Okay, did you put an SD card in the body and shoot some raws? Take the card home and extract?
marike6: Frankly, this review as a hachette job. I've always defended them against the claims of bias, but after this absurdly negative, nick-picky review, I feel zero compunction to defend anyone anymore.
Erroneous claims that the Df is the same price as the D800 and a high number of nit-pick Cons that were never mentioned for other reviewed cameras, i.e., 1/4000 max shutter speed, SD card on bottom, etc.
It could be that there are few if any DSLR shooters on staff but there are several extremely positive reviews of the Df on-line by top pros, see Sam Hurd, Nasim Mansurov, et al. But DPR couldn't be bothered talking about the the extremely quiet shutter, great battery life, or the fact that the Df beat the score of the legendary D3s the previous low-light champ on DxOMark. Instead they lamented the lack of time lapse mode. :-) The phrase "jumped the shark" comes to mind. Sorry guys.
And you're sure those reviews were "bang on" because you've used the Df extensively of course.
At ISO 12,800 the D800 has serious magenta and cyan blotching problems, this is not simple noise.
And no resampling does not remove noise, it makes the photo (and the noise) smaller.
For someone who claims to own a D800, you don't seem real familiar with it. It's a plenty good camera but has significant problems above ISO 6400, in fact lower.
The D800 never seems to have the dynamic range of the D610 or D4/Df either.
(the extended version of above)
Sorry I tried it and "downsampling" doesn't work, links aren't going to prove me wrong.
I read the links:
Two problems with the assertions in the second link: "Adaptive noise filters..." and " correlation". The problem with the first is that's an added filter, well okay, except that's not downsampling that's adding a filter. Then the second means a kind of non-real world event of course not usually captured by a photo. If a pixel really stands out against a smooth even background (non-correlation), yep chances are it's noise and a filter (that word again) can likely identify that noise, but that's an added filter that's not exactly in the Photoshop resampling options.
And did you not read through the comments from the first link?
So my results stand and my eyes and my testing with real world photos and PhotoShop CS6 don't misrepresent what is possible.
There's a reason noise reduction done well is hard.
Two problems with the assertions in the second link: "Adaptive noise filters.." and " correlation". The problem with the first is that's an added filter, then the second means a kind of non real world event of course not usually captured by a photo.
Saffron_Blaze: I am a wannabe professional, not a wannabe enthusiast retro shooter. Would someone please put the lovely sensor in a camera that supports all its capabilities?
"objectively superior"? Um, you do know the meaning of those terms right? DXO sensor scores are next to worthless.
I don't care about "resolution" and I don't imagine that you'll be printing many posters either. If I want "resolution", I'll use a Leica S2, much better lenses and sharp across the entire sensor plane.
There's no argument about high ISO lowlight shooting with the D800 versus the Df/D4+D3s, the latter 3 are all better.
I never wrote that someone shouldn't buy the D800, just that someone shouldn't buy a D800/D610 for the best high ISO lowlight results. And don't imply that I did so.
I don't own the Df. Nor do I love it. However it sure has some better performance characteristics than the D800.
You really don't seem to get this point: I'm not contemplating a full framed camera, I'm contemplating what full framed camera I'd buy for my purposes. If the D800 works for you buy it. It doesn't work for me, and if I had the money it's not the 2nd FF body I'd buy.
If the trick is easy to find, then provide the link.
No, I don't have to be a believer in a pancake earth, I've tried the trick and it doesn't work. That's taking a D800 raw, extracting, in ACR with NR set to zero, then halving the tiff file--with all six options in Photoshop. It doesn't help. Unless you mean that the pixels representing the noise occupy less space, but then so too do the pixels representing a clean signal; that's not NR--that's just smaller representations of the noise. It still be there, while it's less there starting with a file from the Df shot.
Now what can work is printing beyond 300ppi.
You have a strange concept of the math of averages. If there are a hundred pixels 90 red and 10 blue, then you halve that and maintain the averages you still have the same percentage of R+B at output time.
There's no debate: The D4/Df's sensor is much better at high ISO lowlight, low noise output, shooting than the not great sensor in the D800/A7r. And resampling down the D800 files doesn't get the results from the D4.
That reads like extra post processing, got a link? I've also seen it asserted that vibrance is easy to recreate if your photo editing software doesn't have the feature. It's not easy to recreate.
I've never been impressed with the DR of things I've shot with the D800. The D600/610 are plenty good though. I'm not exactly interested in a lab test for DR, I'm interested in results. And my results put the D800 down with the APSC D7100, while the D4, D3s, D600, Df all look a good bit better.
