Nikon D750 real-world samples gallery posted

I agree with Greg about the 85mm f/1.8 Nikkof. It s a great lens and compared to the cost of the f/1.4 version, a bargain.

I see it in Raw conversions using Capture NX 2. There's a D800 ad shown in print magazines of a woman standing in a forest with a dress that has a forest print on it. Interestingly, this image has exactly the same issue. All the greens look the same, when there ought to be a lot more hue variation.

It matters a great deal, and is used by manufacturers to get you to buy yet another DSLR. As is Face recognition. As is Shutter lag. As is a partial (i.e. D800E) removal of the low pass filter. As is a total removal (i.e. D7100 and D810.

When you put together all of them it becomes obvious we are being taken for a ride yet again, since the noise which at 800 and above is BAD on a D600/D610, is a sure way of getting you to buy a D750.

But try waiting now until they "give" (Ha!Ha!!) us all in combination, something they have been able to do for years.

36MP in a small D7100 size body, low noise, face recognition, no filtration of any kind , a twist and flip screen that folds into the body for protection, and a vertical grip that is the same as the model before it!! Fat chance, but HowaboutMORE!!

munro harrap:

Why do you think I own a D610? Or any DSLR?

The fact remains that the D750 is a much better high ISO body than the D610 and has much better AF. (Raw, the D610 is easily useable above ISO 800. Also optically better lenses always help with higher ISOs.)

Why the request for more MPs in an APSC body, what on earth makes you think I chase higher MP counts?

Now speaking for myself, the D750 is one of few DSLRs that have interest for me, others would be: The D700, the D4s, the D3s, the Df and the Canon 6D. See a pattern?

I had two,refurbished from Nikon. The first went back because although it turned on, it refused to take a photograph in P,A,S,and M modes...
It was replaced with a second, which went back, although it worked fine because pictures had shadow noise at 800 and above that exceeded in size detail resolved,and dont ask for a sample as My Gallery identifies them as having more than 30,000 pixels on an axis (obviously impossible) with the best noise reduction I can do, and says error and will not upload them.
Under the same conditions a D800 produces noise too, but it is less than the detail resolved.
I believed the reviews, unfortunately, but tests here and elsewhere are of an incredibly boring evenly lit tatget, not at all real world stuff, so you cant tell.

http://www.dpreview.com/galleries/9632689308/photos/2471113/_dsc6788

6400 1/100

Hahahaha! You are a funny Guy :-)

I was thinking of Nikon 85mm 1.8G. I have read that since this lens lacks an aspherical element, it suffers a little bit from the focus shift. I do not know how sever it is though. Have you had such a problem with this lens?

Lovely, Rishi, thanks! You restored my peace of mind! Compliments, Hermann

Rishi wrote: "Yes, one should never, ever break rules in photography, or any art-form for that matter."

Did I notice a hint of irony? Of course, I am fully aware of those artists of genius who deliberately broke the rules creating new exciting things and eventually setting new rules by themselves.

It is just that a horse never should be photographed using a wide-angle lens ;-))

I appreciate your work, thanks. Hermann

Thanks @Rishi,

I was going off the lead shot for the gallery and a similar shot of the same horse and rider on Ami's IG profile. Hope you had fun in PR!

@Rishi LOL! Check out my Flickr account for some shots that were obviously taken by Henri Cartier-Bresson.

Ellis! You're awesome. Was great to meet you after having followed so much of your work in the past, and I learned a lot from the stories behind your images. :) Look forward to future encounters!

@Rishi, I haven't really use my D300 Auto Area AF mode but I thought it would detect faces as the manual says. So it's good to know it's really not that good and reliable and maybe I should leave it alone. Thanks!

@Rishi
With this advanced AF system, I got eager to give it a try. Can it focus with depth of field preview active? This is the only way I can think of to avoid the heinous focus shift problem.

