The Myth of Megapixels?

Angelo

an interesting read...

http://www.nytimes.com/2007/02/08/technology/08pogue.html?_r=1&8dpc&oref=slogin

Andy

Angelo wrote:

an interesting read...

http://www.nytimes.com/2007/02/08/technology/08pogue.html?_r=1&8dpc&oref=slogin

http://www.dgrin.com/showthread.php?t=53437

Angelo

a day late and a dollar short...

story of my life :cry

Andy

Angelo wrote:

a day late and a dollar short...

story of my life :cry

No, they are different stories

rutt

The big picture in the megapixel department isn't whether today's 4/8/12/16/24 megapixel camera has enough resolution for whatever sized print or online use.

The big picture is that sensor density can continue to increase with Moore's law (exponential growth, which has fueled the computer tech boom.) So at some point we can put more on the sensor than we need just for resolution. What is that point? Who cares! We are going to exceed it and leave it behind in the dust.

What then? One option is High Dynamic Range via multiple sensors for each pixel for different light ranges. Fuji does this in a sort of limited way. The Stanford Light Field camera which focuses after the fact is another idea. The Foveon sensor which measures color at each sensor point is yet another idea. Eventually, we'll get all these things.

And more. What else?

System

But, is increased sensor density a *feature*?

Several sources assert, and I tend to believe them, that since increased density -- in the same space -- leads to smaller pixels, that inherent noise goes up, and therefore you don't necessarily get better pictures.

My Oly, for example, is only a 4MP.

But it has a 62mm lens front, and all that F2.0 glass makes for really nice shots, or so I think. They're certainly good enough for what I do with them.

Certainly, if I had to shoot for a magazine double-truck, I would want 10 or 12MP. But I could just about get away with a single page, uncropped... as long as it wasn't Playboy. :-)

mercphoto

Flawed experiment

Angelo wrote:

an interesting read...

http://www.nytimes.com/2007/02/08/technology/08pogue.html?_r=1&8dpc&oref=slogin

Interesting indeed, but I agree with the "angry group" who said the test was flawed because they author down-rezzed the 13MP shot to 5MP before printing it. He was trying to compare Oranges to Apples at first in an attempt to see if Oranges were any different than Apples. But he rigged the test in such that he was comparing Oranges to Oranges and surprised that the two fruits tasted the same. Well, d'uh!

I do not understand what he was trying to prove by not printing all 13 million pixels to see if the extra pixels made a difference.