Sensor Shootout: Panasonic GH2 vs Panasonic GX1 vs Canon 1Ds Mark II
As many of you are aware by now, I have recently made the switch from Canon DSLRs to the micro 4/3 system. I’ll all the details for another post, but it was primarily to reduce size and weight. So over the last year or so have sold almost all my DLSR equipment and transitioned entirely to Micro 4/3, namely the Panasonic GH2 and GX1. These cameras are at the top of the Micro 4/3 format of mirrorless Compact System Cameras and are large sensor interchangeable lens cameras that can provide very high image quality in a very small package.
I had been extremely pleased with the image quality out of my new cameras, and surprised that in print I essentially can’t tell them apart from my old Canon files. So, before the very last of my DSLR gear goes away in the next week or so, I decided to set up a little test to see in reality how they perform vs what I was shooting. So I have pitted my little Micro 4/3 cameras against a legend in the camera world: The Canon 1Ds Mark II.
Full Frame Legend
The Canon 1Ds Mark II was the second professional grade full-frame (35mm size) camera released by Canon. When it was released, people were astounded by its extremely high resolution and amazing image quality. When it was released, it retailed for $8,000 US. It is certainly not a new camera, having been released way back in 2004. Obviously, modern full frame cameras like the Canon 5D Mark II, Nikon D700 and upcoming Canon 1D X and Nikon D4 will outperform this body.
However, I view it as a benchmark camera for a few reasons. First, when it was released, the 16.7MP resolution was so far above the competition that it remained Canon’s flagship until 2007. Second, throughout the mid and later 2000s, this camera was used for a HUGE amount of professional work, from landscapes to fashion. The 1Ds II has seen photos published in the best photo magazines and has been used for gallery exhibitions by a large contingent of photographers. Finally, while current full frame cameras are better, it is easily safe to say that the image quality out of this beast is good enough for professional use and making large prints (I have personally made 40×30″ prints from this camera that look outstanding). If many of the best fine art and landscape photographers were able to use this camera for years, it should be good enough for the vast majority of photographers. In fact, the camera has an almost legendary status in the photographic community and many serious hobbyists and professionals alike still rely on this camera for their best work.
Now, before the flaming comes on in the comments, a few observations:
1. I know this test is not indicative of ALL shooting conditions. I tried to set it up so that it would give an indication of performance in a variety of conditions, but obviously there’s a limit to this. I do not have the time or inclination to re-run this test four or five more times under additional circumstances.
2. My conclusion is about image quality, not any of the other advantages here…I know full frame is capable of shallower depth of field and micro 4/3 is far more portable. DSLRs have better continuous autofocus (and Micro 4/3 has more accurate and faster single shot autofocus)…that’s not the point of this…mostly.
3. Make your own conclusions. I’ll present my thoughts on the data, but everything is here for you to look at.
To best test these cameras in a situation that I would best see their strengths and weaknesses, I had to decide on a setup. I chose to do a still life, lit by window light. I had an extremely overcast day, which ensured that the light would stay constant outside during the test, and the directional and somewhat dim light would ensure a high dynamic range would be present in the captured area, but I didn’t want harsh shadows. Indeed, when processing the RAW files, the dynamic range of the scene exceeded the available data for all three cameras, which was achieved by shadowing some areas and introducing glass and plastic surfaces which caused large areas of highlights to examine. I also made sure to include some blue items, some red items and I put my color checker chart to both provide a common white balance target and to see if any obvious color casts would be present. I didn’t expect much variance in color, however, as you’ll see next.
To put all cameras in a situation to succeed, all three cameras were processed using custom RAW profiles generated specifically for each individual camera, using the Color Checker Passport and Lightroom. In this way, each camera’s color should be very close to each other, or as close as they can be given each sensor’s performance. Since I use these for all my shooting, I didn’t have to do anything special here. Each camera’s shots were white balanced in Lightroom using the same square of the color chart, and all develop settings were identical. After RAW import, I recovered as much data as I could from each file, which included recovering shadows until they did not clip any more, (which was, using Lightroom 4 beta, +24 blacks on the GH2 and GX1 and +22 blacks on the 1Ds II). I also recovered as much highlight data as could be done using the Whites slider (-100). I used the automatic chromatic aberration removal in LR 4 beta for all shots (though neither lens has a significant amount of CA).
Each camera was set securely in the same spot, with each shutter fired using the 2 second timer, with mirror lockup on the 1Ds II. I used the camera’s internal meter at -0.3 EV. While this yielded images extremely close to each other, the GX1 shots metered very slightly darker than the GH2, and the 1Ds II metered ever so slightly lighter than the GH2. There ended up being about a 1/3 stop difference between the GX1 and 1Ds II shots, with the GH2 sitting 1/6 stop between them. Keep note of this, but ultimately, I don’t think it has much of an impact on the results. I didn’t do a straight conversion of shutter speed, f/stop and ISO as the rated ISO is different between different digital cameras, and since natural light was used, I needed the cameras to compensate for possible changes in outdoor light. While the outdoor light was consistent, I couldn’t guarantee against minor fluctuations.
I used the Panasonic Leica 25mm f/1.4 lens at f/5.6 on the GH2 and GX1, and I used the Sigma 50mm f/1.4 lens at f/11 on the 1Ds II. These apertures were chosen as they provide for adequate depth of field to evaluate more of the scene, and they yield identical depth of field for each shot. Since the sensor is smaller, with an angle of view crop factor of 2 compared to the 1Ds II, I needed to close down two stops on the 50mm lens to achieve the same depth of field. This is because I was shooting with double the focal length at the same distance. (Look up a depth of field calculator if you are confused). Also, since each camera is roughly 16 megapixels, these apertures have the same effect from diffraction, so should be an apples to apples comparison.
Next: The test…