Yes, there is no face detection in AF in the D600, except in Live View. I have retyurned mine because of this and due to a second sample having the most appalling high ISO noise. I have repeatedly tried to upload samples to Gallery here but error code comes uip saying they are more than 30000 pixels along any axis, so Rishe, even if I wanted to demonstrate how to use a wideangle etc to you, I am at present unable, but if you check out all figure and fashion photographers, their stuff might be cropped and ruined in mags like Vogue, but in the originals the bodies are all there. I am a basic reportage photographer and work often in chaotic crowded situations, but something always brings me up short and rejects any picture where bits are cropped off. Your stage pictures crop people unecessarily as well. With all that space there is no excuse. Take em, but dont show 'em!

Next time I want the narrow opinion of your full-bodies-only, no equestrian distortion photography, and my name mispelled, I’ll ask. Meanwhile, others continue to enjoy the shots. I’m sure you think they’re all wrong, and you’re right. I believe there’s a term for that...

Also, let me know when 'I don't know how to use the internet' becomes a valid reason for not showing any of your own work while merrily criticizing others'.

You take whatever pics you like , dont let me stop you, but I have tried to upload from here and I keep getting an error message about 30000 pixela along any axis-from Dpreview. Now how a 24MP image CAN exceed 30000 pixels along any axis I know not!! It does not say 3000, it says 30,000. I am a duffer on the internet I agree, but it is not me stopping me uploading right now. I shall go now to my My Gallery and try again. Meanwhile can you let your site chappies know the 30000 pixel score? just in case.

Yep, IT is still doing it. This deprives you, not me, so for all our sakes, please get it fixed

Dear Scott, There is a strong demand to see on how a D750 behaves. Journalists have the job to discover things quickly, uncover, and come to the point. I am only the messenger. It is up to you to listen to your customers and the competition. And create the Online media to the markets, your corporate goals, and purposes.

If this gallery, presented here, is ok or better in your opinion; it is your decision as Product Manager. One should it calm down, now. Markets will decide.

Ken's shots (1) don't show low-light image quality, (2) don't exemplify "mural-size" image quality. I have nothing against the camera, only I'm not sold on it -- not for a nut of $2300 plus lens. Ken's review is fine, only it's too short.

No axe to grind with KR but the shot seems to mostly highlight the cameras HDR capability. Are those wavy lines at the top of the red brick buildings old signs or moire'

KR always shoots in jpeg basic. He says all pros do.

His family shots are almost always in small-basic and landscapes in large-basic.

http://www.kenrockwell.com/tech/raw.htm

"I never shoot raw. Why would I? Raw is a waste of time and space, and doesn't look any better than JPG even when you can open the files."

one zero too much:) ......... but of course, if I could afford I would take FF, no doubt, but if I could afford MF would be even better:)
Apropos, what is the point of shooting with 20mm at F2.0? anyway I still cannot figure out (except commercial) the concept of D750. I would be happier if Nikon boys could get rid off that peach skin tone;)

You could, yes, but do realize that you'll run out of Raw buffer in something like 2.3 seconds, according to Nasim Mansurov.

The D810 buffer will last you substantially longer.

I actually kept hitting the limits of the buffer in my shooting with the D750.

But for many, 2.3s of burst shooting will be fine.

Rishi Sanyal:

The point remains that the raw buffer on the D750 is significantly better than that on the D7100. The D7100 would be hard pressed to shoot more than about 5 raws in 2 seconds before the buffer slows down a great deal.

The D810's raw buffer is a whole different story.

@HowaboutRAW

I think you missed the point. I do not have to shot RAW all my life and then spend time on PP. If I pay a lot of money for a camera then I expect the picture coming out exactly as I want.
If you are unable to set the camera to have 95% pictures as you want, then the camera is rubbish. In this regards Nikon is rubbish as I cannot tweak the picture in the camera to get the results as I want. It is only possible to do via computer and then bring the profile back. This is what I do and I shoot jpgs and I am a happy bunny. If I go for something special then I consider RAW. You see there is plenty of people who do not know how to set up the camera so they shot RAW but there are others who know and they are happy with JPEGs. We have XXI century and the cameras shall be better than phones:) all the best:)

well, you probably know everything:) but you look like a PP-boy not a photographer or you are joking?;) perhaps .... all the best.

