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). We found that measured ISO from the E-5's measured ISO is within +/-1/6 EV of the indicated ISO, across its entire sensitivity range.
Noise and Noise Reduction (JPEG)
At its default (Standard) noise reduction setting, the E-5's gives excellent image quality at the low end of its ISO scale, and - considering its relatively low pixel count - resolution is very good. At ISO 800, however, noise reduction really kicks in, and takes the edge of fine detail. This is especially noticeable in comparison with the other cameras in this table, the best of which - the Nikon D7000 and K-5 - still describe plenty of detail right up to ISO 6400. By this point, the E-5 has hit its ISO ceiling, and image quality is decidedly shaky.
Here we're looking at the E-5's NR settings in comparison against one another. The E-5 offers four noise reduction settings (including 'off') but even when turned off, you have to look pretty closely at the JPEG output to see much noise at ISO sensitivities below ISO 800. Above ISO 800, luminance noise is very noticeable, and becomes more intense as ISO sensitivity is increased, until by ISO 6400, 'salt and pepper' luminance noise has a very destructive effect on image quality.
As you can see, higher up the ISO scale things are a lot smoother when noise reduction is set to 'Standard', but one of the risks of noise reduction is that it also reduces detail. Unfortunately, at ISO 800 and above, the E-5's noise reduction system starts to have a deleterious effect on image quality, until by ISO 6400, almost no fine detail is present at all. The situation is better with NR set to 'low' and - conversely - fairly dire when set to 'high'. By comparing the 'samples' and 'graph' view, above, you can compare the E-5's image quality to its measured noise levels. Notice how the more detailed samples, taken with NR 'off' also give the highest measured noise readings.
RAW noise (ACR 6.3 noise reduction set to zero)
The amount applied 'under the hood' in this way is not high, but it does vary by camera (Adobe is attempting to normalize output across different sensors). What this means is that 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) but up to ISO 400 the differences are marginal. Looking at the color patches though, it is obvious that although detailed, the E-5's output is significantly 'grittier', and gets even more so as the ISO sensitivity is raised. At the highest ISO settings of 3200 and 6400 (equivalent) the E-5 is clearly lagging behind the competition in terms of noise, and a glance at the noise graphs confirms that measured noise (of all types) is significantly higher from the E-5 than from the other cameras in this comparison.
It is worth remembering at this point that the E-5's sensor first saw the light of day in late 2008. It is clear from looking at the samples and graphs on this page that the rest of the market has simply moved on.