How Important are Megapixels?

Whenever I look at a new camera (purely for interest these days), the first stat I normally read is the megapixel count.  There seems to be a lot riding on this one number, as though it somehow conveys the quality of the images you will obtain and the performance of the camera in different conditions.  We’re also taught generally that bigger is better.

Sony recently announced its 61 megapixel flagship.  61 megapixels is surely “better” than the 24 megapixels of my Fuji or the 20 megapixels of my aging but trusty Canon.  On all counts, nothing could be further from the truth.  Here’s why.

First, let’s define a megapixel.  It is, of course, 1,000,000 pixels.  A pixel is the smallest unit of light capture typically associated with a digital image.  Each digital sensor has its photosites (physical units of light capture) arranged in rows, horizontally and vertically.  Depending on the sensor design, each photosite can record up to 3 different colours (R,G,B).  The total number of units of light capture horizontally multiplied by the total number of units vertically, then divided by a million, provides the megapixel rating for a camera.

There is a relationship between megapixels and image size, file size and resolution.  But surprising to me, they are not one to one.  My 24 megapixel Fuji produces a 40-50 megabyte image, 6000 pixels in width and 4000 pixels high.  In fact, it’s quite typical for the file size in megabytes to be approximately two or even three times the number of megapixels.  This is because cameras record three colour channels as already noted, and depending on sensor design, each unit of light capture can contribute up to 3 channels of colour information, making the downloaded raw file size 3 times or more larger in megabytes than the megapixel count.

But even more important is the sensor size on which all of these units of light capture are arranged.  Fewer megapixels does not necessarily mean a smaller sensor, as witnessed by my Fuji’s APS-C sensor at 24 megapixels vs. my Canon’s full frame sensor at 20 megapixels.  The full frame sensor is approximately 50 percent larger than the crop sensor.  This of course means that each light capturing unit is much larger on my Canon.  The more light captured with the same exposure, the better the image quality (less noise, less distortion, less aberration).  This alone proves that more megapixels isn’t always better.

Ok, that’s the technical stuff.  Megapixel counts are typically the first thing published about a new camera – in fact, it’s often in the headline.  We are predisposed to see bigger and bigger as better – somehow it means you are ahead of everyone else.

But what are the advantages of more megapixels?  It seems to me that there are more disadvantages than advantages.  I’ve already talked about one.  By design, more megapixels means more data.  More data means more storage needs both in-camera and out, which means more hardware and more money spent.  Computers have to be more powerful to handle it, software more efficient.

On the flipside, the level of detail increases with more megapixels.  A single shot or a panoramic shot using a camera with a high megapixel count can reveal the most extraordinary detail in the capture.  And this is where the greatest confusion resides.  The level of detail is relevant only in those situations where you need more detail.  Sure, you can stitch a 4 frame 61 megapixel pano together and get more than 200 megapixels.  With that, you can zoom in on your computer and maybe see the people sitting on the balcony on that high-rise in your cityscape.  Or you can print a billboard size print and it will still be a stunning shot to those driving by.

But how many of us do either of these things – or care about either of these things?  My photography club asks for submissions at 1920px x 1080 px, for a total of 2,073,600 px or less than 2 megapixels.  Facebook and Instagram maximum images sizes are less than that.  Even delivering a nice 24 in x 36 in wall-sized print to a client or family member can be done with either of the 20+ megapixel cameras I already have.

And when you do need more detail, your answer is above:  stitch together a series of images at a lower megapixel count to increase the final count and thereby the finish options for that image.

There are professional photographers who are known for their expansive works.  Ed Burtynsky comes to mind.  He is often in situations where more than 200 megapixels are required to produce the works that he puts on display.  But we are NOT Ed.

So the next time a fancy ad yells out the latest megapixel count, skip to the fine print and check out the reviews, not for image detail, but image quality, low light capture, dynamic range and all of the really important stuff in the next generation of camera gear.  Can you carry that gear easily, are the ergonomics what you need, do the menus make sense?  How well does it focus, how quickly does it focus?  All of these are much more important that that shiny really big number (which by the way is also how they charge the shiny, really big price).  Don’t be fooled.