... and thanks, I prefer vinyls dude:)

I think you got too hot, calm down.
Remember, M is not for Mature, A not for Amateur and S not for Stop just learn how to take light reading and forget about RAW but I assume you use RAW as you have no clue how to take itl. More photography less PP and happy trolling ;)

SkilakDeZoo:

Yeah sure, I'm annoyed.

But rubbishing cameras that don't do what you want without any adjustment to the file later is silly.

What on earth does: "but I assume you use RAW as you have no clue how to take itl" mean?

how to measure the light, or exposure:) sorry I was spaced out a bit;)
10 years ago I went through RAW with D70, I did figure out the settings for JPEGs, everything was OK till D7000 :( D7100 is not any better so 10 years later where memory is supper fast and cheep as potatoes I have a right to expect more???? Apropos, in music I only care about what I can hear :) 3 Hz or 30 000 Hz not bothered :) I guess you are not one of those who thinks that 90dB + 90dB is 180dB?;)

What on earth are you talking about?

Raw is really helpful.

Also from audio are overtones and undertones, and higher resolution files if digital.

Again, I think you do not understand or you do not want to understand:) RAW is helpful but not a must. It is a mean to help you to figure out the settings in your camera and then you can shot with smaller size files. Is it so difficult to catch?
"overtones and undertones" what the f... is that??? subjective acoustics? Mate, tone is a pure sound power at particular frequency. The only thing you could speculate in this field is the sampling method and accuracy. I think you better stick to photography as acoustic is not your strong filed;) unless we are talking about audio acoustic for fan-boys?:) by the way Technics is back, cheers

SkilakDeZoo

No, raw is not:

"It is a mean to help you to figure out the settings in your camera and then you can shot with smaller size files."

This claim is simply wrong.

You can look up overtones and undertones, it's vinyl thing, or more advanced digital music play back. You really should know this.

if I was not an acoustic engineer I would. If a group of people developed those terms for their needs then OK. Acoustic is pure and cruel physics unless you want to enter into subjective acoustics where exactly those terms, I think, belong, unless we are talking about turntable, stylus and vinyl own imperfections;) but I think we have diverted from the main subject;) cheers

SkilakDeZoo:

No what I'm talking about isn't only a turntable thing.

And the terms overtones and overtones are very real physics of sound.

Figure it out.

.. of sound perhaps yeas as a bunch of harmonics but not for a tone itself.

SkilakDeZoo:

Harmonics that digital audio engineers ignored for years is part of it.

By the way, do you think Sony will abandon A mount in future in favor of the FE mount or this new A77MKII is a good sign for future of A mount. Actually, yesterday, when I was watching "The Future of Photography", I was hoping you ask the Sony guy this question.

Two questions then : is the lossy RAW file of the A7 an issue? Can you recover shadows without introducing lots of noise?
(2) Is PDAF consistent? I mean if I employ the AFMA technique that you use, can I always have the eye in focus at f1.4 assuming no hand/camera shake? As you correctly mentioned, I manually focus because I do not know of a reliable and consistent AF system. If I am introduced to one, most probably, I will go for AF which is of course more convenient.

Lossy Raw is not ideal, and is something we hope Sony eventually addresses. That said, I've only noticed it at very harsh contrast borders, and it hasn't generally been a huge problem. Here's an example from the A7S; download the Raw file, and open up the shadows or raise the exposure, and look for areas around the arches (bottom left) or Space Needle. You'll see compression artifacts in the skies around these structures.

As for recovering shadows without lots of noise, I assume you're talking about Raw dynamic range at base ISO. Here the A7R and A7 are very good (like the D800), and the A7S is about a stop worse (it has noisier shadows than the A7/R). However, the A7S is still considerably better than, say, any Canon DSLR in this regard; its Raw DR is more on par with what you might get with a Nikon D4S. Still very good, just not the absolute best. And still far better than some of the competition.

Thanks for the information that you shared with me.

If you want to manual focus, you are best off getting a A7-series camera with Zeiss Loxia lenses. Try it and you will see why.

