ISO Accuracy

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-308), middle gray matched. We estimate the accuracy of these results to be +/- 1/6 EV (the margin of error given in the ISO specifications). We found that measured ISO from the G3 matches the camera-stated ISO throughout the range (so ISO160 = 160, etc).

Compared to...

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.

Here, we're looking at the Panasonic Lumix DMC-G3's noise performance compared to three of its closest competitors, the Panasonic Lumix DMC-GF2, Samsung NX100 and Nikon D3100 at default noise reduction settings.

Beginning at ISO 400, differences between the G3 and the GF2's previous generation sensor start to appear, with the G3 showing less noise and more accurate image detail. At ISO 1600 the G3 pulls further away from both the GF2 and NX100, which is clearly the worst performer of the bunch. This trend continues as ISO increases, and by ISO 6400, only the G3 and the D3100 yield a satisfactory image to noise ratio.

Looking at the graph, the G3 appears to compare quite favorably to the class-leading performance of the D3100's APS-C sensor up to ISO 800. When comparing the samples, there is not a whole lot separating these two cameras. Indeed, the only other Micro Four Thirds camera we've seen give a similar result is the more expensive Panasonic Lumix DMC-GH2. At higher ISOs, however, the G3 simply cannot match the combination of low noise and fine detail of the D3100. Although the graph would seem to indicate the Panasonic models 'catch up' in noise performance, the samples emphasize the point that at this stage, much higher degrees of noise suppression are being applied to the Micro Four Thirds images, which of course, blurs detail.

Noise and Noise Reduction (JPEG)

The G3 offers five levels of noise reduction ranging from -2 to +2. The default setting is NR 0.

Even at base ISO (160), both the samples and the graph clearly show how aggressively the default setting of NR 0 suppresses noise compared with NR -2. We suspect Panasonic is attempting to strike a balance between image detail and noise suppression that will provide pleasing results for users who have no intention of adjusting NR settings. And to be honest, we don't find much fault with this approach, given the results. If you consider the NR -2 setting analogous to the NR off setting on competitors' cameras, the G3 does display more noise at lower ISOs. However, at ISO 800 and above, we find the results noticeably superior to previous Micro Four Thirds models. At NR +2 the graph can be misleading as it shows the G3 to be significantly lower in noise than competing cameras. The reality, however, is that the unusually high degree of noise suppression comes at the cost of image detail, while also introducing artifacts as the ISO increases.

RAW noise (ACR 6.5 beta, noise reduction set to zero)

Here we look at the RAW files processed through Adobe Camera Raw (in this case version 6.5 beta). 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.

All the cameras show signs of noise even at the lowest sensitivity settings (remember these samples have noise reduction turned down to zero in ACR). Yet this comparison makes clear the advantages to be gained with the D3100 APS-C sensor. The G3 is no match for the D3100 which returns cleaner output, with more detail and less noise.

Beginning at ISO 400 it becomes clear by comparison, just how much noise reduction is being applied by the G3's JPEG engine. Above ISO 800, however, the improvement in detail rendering and noise suppression over the GF2 and NX100 are evident, just as they were in the JPEG comparison.

A glance at the graph again gives a false impression that the G3 is less noisy than the D3100, as the latter renders fine detail more accurately. Let's be clear, however, the ability of the G3 to get as close as it does to what we regard as the D3100's class-leading performance is a significant achievement.

Interestingly, and as we first saw in the GH2, setting an NR value of +2 appears to apply significant smoothing to RAW files from the G3 at ISO 800 and higher. As you can see, the 'noise' and 'detail' have been significantly blurred in the RAW files captured at NR+2 compared to those shot at NR 0.