D600 is out

@Ted Spencer,

Yes, nobody has done an equivalent exposure comparison yet. Would be interesting to see how they both deal with same illumination and shadow to highlight detail. Rather than ISO to ISO.
 
Overall, the Nikons are better performers than Canon 1st Gen, and behind the 5dmk3 Canon in some integral video features and in camera video gain noise processing. But, I'd use a D600 over a 5dmk2 any day.
 
@Ted Spencer,

Yes, nobody has done an equivalent exposure comparison yet. Would be interesting to see how they both deal with same illumination and shadow to highlight detail. Rather than ISO to ISO.

There's a video clip of D800 vs 5DII illustrating and compensating for exactly this that I saw recently. I think I found it here in a D800 thread. No time to look for it now. Maybe someone else can link to it here.
 
thanks for those links, but... wow, that's a lot of clips to watch!!

let's see...

* can't judge sharpness or aliasing/moire or codec issues with youtube clips
* low light seems great (http://www.youtube.com/watch?v=tR7f78RSJWw&feature=channel&list=UL) though not as great as in the 5D3 (http://www.youtube.com/watch?v=K8KIBrCvxy0&feature=player_embedded)
* some clips don't have that contrasty look of the D700, and are closer to what I'd expect from the D800 (http://www.youtube.com/watch?v=j-lyn0s1af8&feature=player_embedded)
* AF sucks (as expected) (http://www.youtube.com/watch?v=U51OYv5_wAA&feature=player_embedded)
 
So far it seems like the D600 moires a lot, but the higher resolution means it manifests itself in different ways than I normally deal with on a 5D2.

J9Ss3.jpg


Look at the guys on the left or right side. At first glance I thought there was noise on their shirts (I shot this with flaat 12), but when I actually watched the video there is moire on their clothes fabric. It's fairly under control though - you almost don't notice with the jacket, and you notice with the blue polo but it's not like there's huge wavy lines running across, it looks like an odd noise pattern.
 
So far it seems like the D600 moires a lot, but the higher resolution means it manifests itself in different ways than I normally deal with on a 5D2.

J9Ss3.jpg


Look at the guys on the left or right side. At first glance I thought there was noise on their shirts (I shot this with flaat 12), but when I actually watched the video there is moire on their clothes fabric. It's fairly under control though - you almost don't notice with the jacket, and you notice with the blue polo but it's not like there's huge wavy lines running across, it looks like an odd noise pattern.
hmmm... don't know if I like that. Do you have a video version online or downloadable?
 
Overall, the Nikons are better performers than Canon 1st Gen, and behind the 5dmk3 Canon in some integral video features and in camera video gain noise processing. But, I'd use a D600 over a 5dmk2 any day.

I think we should factor two things. Firstly te Canon 5d mark 3 is about 2/3 stop less sensible than the Nikon (at least the D800), so Iso 2500 is more like Iso 4000 on the 5d3 compared to the d800. Secondly the ISO has to be tested at same output resolution. This is why DXOmark numbers is calculated for a 8 megapixel image. So if you normalise the two resolution the D800 vs 5d3 are much closer because you can apply a lot of noise reduction to get to the same resolution. This is what the nofilmschool example shows.

The mythical Canon 5d3 low light quality is in fact a lot of in camera NR that is inbuilt in the camera to the detriment of detail, the high ISO looks more like SD resolution. This can be traced to the photo side. When it came out Canon said that the camera was like 1.5 o 2 stop better in low light. But it was only in jpeg with lot of in camera NR (loss of detail etc) and in RAW it was barely better the the 4 year old 5d2.
 
There are a lot of reason why I thought the D800's image was (much) more pleasing than the 5DIII in the various camera comparos I've seen, and heavy amounts of in-camera NR on the Canon would seem to explain some of them. While the Canon did look *way* less noisy, especially at *very* high iso's, it was quite a bit softer at all settings, and had a muddier look even when exposures were closely matched by eye (which meant raising the Canon's iso almost a stop above the D800s). So I'm pretty comfortable with the tradeoffs with my D800: superior image quality in all but conditions requiring iso's above 1600 or so, where 3rd party noise reduction solutions (Neat Video etc.) can greatly extend that. As I continue to use the camera, it will be interesting to see how many occasions I find myself wanting to shoot video above 1600 iso...
 
superior image quality in all but conditions requiring iso's above 1600 or so...

As someone who has actually used both cameras and not just youtube'd them, I disagree slightly. The 5dmarkIII is slightly better in video, in all circumstances. Compression is better, no aliasing at all, maybe a smudgeon not as sharp (very small difference though, I attribute it to aliasing mostly on the D800). 720p that is actually usable. It is not with the D800.
 
Fair enough. I've not actually had my hands on a 5DIII, or any of the 5D series for that matter.

Some of the clips I saw were accompanied by commentary that supports much of what I asserted however...

Both great cameras though, for sure.

As someone who has actually used both cameras and not just youtube'd them, I disagree slightly. The 5dmarkIII is slightly better in video, in all circumstances. Compression is better, no aliasing at all, maybe a smudgeon not as sharp (very small difference though, I attribute it to aliasing mostly on the D800). 720p that is actually usable. It is not with the D800.
 
Back
Top