@ Rishi Sanyal

By the way, which one do you prefer for AFMA; FoCal or Lensalign FocusTune?

Yeah, exactly. Our studio scene is better suited to looking for tiny, pixel-peeping differences in, for example, ISO performance.

I did notice that the really cool one with the horse on the beach has motion blur. I think the shutter speed was too low.

Thanks for the comments, Rishi. Just for comparison I viewed the D810 samples on the same monitor/PC/browser and they're very sharp. Maybe more of a camera thing I guess, or the lens and aperture (I looked at mostly center of the images). Great images BTW.

What a bunch of sh.. I do never listen to what DPR says, i note there objections and finds, then I look the samples, that I often criticize, since some people there have a bad taste and often the shots are of focus. I get a camera, rent one, and then test it myself and make up my mind. I see then if DPR's criticism is justified, and in more than 95% of all critics it is. After that I shoot my galleries, that are then most much more sharp, show the subject that fits me, and I rework Raw my way. After that, I can say if a camera fits me or not, despite all what just any site or magazine can say aboput it. I shoot MF only, that is why I use Sony's mirroless bodies and a Fuji S5 with a split prism matte in it. Some of those pictures are quiet stunning. Rishi has a way to shoot from very unusual angles of view, probably part of his shooting style. But after, all, good work and tks for the excellent rewiew of the D750.

I totally agree with 57even. .....Notice however that I am far away from being an expert.

Hi 57even, We got quite an emotional debate. This is more refreshing instead of discussing the best location of pixels. You lead controversial threads which brings much needed life into action. Thank you.

And Rick got an impressive set of lenses I found appealing by the smart composition. It showed devotion and money spent into a good direction. Have a nice weekend.

Hi Thomas,

Thankyou.

And a good weekend to you too.

"It would also be nice if DPR would put the camera up on a tripod for a few shots and try to go for optimal IQ."

Every now and then, DPR posts a shot of the Seattle skyline, rich in edge and corner data. I like that. Gordon Laing of Camera Labs often offers similar kinds of scenes.

Just sayin'.

And a good weekend to everybody!

Pat

And as for "utilizing all of the measures landscapers employ to maximize IQ." -- that's exactly what'd mitigate differences between systems. For example, if I had a smaller sensor which I knew would produce noisier shadows b/c of photon/shot noise, I'd definitely use a graduated ND filter or multi-shot noise averaging in a high DR scene. Something I wouldn't have to do with, say, my D810. So what compensatory technique should I use for what camera to show it 'in its best light'?

That's the whole reason we develop controlled tests. You strip away variables, variables like changing light, changing scenes, different days, different exposures, etc. etc.

I realize the limitations of our studio scene.

Let's try this - what exactly is it you wish to know? About, say, the D750?

Also, the sunrise shot was taken on a tripod, with a 4-stop Daryl Benson reverse graduated ND filter. The sunset shot from the fort with the sun bursting behind the clouds and the palm trees below was also taken on a tripod to minimize shake (not that you'd have any at 98mm at 1/400s).

I believe that's exactly what you were looking for. So why the complaints?

Good morning Rishi. I'm chill baby.

So, there was a sunrise shot on a tripod? I may not have gotten that far. Why the complaints? I didn't know there was such a shot. If DPR is going to provide shots like this every time, I am willing to look at the images but may I suggest that you document which images have enjoyed the extra care, tripod, LV, mirror up, remote release. That way I and others can go straight to them. The only other thing for testing purposes would be to use a decent lens for this particular "extra care" shot.

Part 2

For lens testing, you will find a link in my forum posts to various comparisons I have made that will give an idea of the correct landscape to use. I use the same scene every time just like DPR does with the studio shot. Wouldn't hurt for camera reviews either. Find a scene location that has as much repeatable features as possible (within the limits of nature of course). I shoot from a bridge in my town that overlooks natural scenery as well as buildings. The height of the bridge puts all corners at infinity up to about 35mm so you can judge corners like a flat wall at least with the WA and UWA lenses. Also provides enough detail to judge sensor detail capture.

