The actual sensitivity of each indicated ISO is measured using the same shots as are used to measure ISO noise levels, we simply compare the exposure for each shot to the metered light level (using a calibrated Sekonic L-358), middle gray matched. We estimate the accuracy of these results to be +/- 1/6 EV (the margin of error given in the ISO specifications). Note that these tests are based on the sRGB JPEG output of the cameras, in accordance with ISO 12232:2006, the standard used by camera manufacturers. We found that measured ISO from the D3100 is roughly 1/3EV higher than indicated - so ISO 100 = ISO 125 (approx). This holds true throughout the entire ISO range, but a discrepancy this small has little practical impact upon everyday photography.
Noise and Noise Reduction (JPEG)
This is our standard studio scene comparison shot taken from exactly the same tripod position. Lighting: daylight simulation, >98% CRI. Crops are 100%. Ambient temperature was approximately 22°C (~72°F).
Note: this page features our new interactive noise comparison widget. By default, we show you the default noise reduction settings of the camera tested, and three other models of the same class. You can select from all available NR options, and from other cameras. The 'tricolor' patches beneath the familiar gray/black/portrait images are taken from the same test chart, and show how noise impacts upon blue, green and red areas of a scene.
The Nikon D3100's default noise reduction setting is 'On' and we'd be happy to leave it at that setting. The camera does a good job of keeping noise in check without too much of a negative effect on image quality. The effects of noise reduction (both in terms of detail loss and noise suppression) are only really visible above ISO 1600 and even at 3200 the results are excellent.
From ISO 6400 noise is increasingly allowed into the image but even at this point the attempts to keep it down don't result in unreasonable amounts of detail loss. Only at the ISO 12800 equivalent setting (Hi2 in Nikon terminology) does the camera lose its struggle against noise - as we're used to seeing in cameras with this sensor size - there's just too much noise to cope with and the image quality drops dramatically. However, this is an ISO setting that's still pretty ambitious, so we're not at all surprised by this result.
RAW noise (ACR 6.3 - noise reduction set to zero)
Here we look at the RAW files processed through Adobe Camera Raw (in this case version 6.3). Images are brightness matched and processed with all noise reduction options set to zero. Adobe does a degree of noise reduction even when the user-controlled NR is turned off.
The amount of NR applied 'under the hood' is not high, but it does vary by camera (Adobe is attempting to normalize output across different sensors), so inevitably we are still looking at a balance of noise and noise reduction, rather than pure noise levels. However, the use of the most popular third-party RAW converter is intended to give a photographically relevant result, rather than simply comparing sensor performance in an abstract manner.
The three APS-C cameras in this comparison are all so close as to make no difference (which to a degree tells us as much about Adobe's ability to normalize the various cameras' outputs as about the cameras themselves). The E-PL1, despite its slightly smaller Four Thirds sensor, keeps pace at ISOs up to 800, but above that starts to show visibly more noise, particularly in the blacks. The ISO 6400 and 12800 shots show just what the cameras' JPEG engines are having to cope with; pretty much all of the detail is swamped by noise at the highest setting.