Yet again: It's not simply cyan and magenta "noise"; it's big blotches of both--so not normal noise, I'd be impressed if the trick works, but calling it "noise" kind of suggests we're talking about different things.
reinish: I am allways dreaming of Nikon FM2 to be reborn in digital age, and I tought that dreams has come true ..... but this DF is such a dissapointment, becouse of its size, it`s HUUUUGE, it looks like clasical film Nikon fed by force with leftowers from Tour de France.
Common Nikon , go and by Panasonic GM1 , dismantle it and learn how to make things compact
there they bouth are side by side
Well for starters: reduce the sensor size by three quarters, oh wait.
Maybe Nikon could improve the optical quality of its lenses instead of wasting effort chasing the tiniest. You see there Nikon has some catching up to do. And not simply to Leica and Zeiss.
Sensible1: It's carefully targeted people segment camera. They, rather few i believe, love it. For the rest it's crap. The Nikon Df.
Pray tell why do you think this Df “crap”? Be specific, and try to avoid adjectives in composing an answer.
Also, of course, you’ve used this body, right?
Frank C.: I hope Nikon learns from this episode and wises up unlike that other stallion of the Japanese economy, Tepco
Why? The Df appears to be an excellent light, quiet camera body.
Nikon hasn't told lies about its performance.
antoin: I'm sorry but this camera does not deserve a DPreview score of 81%. Using the DPreview compare tool and adding the Pentax K01, I can see that the DF has approx. 30% more "value" compared to the K01 -that's 30% more bang for buck. I bought a new K01 at a reduced price of AU$299. Compared to the DFit has shake reduction,full hd movies, inbuilt flash, albeit the K01 is an APS-C format compared to the full frame Df. The Dxomark sensor score for the K01 is 79, which puts it near the top of the APS-C rankings.Yet DPReview gave it a score of 69%, and comparing that figure to the 81% for the Df, well its a bit of a joke. Regarding the Df - "Disappointing AF performance drops off in moderate light", I would imagine this would be a deal breaker if you are trying to get the most out of its excellent sensor in low light situations.
If you don't care about the lack of a VF, the K01 is a good APSC box. But that sensor isn't close to the high ISO performance of the sensor from the D4.
MarkByland: Interesting review. I have questions about the low light AF performance of this versus the D600/610. With the same AF engine, why the difference in performance? Seems the 600/610 rated satisfactorily. Why the lack of similar performance in the Df? Were all AF modes tested? Phase detect, etc? Do they perform differently with older lenses under varied circumstances?
I'm not willing to fault this camera based upon retro appearance merit, alone. I understand its place in the market. I think the retro movement is capturing what got skipped over by all-things-plastic at the onslaught of digital. I also think full backwards compatibility with all F mount lenses is a major feat to have accomplished. The fold-away aperture control tab is top notch engineering, IMO. If some one were looking for the finest thing to hang their collection of Nikkor glass off the front of, they've found their calling.
I think it changed a while ago.
No down sampling does not remove noise, if noise removal were that easy there'd be little interest in sensors like that from the Nikon D4 or Canon 6D or Canon 1D X.
Instead of quoting something you read somewhere as established, test it out with your own noisy raw. Extract to tiff with NR set at zero, then go thru the 6 downsampling options in Photoshop. None really help with noise control, all make the picture smaller though.
Not "simple math", but a made up model based on little understanding of NR.
If the D610 servers your purposes, by all means use one.
Boring, the sensor in the D800 doesn’t have particularly good dynamic range.
And as you know the D4's sensor significantly outperforms the sensors from the D800 and D610 in one particularly important area.
If the D800 serves your purposes by all means then shoot with one.
You clearly don’t know much about high ISO lowlight shooting. And one of the things this sensor allows for is much faster shutter speeds with zoom lenses, so that has an obvious advantage beyond what would normally be called “lowlight”.And it aint simply “magenta noise” that’s a problem with the D800 above ISO 6400, it’s magenta and cyan blotching. Since you clearly don’t understand this point: Blotches are much harder to remove than high gain noise.
Right, that "whole package" D4 costs a good bit more than this quiet+light body with the same sensor.
Right, well that's a place to use film+big negatives.
With some MF lenses (newer Zeiss) and manually focusing with an AF lens, the focus indication on the Df is a red flash across the view finder.
I remember some thing like you describe with perhaps the D3s body.
Thank you for providing an example of where a Df would be useful and where a D800 has some limitations.
plasnu: I mostly agree with the review. This is not a photographer's camera, anyway. Instead, this is for CAMERA enthusiasts after all. Or for someone really wants that D4 sensor.
Have you shot with the Df?