ISO Sensitivity / Noise levelsISO equivalence on a digital camera is the ability to increase the sensitivity of the sensor. This works by turning up the "volume" (gain) on the sensor's signal amplifiers (remember the sensor is an analogue device). By amplifying the signal you also amplify the noise which becomes more visible at higher ISO's. Many modern cameras also employ noise reduction and / or sharpness reduction at higher sensitivities.
To measure noise levels we take a sequence of images of a GretagMacBeth ColorChecker chart (controlled artificial daylight lighting). The exposure is matched to the ISO (ie. ISO 200, 1/200 sec for consistency of exposure between cameras). The image sequence is run through our own proprietary noise measurement tool (version 1.5 in this review). Click here for more information. Room temperature is approximately 22°C (~72°F), simulated daylight lighting.
Nikon D5000 vs. Canon EOS 500D vs. Olympus E-620 vs Panasonic DMC-G1
- Nikon D5000: Nikkor 50 mm F1.4G lens, Aperture Priority, Manual WB,
Default Parameters. Standard Picture Controls, Active D-Lighting Off, NR Normal, JPEG Large / Fine
- Canon EOS 500D: Canon 50 mm F1.4 lens, Aperture Priority, Manual WB,
Default Parameters, Standard Picture Style, NR Standard, JPEG Large / Fine
- Olympus E-620: Olympus 50 mm F2.0 Macro lens, Aperture Priority, Manual WB,
Default Parameters, Normal Picture Mode, Gradation Normal, Noise Filter Normal, JPEG Large / Fine
- Panasonic DMC-G1: Olympus 50 mm F2.0 Macro lens (via adapter), Aperture Priority, Manual WB,
Default Parameters (Normal), High ISO NR Normal, JPEG Large / Fine
|Nikon D5000||Canon EOS 500D||Olympus E-620||Panasonic DMC-G1|
These are the noise results for the default JPEG outputs from each camera (we'll investigate the D5000's different noise reduction settings on the next page). The D5000 does an admirable job at keeping noise at bay up to ISO 800 - its softer JPEGs never compete with the EOS 500D for detail but they're comparable in terms of detail lost. It's only from ISO 1600 and up that the D5000 shows much in the way of noise or noise reduction. The ISO 3200 result is starting to get pretty smeared and the showing noticeable noise.
ISO 6400 equivalent is noisy and smeary for both the Nikon and the Canon. On the Nikon, the highest setting is an ISO equivalent value which is likely to mean that the sensor is being run just as it would be at ISO 3200 but exposed as if it were ISO 6400, with all the values pulled-up to give the correct brightness. Canon defines ISO 6400 as an 'expanded ISO' which may involve the same process or just denote that Canon does not feel it is of sufficient quality to make it an option by default. Either way, the results are similar.
With ACR taking away any of the clever processing done by the manufacturers' JPEG engines, the D5000 does well in terms of noise, with neither chroma nor luminance noise becoming too obvious until ISO 800 (which is consistent with our experience of when the selectable noise reduction processing starts to kick-in). The D5000 seems to have the edge on its immediate peers in terms of noise.
Bear in mind that when looking at the images at the same output size, the EOS 500D's higher pixel count means the visual impact of that extra noise will be reduced. However, since the difference in resolution is small and the difference in noise at the pixel level is fairly large, we'd conclude the Nikon will retain an advantage in terms of most real-world usage, albeit a fairly small one.
|Nikon D5000 RAW||Canon EOS 500D RAW||Olympus E-620 RAW||Panasonic DMC-G1 RAW|
Raw Noise graphs
Because the we've turned ACR's noise reduction down as far as possible ('Off' may not mean that no noise reduction is done), the graph results mirror what we've seen from the crops above, with the D5000 doing a good job across all settings.
|Nikon D5000 RAW|