Such a scene would not be as informative for 200+ mm lenses I would imagine.

Rick: This sort of infinity test is something we're looking into for lens tests (not camera tests). Of course, it'll be of limited value b/c of copy variation, mount differences, and uncontrolled scene contrast (since contrast changes with time of day, cloudy vs. sunny, etc.).

"it'll be of limited value"

Not that limited, in my humble opinion. Let interested persons vet the tests of Gordon Laing at Cameralabs.com.

Thank you and God be with you.

Pat

Rishi, I promise you that such a scene will provide much of what an experienced photographer will need to assess across the frame sharpness. As for time of day and the type of day, you'll have to pick a time of day and stick with it and shoot on sunny days only. Yes, this may require shooting some samples on different days. I usually shoot my comparisons/tests during early morning hours where there is still shadow that can be analyzed and mostly cloudless days. These changing conditions are present in your general samples already so not sure what the issue is. This isn't meant to compete with a "studio test" anyway.

If a landscape scene is of limited value in assessing a camera's sensor with regards to the capture of detail and of DR, I can't think of a better one.

Rick: All I'm saying is that it's harder than you make it seem. Having shots thousands upon thousands of infinity scenes for my own lens tests in buy cameras/lenses, I am telling you that these tests are of very limited value.

Have you read Roger Cicala's work over at LensRentals with regard to copy variation, differences in mount tolerances, etc.? We'd have to first get a good copy of the lens (trust me: manufacturers don't vet their lenses before sending them to us); then, there's no guarantee that that lens happens to work well with the body we have. A good example is the Nikon 85/1.8 studio lens we have. When we published the D810 studio scene, people talked about how one side of the image was soft (even at f/5.6). Curiously enough, the same lens on our D800 was fine. Was the sensor in the D810 slightly skewed? Was it the mount?

Oh, & for these shots to work with our widget, scene alignment needs to be pretty good.

Things get complicated very quickly. Please appreciate this.

Let's look at another example: we just got a copy of the Sony 16-35mm f/4 OSS lens in. I have a personal A7R, and I started off taking some shots of exactly what you're asking for. Only to realize that this lens on this body shows softness in the almost 1/4 of the frame on one side. Everything else is sharp.

Is it the lens? A lens element? The lens mount? The body mount? Did you also know that a decentered element doesn't have to make only one side soft - it can make both sides soft? In which case, how do you figure out if it's a decentered element, or if that's just as good as the lens gets?

These are the sorts of questions we're asking - and we're making some headway on out to get around some of these issues. Simply sending lenses and bodies back and forth to try over and over again is not a viable option - we are already fairly short-staffed when you consider the number of projects running in parallel at the moment.

All this said, we have been putting together some widgets for exactly what you ask - infinity shots for lenses (I assume you're asking for lenses, not bodies). Simply b/c we're very curious about how different lens+body combos stack up against one another. For example: what kind of sharpness can you expect to get from a 16-35mm lens across Nikon, Sony, and Canon systems?

And before publishing, we have to be careful that one test doesn't favor one lens/system. For example, a sunny day with low-angle sun will create a lot of contrast in the scene, which may make the lens shot then look favorable compared to a lens/system shot on a cloudy day, or at another time of sunny day (b/c remember that MTF is a function of both input and output contrast).

We'll carefully consider if it's worth publishing some of these tests. Some of this stuff has already been shot, literally out of my own curiosity to see how the different systems compare.

Okay.

"If a landscape scene is of limited value in assessing a camera's sensor with regards to the capture of detail and of DR, I can't think of a better one."

I can. A controlled one in the lab. There is no way you can compare Raw DR between cameras from different sunsets/sunrises, shot on different days, or even shot at different times within the same sunset.

You can do side-by-sides like this. I did to compare the Raw DR of A7S vs A7R (which, sadly, I just haven't had the bandwidth to publish yet). Even there, I had to bracket shots - both to control for any transmission differences between the lenses on the two bodies, and to find the shot where the highlights are just short of clipping (since you can't determine that at the time of shooting).

