The GF5 versus GF3 versus G3 comparisons are starting. What will be the results?
Well, taking Panasonic on their word (the GF5 sensor is derived from the G3) we can make a few basic guesses, without even testing the cameras.
Warning: the following applies to raw files. Because of the new JPEG rendering schemes in the GF5, all bets are off for JPEGs.
- A GF5 and G3 will have the same amount of noise overall. Why? Because they have the same capture area and use the same technology, so the same amount of light falls on the sensor and the amount of that gets converted into image data is basically the same. One of the myths of digital is that more pixels = more noise. Not exactly. On a per pixel basis, yes, you might come up with different measurements. But that's not what's important. If we're going to create an 8 x 10" print from raw files from the GF5 and G3, we should get the same amount of noise in the print: same capture area, same sensor technology, same magnification. There aren't any unequal variables here, so the results should be pretty much the same.
- A GF5 will be slightly better than a GF3 overall in low light (high ISO values). Why? Because the efficiency of the G3 (and thus GF5) sensor is slightly higher than the GF3 sensor and the read noise is definitely lower. The unequal thing here is the sensor technology, which has moved forward. Not dramatically forward, as the Panasonic sensor tech doesn't yet match what's happening with some Sony/Nikon/Canon sensors in terms of efficiency and read noise, though they're getting closer.
Bottom line, I'm not expecting anything dramatic at the raw level. A GF5 and G3 will likely have the same amount of noise overall at base ISO. A GF5 will be slightly better than a GF3 at higher ISOs. But unless you're stretching your output to or past the maximum size 12mp implies (15" print), I doubt you're going to see a lot of meaningful difference in raw files.
As I noted above, JPEGs may be a different game. I'm keeping my fingers crossed that Panasonic has upped their JPEG rendering game considerably. For every Panasonic camera so far, the JPEG rendering has been behind what Olympus has gotten from the same sensor. It's actually my biggest complaint about the Panasonic m4/3 cameras so far: they under deliver on JPEG quality. It's easy for me to take a raw Panasonic file and make a better looking JPEG than I can get out of the camera. It's difficult for me to do that with an Olympus raw file. Given that a lot of the mirrorless shooters value JPEG results, that's been one of the defining differences between the Panasonic and Olympus m4/3 cameras to date. Let's hope the gap has finally been closed.