Pixel Density - Does it Matter?

Way down on my list of photo concerns when composing a shot in the field is pixel density.  After all, after the camera purchase I can't adjust it.  It's not part of the "exposure triangle".  In fact, after the advertised pixel count - what is there?  And of course, the related male driven characteristic - more is better.  Right?

Like a good geometry test question, let's start with the givens.  Sensor Size.  Sensor size, by manufacturer, is always the same size.  FX (full) and DX (APS-C) have the same dimensions.  FX = 36x24 mm.  DX = 24X16 mm.  Most everything else is a variable.

Let's get to the pixel packing.  The more cows that are in a corral, the more dense the cows are.  (With due respect to Gary Larson and the Far Side, this is physical density, not mental.)  And in the same way, the more pixels in a sensor, the greater the pixel density.  And does this matter?  Is more really better?  Or just different?

So, with the given information we can conclude that higher the advertised pixel count, the smaller the smaller the pixel and the higher the pixel density.  So now we can equate higher pixel count is a smaller pixel.  Not all pixels are created equally.  Is that a surprise?  Or something you don't think of?  

Does pixel density matter?  Very simply - YES.  Can you do anything about it in the field?  No, other than to understand the concept and be mindful.

We're in the constant search for the sharpest possible image.  There are time when we're looking for blur, but that is an artsy thing.  So let's stay for the sharp thought for now.  A sharp image is an image where points of light exist on more than one pixel.  If a point of light exists on multiple pixels - not so sharp.

And what are the best ways to eliminate blur?  Assuming the blur is motion related and not a focus issue, increase shutter speed and better technique.  

As with most situations where there is more than one element involved, there are trade offs between the elements.  More pixels simply mean a sharper shot is possible.  And if that is true, a larger shot is possible.  But the trade off is you need to do something to keep that one point of light on a single pixel, simple terms - speed.

Is this all going somewhere?  That might be helpful?  Yes.  The higher pixel counts present a number of issues.  Increased files sizes, buffer size issues.  On the plus size, sharper images on a larger end format.  Some folks still swear by their 6 mp camera.  With proper technique and a small end format it will work flawlessly.  But the fact is you will not be able to get a wall sized portrait with 6 mp.  Most enthusiasts will top out at a 11 x14 print, which means you need 3300 x 4400 for a 300 dpi image.  Theoretically, you as an enthusiast could get by with a 14 mp camera.  No cropping.  But, 18, 20 or 24 mp camera - go for it.  For digital imagery, the math get more complicated.  And I don't do that well.  But I use a 27 inch, 1920x1080 monitor.  It was gaming quality when I bought it a few years ago.  These days, pretty average.  Anyway, video drivers do a great job with my cropped shots.  The upshot is, to me quality is best defined at the print level.  Which few of us use any more.

So what's the point of this?  The more pixels, the more chance to blur an image.  Or get a sharper one that the end format will not be able to tell.  Or make a wall print.  Those are pretty indisputable facts.  

I think that the more we know how the camera works, it benefits us in many ways.  Usually, in my case, it is for forensic purposes as to why the photo was messed up.  For me, the one element I start with is speed.  When speed is not the primary concern, then it is aperture.  ISO, as discussed previously, runs a distant third.  In search for the right speed, we all start at the SWAG Reciprocal Rule.  In my 35 mm days, this worked well.  The technology is so much different now.  Now, about the only thing the Reciprocal Rule is good for is maybe to double it.  For starters.  In the 35 mm days, a 1/400 sec speed was getting in the nose bleed areas - like anything over ASA 400.  Today's cameras have evolved where speeds over 1/1000 th of a second are a starting point for me.  And it all starts with the sensor - and the pixel density.