Anyway, there's a reason lab tests exist. You have to control for variables. That said, as I mentioned, we're working on designing a Raw DR test in the lab. It remains to be seen if it's even feasible (no one's succeeded to this day, AFAIK).

I wonder if it also has the smart exposure function that tony northrup showed in his review of the d810.

in that review the camera adjusted from full image matric to something like point metering on the fly when using this face detect system.

Always good to read your comments, Rishi. Sorry to say it, but about the only dpreview editor who seems to know what he does.

Thanks Rishi. Nikon USA's spec page for D810 clearly mentions face priority AF as an option for 'AF-area mode'. This was not obvious based on the D750 page. Hence the question.

This is mainly if I am handing the camera over to someone for a snap-shot and I don't have to worry about explaining half-press, focus / recompose, AF-on button etc etc.

No worries @rkumar, and no one can blame you for not knowing. It's our (and brands') jobs to explain these things, and it's a bit of a collective failure on our parts for not clarifying these things.

It's also that these are complicated devices, constantly getting new features. So the Face AF, and 3D focus tracking as well, have kind of been lost in marketing talk b/c they were introduced years ago. And they do actually work. They were talked about back then, but I guess did not become known widespread. And now there are some of these amazing features that people just don't know about.

We're working on some clarifying content/articles to explain these differentiators of AF systems.

Your example is a great one by the way - when I hand off my camera to a relative to take a picture of, say, me, I want to know that the camera will detect my face properly and focus on it without the shooter having to know what to do. Face AF allows for exactly that.

Note though that in our testing, face detect AF doesn't work too well for critical focus (e.g. 35/1.4 for a close subject, where you need to nail the eye, not just the face), or in low light.

We're trying to test these things more rigorously.

Thanks; it's fun and great to engage. But to be fair to all of us at DPReview: we do try to engage with the stories we're responsible for. And when we can't, it's usually b/c we're already over-extended. Which is not unusual, all the time, for us!

Spectro - can't blame you; it's not that obvious even in Nikon's own specifications listings. Possibly b/c it was introduced some time ago and has been around and I guess they don't feel a need to reiterate it? Which is a shame, b/c I feel like a lot of these features such as face-detection and 3D focus tracking are just not well known. And yet they really work.

Here's where they describe the ability of face-detect in 'auto area AF':

http://www.nikonusa.com/en/Learn-And-Explore/Article/ftlzi4po/af-area-modes.html

Maybe the AF needs calibrating then... :D

That's interesting. I use minimal sharpening in Lightroom but I'm still used to seeing slightly better pixel level sharpness in the depth of field (it's mostly pretty obvious where you focused). I wonder if it's because the D750 has an anti-aliasing filter, and I've grown accustomed to the extra crispness you get without one (my weapon of choice is a Sony A7R these days).

The other thing is this: certain browsers don't seem to resize images properly in the gallery viewer, not sharpening after downsizing. That may have something to do with it.

Try downloading the original & viewing it in your favorite viewer. Still unsharp? Also, please don't expect tack sharp images at f/1.8 where the subject is near a corner. Even my Canon L primes need loads of sharpening on a 22MP sensor to look sharp at sub f/2.0, especially off-center.

You need to try them side-by-side, there is a difference. But a D90 is a nice camera, I got many very good images from a D80 ...

I think a side-by-side test would be the only way for me to know. These pics are good but not how I would have done them, or even the sort of subject I get to photograph!

The D90 is excellent but mine is falling apart, my choice is the D7100 or a shift up to full frame? decisions, decisions!

You are right in theory, but in practice I noticed with the cameras I owned (Nikon Coolpix 3100, Panasonic Lumix TZ8, Olympus PEN E-PL1, Fujifilm X10, Canon 1100D) that the bigger sensored ones tend not to overexpose the sky, which leads to more pleasing images

MG89, you may have experienced a narrower dynamic range meaning in your case if the subject was exposed correctly the sky was blown out.

don't think it follows only FF cameras can be engineered to have a wider dynamic range.

Canon seem to suffer here on FF compared to others for example so I don't think FF guarantees good DR.