Night Shootout: Panasonic GF1 vs Canon 5D Mk II

When I was talking to a fellow travel photographer, Paul, to organise a bit of a test of these two cameras, I was expecting the scene to look a little bit like the one above, which I shot 2 days later.

The plan was to for me to take a few night shots of Zagreb, while Paul put together a time-lapse video, then at the end of it shoot a few frames that were as close as possible so that we could get an idea of just how much difference there is between the images from the two cameras.

Instead of nice clear blue sky however, as the shooting progressed we ended up with lots of cloud by the time it got around to do a comparison. Nevertheless I've put together some comparison shots.

Have a look at the full post for details and samples! (If you're not interested in reading a technical comparison of 2 cameras... no need to read further!)

No... Its not Fair!

This "test" was aimed at getting an impression - not really to provide a "fair" comparison. We were aiming to get the exposure value and depth of field as close together as possible.

The Cameras:

	Canon 5D Mark II	Panasonic DMC GF1
Lens	Canon 17-35mm f2.8 L	Olympus 9-18mm f4-5.6
Weight (body)	810g - 28.6oz	285g - 10.05 oz
Weight (lens)	545g - 19.2 oz	275g - 9.7 oz
Price (Body)	$2499 (body only)	$749 (with 14-45 lens)
Price (lens)	$1,350.00 (used)	$474.99 (new)
Raw file size	29.2 MB	13.9 MB

Settings used:

	Canon 5D Mark II	Panasonic DMC GF1
Aperture	f6.3	f4
Shutter speed	30 seconds	15 seconds
ISO rating	125	100 (125 equiv)
Mounted:	Tripod	gorillapod
White Balance	Auto	Auto
Focus	Auto	Manual

Conditions:

In many ways the conditions for the test weren't exactly conducive to taking nice night-time photos. The lighting for the streets comes mainly from Gas street lights, but at this time the clouds are absorbing a lot of tungsten light pollution from the city.

Although we'd planned to get identical shots, my version of the image with light trails is blurred due to bumping the pole that my gorillapod and camera were hanging off! As a result this image is taken slightly (ax. 1 min) later with different cloud formations.

As you can see from the default images produced by Lightroom, both cameras have difficulty with the White Balance.

Hot Pixels:

I wasn't expecting to see hot pixels in either image. When using the Olympus E30 at night with similar settings, there simply wouldn't be any. At 45-60 seconds they start to creep in.

The GF1 and 5DII images both have hot pixels that are evident in the images - in the case of the 5D these are controlled quite well by the color NR in Lightroom. On the GF1 they're not that evident, but you can still see traces of artifacts where they've been and they result in slight image degradation. At 30sec on the GF1 they're very prevalent. Both images are however taken at the end of a fairly long nighttime shoot so both cameras are well and truly "warmed up".

Compared to the Olympus E30, it seems that the much smaller body on the GF1 comes at the cost of some signal processing and probably more tightly packed circuitry which tends to increase overall heat.

Sharpness:

There is one crop below that stands out as an area where the GF1's image is clearly sharper. This is taken from the right part of the frame which is also one of the areas closest to the camera.

While on explanation could be that it shows corner softness in the Canon's images, this isn't consistent with the crop taken from the other extremity. The right part of the image is however physically closer to the camera.

Pixel-peeping the entire image, this difference can be seen across the whole image along the paving right up to a point just past the first set of flags. Before this point, the GF1 image is sharper, after it the Canon. Clearly this difference is a lens and focus issue. I suspect that the point at which "infinity" starts on the Olympus lens is closer than on the Canon.

Dynamic Range

An area of clear difference between the two is in Dynamic range. This can be seen in both the highlights and shadows. The 5D image retains detail in the clock and there is also significantly more shadow detail - if you try to recover the highlights in the clock, you can get back almost all the numbers. On the GF1 you get no more detail than can be seen in the crop.

In the shadows, the 5D image retains a significant amount of detail in the dark areas, where there isn't much information in the GF1 files. Its possible that some if this difference is due to the different focus distance however.

Lenses Used

The Canon lens used is an older lens design that isn't currently in production. One would expect one of the newer lens designs to perform better. Having said that, the comparison is against an Olympus lens that is rated as one of the consumer grade by Olympus. All in all, it holds up pretty well.

Conclusions:

The GF1 wins on:

• Price
• Weight
• White balance

The 5D II is clearly better in:

• Dynamic Range
• Image size

On these tests its hard to be definitive about resolution, but I think one can safely assume that with the appropriate lenses the 5D should be comfortably ahead there too.

Obviously there are many features that the 5D has that aren't relevant to this test. As I indicated from the outset, it isn't aimed at being fair, and isn't aimed at making a statement about which camera is "better" - more to get an impression. Both are outstanding and exciting cameras in their own right.

I'd appreciate your thoughts!