Why Are Larger Pixels on Camera Sensors Better?

[skyglider]

I understand that larger photosites on a digital camera sensor have benefits such as improved dynamic range and lower noise due to improved light collection...but I can't help but think that maybe having larger photosites isn't better.
--- snip ---

I would think that having very small photosites would be better, as long as they can be efficient, because each photosite will then be more accurate because it will correspond to a more precise part of the scene being photographed. So why are larger photosites better?

For prints, my take is:

Dpi for optimum print quality is 300 dpi. An 8 x 10 inch print requires a 2400 x 3000 pixel image to print at 300 dpi. With higher resolution sensors, adjacent photosites are combined to print at 300 dpi. So wouldn't the precision of smaller photosites be negated anyway? As you mentioned, larger photosites results in less noise.

Of course having a sensor with higher resolution than 2400 x 3000 pixels for 8x10 inch prints is good to allow for cropping without having to frame the image exactly every time. So a balance needs to be achieved in this regard.

For other than making prints, I donno. I only shoot to make prints, and displaying on a TV set is a non issue using print quality images. I think the mega-pixel race that the manufacturers have is to sell cameras only, not for image quality. My take anyway,

Sky

 

[BobSC]

In fact, it exactly makes up for it when looked at mathematically. If at a given

And the reason Canon reduced the pixel count on the G11 is?

I don't know. Do you, really?

If you look at the results the intent seems pretty clear.

Canon chose to reduce the pixel count in order to improve the low light performance ("noise") at the expense of absolute resolution.

They decided, that for their target market, fewer pixels = better pictures.

 

[sacundim]

And the reason Canon reduced the pixel count on the G11 is?

I don't know. Do you, really?

If you look at the results the intent seems pretty clear. Canon chose to reduce the pixel count in order to improve the low light performance ("noise") at the expense of absolute resolution. They decided, that for their target market, fewer pixels = better pictures.

And how do you claim to know this? There are plenty of examples of newer cameras that increased the pixel count and the low light performance at the same time, compared to their predecessors. So I don't see how your argument is any better than either of these:

1. The decrease in pixel count caused the low-light performance improvement because the low-light performance improvement came at the same time as the decrease in pixel count. (This is the classic "B happened after A, ergo, A caused B" fallacy.)
2. The fact that the decrease in pixel count came with improved low-light performance lends credence to your prior hypothesis that fewer pixels = better low-light performance. The problem there is that, well, you're ignoring all those other camera upgrades that gave us both more pixels and better low-light performance.

 

[Great Bustard]

I understand that larger photosites on a digital camera sensor have benefits such as improved dynamic range and lower noise due to improved light collection...but I can't help but think that maybe having larger photosites isn't better.

Think of it this way, a larger photosite captures more light, yes, but it also captures light from more angles, thus an individual photosite on a large photosite camera sensor will be receiving the light from more places in the scene being photographed than a smaller photosite. Wouldn't having bigger photosites actually lower the resolving power of the sensor then? Do you understand what I'm getting at?

I would think that having very small photosites would be better, as long as they can be efficient, because each photosite will then be more accurate because it will correspond to a more precise part of the scene being photographed. So why are larger photosites better?

Thank you!

If you want to know why, then look no further:

http://forums.dpreview.com/forums/read.asp?forum=1018&message=32064270

 

[Desmond Ong]

Yes, larger pixel for camera sensor can produce more vibrant and sharper pictures. In my opinion, 12-16 MPs is more than sufficient for general photographics needs as we only wants its size for up to 8R. Larger MPs mainly suits for studio/ads shooting where they wanna to boom up to giant size banner.

 

[Great Bustard]

Yes, larger pixel for camera sensor can produce more vibrant and sharper pictures. In my opinion, 12-16 MPs is more than sufficient for general photographics needs as we only wants its size for up to 8R. Larger MPs mainly suits for studio/ads shooting where they wanna to boom up to giant size banner.

So long as the sensor in the camera that has more pixels is at least as efficient as the sensor with larger pixels (and often even when it isn't as efficient), the sensor with smaller pixels will produce a "better" photo when both are displayed at the same size.

