ISO / Sensitivity accuracy

In a new addition to our reviews we are now measuring the actual sensitivity of each indicated ISO sensitivity. This is achieved 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 Sekonic L-358), middle gray matched. We estimate the accuracy of these results to be +/- 1/6 EV.

As per the Olympus E-330 and Canon EOS 30D the DMC-L1 proved to be about a third faster than indicated, this means that for instance ISO 100 has an equivalent sensitivity of ISO 125.

ISO / Sensitivity noise levels

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 1.4 in this review). Click here for more information. (Note that noise values indicated on the graphs here can not be compared to those in other reviews). Room temperature is approximately 22°C (~72°F), lighting simulated daylight.

Panasonic DMC-L1 vs. Olympus E-330 vs. Canon EOS 30D

• Panasonic DMC-L1: Olympus 50 mm F2.0 Macro lens, Aperture Priority, Manual WB,
  Default Parameters (Standard), JPEG Large / Fine
   
  
• Olympus E-330: Olympus 50 mm F2.0 Macro lens, Aperture Priority, Manual WB,
  Default Parameters (Vivid), JPEG Large / Fine, ISO Boost: On
   
• Canon EOS 30D: Canon 50 mm F1.4 lens, Aperture Priority, Manual WB,
  Default Parameters (Standard PS), JPEG Large / Fine
  

Panasonic DMC-L1 ISO 100 (125 equiv.)	Olympus E-330 ISO 100 (125 equiv.)	Canon EOS 30D ISO 100 (125 equiv.)
Panasonic DMC-L1 ISO 200 (250 equiv.)	Olympus E-330 ISO 200 (250 equiv.)	Canon EOS 30D ISO 200 (250 equiv.)
Panasonic DMC-L1 ISO 400 (500 equiv.)	Olympus E-330 ISO 400 (500 equiv.)	Canon EOS 30D ISO 400 (500 equiv.)
Panasonic DMC-L1 ISO 800 (1000 equiv.)	Olympus E-330 ISO 800 (1000 equiv.)	Canon EOS 30D ISO 800 (1000 equiv.)
Panasonic DMC-L1 ISO 1600 (2000 equiv.)	Olympus E-330 ISO 1600 (2000 equiv.)	Canon EOS 30D ISO 1600 (2000 equiv.)
Panasonic DMC-L1 n/a	Olympus E-330 n/a	Canon EOS 30D ISO 3200 (4000 equiv.)

Sharing the same sensor, although with different image processors, the DMC-L1 and E-330 deliver similar results. The primary difference between the two is the visibility of chroma noise (color blotches) and the level of detail at ISO 1600. The DMC-L1 is better in both respects, its noise reduction algorithm dealing well with the blotchy chroma noise without adversely affecting detail (note that the DMC-L1's default image parameter set 'Standard' has a minimum noise reduction setting).

However probably the DMC-L1's biggest challenge is overcoming the obvious advantage of the Canon CMOS sensor used in the EOS 30D which delivers lower noise and better detail / sharpness through to ISO 1600 and even provides a usable ISO 3200.

Luminance noise graph

The DMC-L1's luminance noise matches the E-330 all the way up to ISO 800 (actual ISO 1000) but then takes a jump at ISO 1600 because of lower luminance noise reduction. The EOS 30D delivers a steady consistent climb in noise and manages to maintain very low shadow noise.

Chroma noise graph

As shown in the crops and discussed above the E-330 demonstrates more chroma noise (color blotches) at higher sensitivities than the DMC-L1.

Noise Reduction graph

By default the DMC-L1 has a noise reduction setting of -2 (the minimum) which is admirable, it means you're getting the most detail (at the expense of some noise) with the option to increase NR yourself if you wish. The L1's noise reduction appears to be most effective in shadows, as 'Black' in the graph below.

Effect of constant Live View over time

One obvious concern when you're using the main sensor to provide live view is the build-up of heat over time, and heat means noise. To test this we took noise measurements at ISO 100 and 1600 over a period of 20 minutes leaving Live View enabled all the time. As you can see there's no clear increase in noise even after 20 minutes.