LUMIX GH3 REVIEW PART 5 FIELD OF VIEW
GH3 has less than GH2 and G5
Author AndrewS April 2013
Introduction Sometimes, in the course of testing photographic equipment, one encounters an unexpected observation. While testing the Lumix 35-100mm f2.8 lens, using the Lumix G5, GH2 and GH3 cameras, I noticed that test chart pix with the GH2 and G5 included a larger area of the chart than those taken with the GH3.
The test I use a simple kind of test chart which is useful for comparing some aspects of the image quality of one camera or lens with another, but does not give absolute readings for technical matters such as resolution. I run the test chart pix in 4:3 image aspect ratio. I found that with the same lens at the same focal length on the same tripod at the same distance from the chart, the horizontal field of view [HFOV] given by the GH2 and G5 was 2.5% more than that given by the GH3. I re ran the test with different lenses at different focal lengths but always came up with the same result. I used zoom lenses at the end of the zoom range to ensure there was no change in focal length from one test to the next. Unfortunately I did not have an Olympus OMD available for comparison testing.
 |
| Same lens, focal length and distance of focal plane from the test chart. The G5 and GH2 include 2.5% more horizontal distance as measured on the chart than the GH3. In this case I used the 35-100mm lens at 35mm. Actual HFOV measurements were GH3 1026mm, G5 1053mm, GH2 1050mm. Viewfinder and Monitor view appeared to be 100% of the final output in each case |
1. For any lens at any given focal length and focussed distance the GH3 gives a slightly smaller HFOV than the GH2 and G5.
2. The actual imaging area of the GH3 sensor has to be smaller than that of the GH2 and G5. Note that we know the total size of the GH2 sensor, also apparently used in the G5, is larger than most other M43 sensors because it was developed to be used with the multi aspect ratio facility of the GH2. I ran all the tests at 4:3 aspect ratio which I had expected would give the same HFOV with any M43 camera. Apparently not.
Published Data The following tables have been compiled from various data sources, mostly the manufacturer or elsewhere as noted.
Sensor dimensions in millimeters and megapixels
Model | Nominal sensor dimensions, mm | Stated total M pixels | Stated Camera effective M pixels | Image size in pixels @ 4:3 AR |
GH3 | 17.3x13 | 17.2 | 16.05 | 4608x3456 |
G5 | 17.3x13 | 18.31 | 16.05 | 4608x3456 |
GH2 | 17.3x13 | 18.31 | 16.05 | 4608x3456 |
GX1 | 17.3x13 | 16.68 | 16.0 | 4592x3448 |
EM5 | 17.3x13 | 16.9 | 16.1 | 4608x3456 |
* All the sensors are given as the same nominal physical size of 17.3x13mm which is presumably not literally the case.
* The GH2 and G5 appear to be using the same sensor.
* The numbers don't add up. 4608x3456 = 15925248, not 16.05 or 16.1.
* Camera data gives the pixel dimensions for 4:3 AR as 4608x3456 for the GH2, GH3, G5 and EM5.
* DXO Mark gives a slightly different set of numbers. [dxomark.com]
Model | Pixel Pitch | Camera effective pixels | Resolution as pixels |
EM5 | 3.73 | 16.11 | 4640x3472 |
G5 | 3.74 | 16.05 | 4624x3472 |
GH3 | 3.74 | 16.05 | 4624x3472 |
What can one make of all this ? Probably that the technology is beyond my limited grasp.
Logically it seems to me that the pixel pitch of the GH3 should be slightly less than that of the G5 but they are given by DXO as the same.
My observations remain however, namely that the GH3 gives a bit less HFOV with any lens than the GH2 or G5.