That is, if you are looking at 100% crops for each, then the photo made from the smaller pixels may look worse, since you are viewing the photo at a greater enlargment, But at the same display size, the photo with smaller pixels will win the IQ contest, especially if NR (noise reduction) is judiciously applied. I mean, it could not be more clear:

http://forums.dpreview.com/forums/read.asp?forum=1018&message=37714016

http://forums.dpreview.com/forums/read.asp?forum=1018&message=39133368

http://forums.dpreview.com/forums/read.asp?forum=1012&message=37549285

 

[ChrisLumix]

Think of it this way, a larger photosite captures more light, yes, but it also captures light from more angles

This may be true in theory (though I'm not even sure of that) but in practice, each pixel is so small that to think of 'angles' is misleading. Think of it more like this : on your TV picture, isolate one single dot on the screen. How much of the total image is that giving you? Effectively nothing recognisable, as it's the total of millinos of dots together that produce the TV picture, not one single dot.

Therefore the effectiveness of that dot in relation to the whole, and the combined effect of that, is what counts. So if a 'dot' is noisy and inefficient, then increase or decrease its size until it is the optimum.

But yes, resolution is determined by the total number of pixels, that part is true. The question is, do you want a massively poor resolution, or a lower but higher quality resolution? I know which I'd go for.

--

Panas0n!c Lum!x FZ-38 (The word "LOSE" is spelled "LOSE"! It's not spelled "LOOSE", ok?)

 

[Great Bustard]

But yes, resolution is determined by the total number of pixels, that part is true. The question is, do you want a massively poor resolution, or a lower but higher quality resolution? I know which I'd go for.

...right here:

http://forums.dpreview.com/forums/read.asp?forum=1018&message=35068712

So long as the sensor is at least as efficient (and even when it isn't, in many cases), more pixels results in higher IQ for a given sensor size.

 

[cberry]

In fact, it exactly makes up for it when looked at mathematically. If at a given

And the reason Canon reduced the pixel count on the G11 is?

I don't know. Do you, really?

I know - they made the camera better

 

[BobSC]

And the reason Canon reduced the pixel count on the G11 is?

I don't know. Do you, really?

If you look at the results the intent seems pretty clear. Canon chose to reduce the pixel count in order to improve the low light performance ("noise") at the expense of absolute resolution. They decided, that for their target market, fewer pixels = better pictures.

And how do you claim to know this? There are plenty of examples of newer cameras that increased the pixel count and the low light performance at the same time, compared to their predecessors. So I don't see how your argument is any better than either of these:

1. The decrease in pixel count caused the low-light performance improvement because the low-light performance improvement came at the same time as the decrease in pixel count. (This is the classic "B happened after A, ergo, A caused B" fallacy.)
2. The fact that the decrease in pixel count came with improved low-light performance lends credence to your prior hypothesis that fewer pixels = better low-light performance. The problem there is that, well, you're ignoring all those other camera upgrades that gave us both more pixels and better low-light performance.

If Canon could have improved the low light performance of the camera without reducing the pixel count, why didn't they do that?

If they could have improved low light performance while increasing pixel count, why wouldn't they have done that?

Instead, they did something completely opposite industry norms and consumer expectations -- the reduced the pixel count.

Unless you can provide some reasonable alternative hypothesis, I think the only logical conclusion is that Canon reduced the pixel count because it improved image quality -- why else would they do it?

I agree that sensors have been, and probably will continue, to improve over time.

Turning up the gain amplifies noise (both photon and electrical). The smaller the site, the higher the necessary gain.

 

[sherwoodpete]

Turning up the gain amplifies noise (both photon and electrical). The smaller the site, the higher the necessary gain.

The smaller the site, the less of a contribution it makes to the overall image.

 

[Mark S Abeln]

You can get decent images with cameras that only deliver one bit per pixel. Film photography is a good example of this, for in black and white film, for you either have a film grain at a location on the negative, or you don't. There are no shades of gray.

I wrote an article on this: http://therefractedlight.blogspot.com/2011/08/1-bit-depth-cameras.html

--
http://therefractedlight.blogspot.com

 

[Leon Wittwer]

Instead, they did something completely opposite industry norms and consumer expectations -- the reduced the pixel count.

Unless you can provide some reasonable alternative hypothesis, I think the only logical conclusion is that Canon reduced the pixel count because it improved image quality -- why else would they do it?

Canon may have been doing a marketing experiment to see if there was some marketing potential in limiting the pixel count. Since many believe, correct or not, that fewer pixels would be better, maybe that would translate into sales. Of course, the imager cost may have been less but that would be just a bonus. Regarding sacrificing image quality, that would only significantly occur it a shooter wanted a sufficiently large print where more pixels (and more resolution) were actually needed. I do have a G11 and with decent light, I can get a good 8x10 or 11x14 print.
--
Leon
http://web.me.com/leonwittwer/landscapes.htm