Full-frame DSLRs are bigger and heavier than smaller-format cameras, but that full-frame sensor does make a difference in terms of image quality. That’s largely because a bigger sensor can collect more light, and more light means less noise and thus better image quality. Today’s top pro DSLRs are all full-frame models, so that’s the way to go if you’re into action or low-light work, not just low-ISO image quality.
If you shoot at a specific shutter speed and ƒ-stop in a given amount of light, the exposure will be the same for any format; the same number of photons per square millimeter will reach the sensor. But the full-frame sensor is larger than smaller sensors, so the full-frame sensor will collect more total (total light = photon/unit area times total sensor area). And more total light translates to a better signal-to-noise ratio because noise increases at the square root of the photon count: If 100 photons reach the sensor, there will be 10 photons of noise, for a photonic signal-to-noise ratio of 10:1. If 10,000 photons reach the sensor, there will be 100 photons of noise, for a photonic signal-to-noise ratio of 100:1. Of course, there are other sources of noise, but photonic noise (which is carried by the light itself and thus can’t be reduced by technology) is the major one for normal photography. More light = more image quality. (This, by the way, is why high-ISO shots are noisier: When you shoot at a higher ISO, you use a faster shutter speed and/or a smaller aperture, which reduces the amount of light.)
Of course, there are some things to keep in mind here. If you shoot at 1/500 at ƒ/2 with a full-frame camera and 1/500 at ƒ/2 with a Micro Four Thirds camera, the same amount of light will fall on each square millimeter of sensor area. Since the full-frame sensor is almost 4X the size of the M43 sensor, nearly 4X as much light will fall on it. But if you use the same focal length on each camera, the full-frame camera will produce a much wider angle of view because the larger sensor "sees" more of the image projected by the lens. If you want to produce the same framing with each camera, you’ll have to use a lens half as long with the smaller format: 12mm on the M43 camera, if you’re using 24mm on the full-frame camera. Now you have the same framing, but the M43 image will have more depth of field because that depends, in part, on the size of the aperture, and ƒ/2 on a 12mm lens equals 6mm, much smaller than ƒ/2 on a 24mm lens (12mm). If you want the same depth of field, as well as the same framing, with both cameras, you’ll need to stop the 24mm lens on the full-frame camera down to ƒ/4. If you then want to maintain the same exposure, you’ll have to slow the shutter speed on the full-frame camera two steps, in this case, to 1/125, so you get the same image brightness at ƒ/4. That’s fine if your subject isn’t moving, but may not work with an action subject. Of course, you could instead increase the ISO setting on the full-frame camera two settings and shoot at 1/500 at ƒ/4, but then you’re reducing the amount of light reaching the sensor, so the image will be noisier.
This works the other way, too. If you want the limited depth of field produced by a 24mm ƒ/2.8 lens wide open on a full-frame camera, you’d have to open the M43 12mm lens to ƒ/1.4—but there are no 12mm ƒ/1.4 lenses for M43 cameras.
For a very thorough explanation of equivalences, go to www.josephjamesphotography.com/equivalence. In practice, if you’re shooting with a given format, you’re setting everything to get the shot you want with that camera, not thinking about what you’d have to do to get an equivalent image with another format. A full-frame camera at a given shutter speed and ƒ-stop should produce better image quality (better signal-to-noise ratio, technology generation being equal) than a smaller format at the same shutter speed and ƒ-stop in the same light (but with less depth of field if you frame the image the same from the same distance).
Full-frame sensors perform better in low light than smaller sensors, in large part because they can collect more light at a given shutter speed and ƒ-stop. In DxOMark.com’s sensor ratings, the top 29 scorers in low-light ISO performance are full-frame ones. The highest-scoring APS-C sensor ranks 30th, the best medium-format camera, 37th (and that through pixel binning, which drops its 40-megapixel sensor to 10 megapixels), and the best Micro Four Thirds sensor, 78th. Note that DxOMark.com hasn’t yet tested the medium-format cameras using the new 50-megapixel Sony CMOS sensor, which should fare far better than the CCD sensors in low-light performance. In overall score, counting color bit depth and dynamic range, as well as low-light/high-ISO performance, full-frame cameras occupy the top 14 spots and 18 of the top 21. The other three are 40- to 80-megapixel medium-format cameras.