little fingernail. Anything above ISO 400 is totally useless, yet many of these cameras use a form of electronic image stabilization that raises the ISO to as high as 3200 in order to counter camera shake. The resulting images are so poor that printing anything above a 4x6 is totally useless. That is the very reason the OEMs are adding pixels to their sensors anyway. They advertise that you can print up to 16x20 with some of these cameras and it is pure folly.

Here's my suggestion: Why not make a digital point and shoot camera with an APS-sized sensor of say six-mega-pixels and leave it at that. Then, concentrate on exposure control and scene modes. Make the cameras more energy efficient and give the end-user a large assortment of accessories to buy.

With DSLR cameras, we have more than enough pixels to do the job. We really don't need more than twelve-mega-pixels, actually. So, why not innovate in other ways and concentrate on better lens technology? Why not work on better exposure control and faster processing speeds? Why not make better LCDs that are viewable from any angle and in bright sunlight?

Adding more pixels just to blur the image to mask noise makes no sense at all.

---- Mickey Maguire