D600 High ISO in DX

Started Nov 23, 2012 | Questions thread
noirdesir
Forum ProPosts: 11,048
Like?
Re: pixel pitch and SNR
In reply to Leo360, Nov 23, 2012

Leo360 wrote:

bobn2 wrote:

Photosite (or even pixel) size has very little connection to sensor efficiency and therefore low light performance. The pixels might be bigger, but you have fewer of them, so in the end the same amount of light gets collected.

-- hide signature --

Bob

There are two different things here. There is photon count per pixel and number of photons collected per unit area. The latter does not depend on the pixel pitch but the former does. And SNR per pixel gets larger with more photons collected by that pixel (photon shot-noise per photon gets weaker). For the same exposure larger pixels capture more photos and, thus, have higher SNR. This is why pixel peeping reveals more noise-per-pixel for smaller photosites. The price to pay is reduced resolution.

With proper down-sampling (bicubic, etc) to the same level of detail one does can hope to recover the SNR back by effectively combining outputs of multiple smaller pixels into an aggregate one but doing so does not entirely compensate for read-noise increase.

The photon shot noise per sensor area will be exactly the same for both the smaller and the larger pixel sensors, assuming the same quantum efficiency.

And the read noise component of the final noise usually is the same or even lower for smaller pixels because noise is not something that gets simply summed up as it is a standard deviation.

In an ideal world, read noise is proportional to pixel size. So, let's take a 2x2 μm pixel with a read noise of 6 e- (ie, a standard deviation of 6 e-). Now compare that to a 2x2 pixel array of 1x1 μm pixels, where each pixel has a read noise of 1.5 e-. Simple statistics tell us that the read noise component for a combination of these four smaller pixels is:

rn(com) = square root of (1.5^2 + 1.5^2 + 1.5^2 + 1.5^2) = 3 e-

ie, lower than that of the sensor with larger pixels. Of course, whether the smaller pixels achieve a read noise reduction proportional to their area depends on how the pixel size reduction is achieved but if it is through a process shrink, you can actually have less noise in your final image with smaller pixels (all else equal, ie, sensor size and QE).

Take a real-life example, the Nikon D600 and D800 which have sensors of pretty much the same generation from the same design team (Sony) using more or less the same design goals and philosophies. Read noise numbers vary a bit over the ISO settings but if we omit the highest and lowest ISO setting we get an average of 3.74 e- for the D600 and 3.1 e- for the D800. And if we round the pixel size a little bit three D800 pixels cover the same area as two D600 pixels. Thus:

rn(D800) = square root of (3.1^2 + 3.1^2 + 3.1^2) = 5.37 e-

rn(D600) = square root of (3.74^2 + 3.74^2) = 5.29 e-

That is essentially a draw. But if we compare this with the D3x we get 5.80 e-, so going from the D3x to the D800 we achieved less read noise per sensor area while having smaller pixels. If we go one step further and look at the Olympus OM-D E-M5 (which is widely assumed to have a Sony sensor as well), read noise 2.62 e- per pixel, we get (for more instructive comparions we take 63 D800 pixels and 100 E-M5 pixels):

rn(D800) = 29.7 e- and rn(E-M5) = 26.2 e-, a small but measurable difference in favour of the smaller pixels.

(All read noise values from sensorgen.info)

Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow