Pixel Size and Actual Photoshop Percentage Comparison Canons/Nikons/Nikon D800

[clicstudio]

Here is a little comparison based on Pixel Size and Actual Photoshop Percentage between 6 Cameras.
Pixel size doesn't really say anything. 16MP vs 32MP is not really double the size. What matter is the actual percentage in Photoshop to make the image reach the next MP size.
As you can see in the chart, there is a 15% increase from 16MP to 22MP. A percentage that won't ruin your image if you want to make it a little bigger... Now, a full 50% enlargement will introduce artifacts, noise and blur to your image, so you probably won't want to do that...

I do believe in per-pixel quality and noise levels but if you are a studio photographer who shoots at ISO 50 or 100 most of the time, you probably don't care and want 50% more Megapixels to play with. I do see the fact that the Nikon D4 is going to cost double the D800 and have  less than half the MP so there has to be something there to make it better but if u want raw huge files the D800 will be the winner no matter how you see it.

[yunusoglu]

Here is a little comparison based on Pixel Size and Actual Photoshop Percentage between 6 Cameras.
Pixel size doesn't really say anything. 16MP vs 32MP is not really double the size. What matter is the actual percentage in Photoshop to make the image reach the next MP size.

16MP vs. 32MP actually IS double the size...
It's just not double the height or width.

[Shnookums]

Pixel size doesn't really say anything. 16MP vs 32MP is not really double the size.

Rotate the Red rectangle 90 degrees and you can put 2 in the D800 rectangle. So yes, it is double the size.

[jchung]

Difference of Length vs Area.

Area = L x W.

MP == Area.

MP != Length.

2 x Area would be equivalent to 41.4% increase in a single linear dimension (i.e. length).

An increase from 16MP to 22MP is a 37.5% increase in area, but a 17.3% increase in a single linear dimension.

Not a 15% increase in either case.

[jweu]

You can take Europe papersize for an example: A0 is a square meter and A1, A2, ... is always a ½ of the previous one. A sheet of paper A4 with 80g per m² has a wight of 4g (=80g/16=80g·½·½·½·½). With portait-oritentation: A4 has a with of about 21cm and the height of A5 is exactly the same because of the page format; the height of A4 is double the width of A5. => For every sheet: width * 1.141… = height .

[Astro]

16MP vs. 32MP actually IS double the size...
It's just not double the height or width.

MP is just a number is has no physical dimension unlike..... cm^2 or m or kg.

so it´s twice the amount of pixels but not size.

[chriswatters]

MP is just a number is has no physical dimension unlike..... cm^2 or m or kg.

so it´s twice the amount of pixels but not size.

If you render the pixels they will have a physical size. MP is a scalable measure of area. Whatever area you assign to 1 pixel, 2 pixels will be twice that area. Whatever area you use to render 16MP, 32MP will be twice that size.

Rotate the Red rectangle 90 degrees and you can put 2 in the D800 rectangle. So yes, it is double the size.

Given the same aspect ratio (ie 3:2), this illistrates the point perfectly.

[Astro]

If you render the pixels they will have a physical size. MP is a scalable measure of area. Whatever area you assign to 1 pixel, 2 pixels will be twice that area. Whatever area you use to render 16MP, 32MP will be twice that size.

well you mix things up here... MP have no unit of area assigned to them per definition.
only if you give an additional information (1 pixel x 1 mm^2) you give them a dimension.

you can print 100 MP as 10x10cm print and you can print 10 MP as 10x10cm print.

so nowhere in a megapixel number alone.... you have information about size.
and that´s what im saying.

[Rexepic]

Why is everyone mixing up sensor size with megapixel size?  Both arguments are correct. If you establish a size for the pixels, then 36 is double 18, BUT this is not equatable with sensor size, even if the receptors(pixels) are jammed in there.

We need the optical physics guy here. Larger sensors have optical physics properties, regardless of megapixel count, that smaller sensors don't (due to light gathering area), but don't confuse that and think that more megapixels is the equivalent of a larger sensor.

It might have the megapixel count of medium format, but it is definitely not a medium format sensor. Hasselblad and Mamiya are not quaking in their boots over this. But it is cool to have the MPs, though I'm not sure that the file size is worth it unless you're trying to print billboard size photos!

[awinphoto]

If you render the pixels they will have a physical size. MP is a scalable measure of area. Whatever area you assign to 1 pixel, 2 pixels will be twice that area. Whatever area you use to render 16MP, 32MP will be twice that size.

well you mix things up here... MP have no unit of area assigned to them per definition.
only if you give an additional information (1 pixel x 1 mm^2) you give them a dimension.

you can print 100 MP as 10x10cm print and you can print 10 MP as 10x10cm print.

so nowhere in a megapixel number alone.... you have information about size.
and that´s what im saying.

Yes BUT with the increased resolution, the 100 MP  will deliver cleaner larger prints than the 10MP will deliver...