|Panasonic DMC-G1||$670||12.1 MP CMOS; 17.3 x 13.0 mm (2.0x crop)||5.0|
|Olympus E-520||$540||10.1 MP CMOS; 17.3 x 13.0 mm (2.0x crop)||4.0|
14.2 MP CMOS; 23.6 x 15.8 mm (1.5x crop)
|Canon 450D||$630||12.2 MP CMOS; 22.2 x 14.8 mm (1.6x crop)||3.7|
|Canon G10||$410||14.7 MP CMOS; 7.6 x 5.7 mm (4.6x crop)||34|
|Fujifilm S100fs||$640||11.1 MP CMOS; 8.8 x 6.6 mm (4.0x crop)||19|
For direct comparisons we always use sharp prime lenses stopped down. Here we have used the Olympus Zuiko 50mm F2.0, Canon EF 50 mm F1.4 and Minolta 50mm F1.4. For the comparisons against the two compacts with their integrated optics, we've used the Panasonic G Vario 14-45mm F3.5-5.6 that comes with the G1.
Studio scene comparison (JPEG)
Panasonic Lumix G1 vs. Olympus E-520
- Panasonic DMC-G1: Olympus Zuiko 50 mm F2.0 lens, Aperture Priority, ISO 100
JPEG Large/Fine, Manual WB, Default Parameters (Standard)
- Olympus E-520: Olympus Zuiko 50 mm F2.0 lens, Aperture Priority, ISO 100
JPEG Large/Super Fine, Manual WB, Default Parameters (Standard)
5.0 MB JPEG (4000 x 3000)
5.9 MB JPEG (3648 x 2736)
The most immediately apparent feature of the G1 over the E-520 is the slight, but visible additional detail the Micro Four Thirds camera is able to pull out of the scene in comparison to the conventional Four Thirds DSLR. More fine detail is visible in every crop, from the horizontal lines on the medals (third crop down), to the more textured rendering of the pastoral scene (crop six).
Some of this difference will be down to it having more pixels to represent detail with but it could also be due to the different approaches the two companies take to the strength of the anti-aliasing filters that very slightly blur the image to prevent jagged edges appearing. We've always suspected the E-520's AA filter to be fairly strong but the slight jagged edges on the paperclips in the G1's fifth crop (and the occasional moiré in our real world shots) suggest it may have gone to the other extreme. Of course there will also be an influence of in-camera sharpening that should be eliminated when we compare the images in RAW.