ruthenium Offline Upload & Sell: On
|
amv8 wrote:
Just for giggles, I actually did a test last night. I photographed a uniform gray wall under a controlled lighting environment with my FF using a 50mm @ f4, 1/40s @ ISO 200 and then swapped to a m4/3 shooting 25mm @ f4, 1/40s @ ISO 200. Both cameras metered the same once I set the FF meter mode to "average across the screen" mode (my m4/3 didn't have that option).
The resultant raw files came out almost identical (maybe .2 ev difference) when looking the raw histograms.
What you observed is a well-known fact that a medium format, a FF, and a cropped-sensor camera at equivalent focal lengths all set to the same SS and f-number produce correctly exposed images (the ISO is calculated by the cameras and should be the same as well).
Question: Where is the catch?
Answer: These images are not equivalent.
Question: What does this mean in practice?
Answer: The images have different noise. The smaller is the sensor, the more noise there will be in the image. At low ISO, the dominant source of noise is the shot (photon) noise. This scales a square root of the signal. Since the number of photons that hit the FF is four times that number for the mFT camera (under the condition of equal f-numbers), there will be two-times higher (more intense) noise in the photos obtained with the latter.
Question: Why I don't see the greater noise in my example above? "The resultant raw files came out almost identical "
Answer: Because at low ISO the signal-to-noise ratio is uniformly large in a uniformly exposed image, and one simply doesn't see the noise. For the same reason, well-exposed images obtained at base ISO do not normally require denoising. The important point is that the noise that we don't see under some circumstances may becomes very visible and troublesome under different circumstances. Noise is a significant concern in photography.
Question: When should I see, and start worrying about, the greater noise in an image taken with an mFT camera set to the same f-number as a FF camera?
Answer: In two cases, concerned with low and high ISO, respectively. The first is when the ISO is low and the image has shadows that are lifted in post. Lifting shadows in the mFT image will reveal the higher noise compared to the FF image. Noise and dynamic range are the two sides of the same coin; in other words, the dynamic range of the mFT camera is relatively poor under the inequivalent conditions. To be specific, the Photographic Dynamic Range of OM-1 II is 9.69 at ISO 200 and that of a FF camera A7IV is 10.64. These a log-2 values; when converted, the dynamic range is 1.93 times better for A7IV vs OM-1 II, as expected. Note this has nothing to do with the size of the sensor, only with the amount of light that hit the sensors (four times greater for the FF camera under the inequivalent condition of the same f-number). At high ISO, the noise is immediately visible in the images. Thus, for example, an mFT image at ISO 3200 would have more intense (higher) visible noise than the FF image at ISO 3200. Note that equivalent mFT and FF images must have ISO(FF) = 4 x ISO(mFT), calculated by the cameras in auto-ISO. If the ratio is different from exactly 4, then the images were taken using non-equivalent apertures. I remember a video from Mike Lane where he said something to the effect that he was uncomfortable with ISO above 1600 on OM-1 for bird photography. Yes, this makes sense to me, as if we take equivalent photos using a FF camera, then the equivalent FF ISO should be 6400. Even if we can go higher, we normally wouldn't want to.
A summary in simple terms. The knowledge of photographic equivalence allows to tell when an mFT camera can theoretically take "the same" photo as a FF camera. This knowledge also allows to tell when this should be impossible. As simple as that.
The consideration of equivalence does not mean that when an mFT photographer and a FF photographer are side by side, taking inequivalent(!) photos, that the FF photos should be better. The image quality is going to be influenced by multiple factors - the optical quality of the lenses, the camera settings and speed, and by the post-processing skills of the photographers.
To the OP, if you are at a beginner's level, you probably are not going to get better photos with the FF kit. Go for mFT, but consider the Oly 300mm f4 lens which I compared with my Sony 200-600 mm lens on A1 and, as a result, stopped using the latter.
|