Sony A6000 vs Sony A6300 vs Fujifilm X-Pro 2 noise
To compare noise levels in low light I photographed this scene with each camera using their full range of ISO sensitivities. I've included Fujifilm's X-Pro 2, another 24 Megapixel APSC mirrorless camera, not because it's targeting the same buyers but because the Fuji sensor will almost certainly be featured in future X-bodies aimed at the same audience as the A6300.
Both the A6000 and A6300 were fitted with the same Zeiss 16-70mm f4 zoom, set to 19mm f5.6 in aperture priority for the optimal quality. I fitted the X-Pro 2 with the XF 10-24mm f4, zoomed to 19mm to match the field-of-view and also set to f5.6 for the optimal quality. I used this lens on the Fuji as it was the only one I had available at the time of testing with a matching focal length. I realise it's higher-end than the Sony lens, but I decided to include it here out of interest. The cropped area is taken towards the middle of the frame where both lenses perform well.
Each camera was using its default processing style and the shutter speeds were matched for each sensitivity - so what you're looking at below is directly comparable. The full view is shown below with the red square indicating the cropped area, presented below at 100%. I've compared out-of-camera JPEGs here but also recorded the scene in RAW for a future comparison when all the cameras are properly supported in Adobe Camera RAW.
As we saw on the previous page, the Fujifilm is applying slightly higher sharpening to what was arguably slightly crisper output from the sensor to start with, resulting in a punchier-looking set of crops. But while initially attractive, if you look carefully you'll see evidence of over-sharpening with edge artefacts which you may or may not find appealing. The bottom line is you can tone-down the Fujifilm sharpening or apply more to the Sony images either in-camera or on RAW files, and achieve a more similar-looking result if desired.
As for actual recorded detail and visible noise, I'd say all three cameras are fairly well-matched up to 1600 ISO. At 3200 ISO and above though, all three cameras have to deal with noise and I'd say all the way up from here to 51200 ISO the Fujifilm does the best job using the default settings. Sure its approach may result in more visible noise, but I personally prefer the retention of detail rather than smearing it out.
As for the two Sonys, I'd say the older A6000 is suffering a little more than its younger sibling, with rougher edges and patchier noise reduction. From 3200 ISO upwards I'd say the newer A6300 enjoys an advantage in quality of around one stop over its predecessor - so at 6400 ISO for example, the A6300 looks similar to the A6000 at 3200 ISO. It's certainly not a standout difference and I wouldn't upgrade from an A6000 to an A6300 based on this alone, but there are of course many other advantages to the new body over the old one. The good news at least is the addition of even more embedded phase-detect AF points on the A6300 hasn't compromised its quality over its predecessor - judging from my two pages of results, the A6300 essentially matches the A6000 for resolution and slightly out-performs it above 3200 ISO for noise and retained detail. Not a bad result considering more of the sensor pixels are devoted to AF duties than before.
Now check out my Sony A6300 sample images to see how it performs under a variety of real-life conditions or skip to my verdict.