Nikon D800 vs Canon EOS 5D Mark III JPEG RAW noise (RAW down-sampling on next page)
Support me by
shopping below |
|
|
|
|
|  |
To compare RAW noise levels under real-life conditions I shot this scene with the Nikon D800 and Canon EOS 5D Mark III within a few moments of each other using their RAW modes and at each of their ISO sensitivity settings.
The D800 was fitted with the Nikkor AF-S 24-70mm f2.8G ED, while the Mark III was fitted with the Canon EF 24-105mm f4L IS USM, both with the aperture set to f8. Both were using their standard processing styles and White Balance was set manually to 3300K.
Auto Lighting Optimiser and Active D-Lighting were disabled. See the contents list above left for all my results pages.
|
The image above was taken with the Nikon D800 with the Nikkor AF-S 24-70mm f2.8G ED lens set to 40mm and the aperture set to f8 in Aperture Priority mode. At its expanded Low sensitivity of 50 ISO, the D800 metered an exposure of 5 seconds for this composition. The Canon 5D Mark III metered an identical exposure, so you're comparing like-with-like below.
I processed both sets of files in Adobe Camera RAW using identical settings: Sharpening at 70 / 0.5 / 36 / 10, Luminance and Colour Noise Reduction both set to zero, and the Process to 2012 with the Adobe Standard profile. These settings were chosen to reveal the differences in sensor quality and isolate them from in-camera processing. The high degree of sharpening with a small radius enhances the finest details without causing undesirable artefacts, while the zero noise reduction unveils what's really going on behind the scenes - as such the visible noise levels at higher ISOs will be much greater than you're used to seeing in many of my comparisons, but again it's an approach that's designed to show the actual detail that's being recorded before you start work on processing and cleaning it up if desired.
If you're viewing this page in isolation I'd urge you to have a copy of my Nikon D800 JPEG noise results open at the same time - and just to be extra helpful, you can click this link to open them in a new window.
Image processing can be a very personal thing, but like my outdoor test results I'd say both the D800 and 5D Mark III benefit from the RAW workflow described above. Both are displaying very crisp fine details without any undesirable haloing artefacts. Again like my outdoor tests, the Mark III benefits the most, with the RAW conversion eliminating the contrast-heavy approach of in-camera JPEGs using the default settings. It's really brought out the finest details, and now the 5D Mark III is showing the cross-hatched pattern in the egg shapes at the top of some of the organ pipes; sure it's not as clearly defined as in the D800 crops alongside, but it's a darned sight clearer than the JPEG comparison.
With noise reduction completely turned off, it's not surprising to find a smattering of grainy textures appearing at low sensitivities: indeed pixel-peepers will just about spot it at 200 ISO, and more noticeably at 400 ISO. At 200 and 400 ISO this graininess is ever so slightly higher on the D800 than the Mark III, but it's so close it's hardly worth mentioning.
What's much more important now though is scrolling down throughout the sensitivity range and noticing how both cameras appear to share very similar noise characteristics at each ISO value when viewed at 100%. Indeed all the way up to 25600 ISO the noise levels appear pretty much the same.
This is a very revealing result. On the previous pages we saw how the Nikon D800 enjoyed a detail advantage at low ISOs, but lost out to the Mark III on noise levels at higher ISOs. What this page reveals is both cameras are actually starting with roughly similar degrees of noise when viewed at 1:1 and the Canon is simply applying greater (or more effective) noise reduction on its in-camera JPEGs than Nikon. Turn that noise reduction off though and both cameras are essentially delivering the same amount of speckles. Crucially though, the higher resolution of the D800 is still managing to capture slightly finer details than its big rival.
This is a great result for the D800 which amazingly manages to pack 50% more pixels onto its sensor than the Mark III without any apparent compromise in noise levels. It also illustrates the importance of performing comparisons of RAW files with noise reduction disabled to see what's really happening behind the scenes.
So a draw in noise but a lead in detail makes it a win overall for the D800 on this page. It also begs the question of what happens if you then take these files from the D800 and down-sample them to the same resolution as the Mark III - and that's exactly what I've done on the next page in my Nikon D800 RAW noise down-sampled results!
Canon EOS 5D Mark III (RAW using Adobe Camera RAW)
Using EF 24-105mm f4L IS USM at 40mm f8 |
|
Nikon D800 (RAW using Adobe Camera RAW)
Using AF-S 24-70mm f2.8 at 40mm f8
|
|
|
|
L (50 ISO) |
|
L (50 ISO) |
| |
|
|
|
|
|
100 ISO |
|
100 ISO |
| |
|
|
|
|
|
200 ISO |
|
200 ISO |
| |
|
|
|
|
|
400 ISO |
|
400 ISO |
| |
|
|
|
|
|
800 ISO |
|
800 ISO |
| |
|
|
|
|
|
1600 ISO |
|
1600 ISO |
| |
|
|
|
|
|
3200 ISO |
|
3200 ISO |
| |
|
|
|
|
|
6400 ISO |
|
6400 ISO |
| |
|
|
|
|
|
12800 ISO |
|
H1 (12800 ISO) |
| |
|
|
|
|
|
25600 ISO |
|
H2 (25600 ISO) |
| |
|
|
|
|
|
H1 (51200 ISO) |
|
51200 ISO not available |
| |
|
|
|
|
|
H2 (102400 ISO) |
|
102400 ISO not available |
| 
Tweet
