Nikon D5200 added to our studio comparison database

Where do I even remotely imply that sharpening images also improves colors? Please contain yourself in daftness when replying or don't reply at all.

Obviously, you can not only change the sharpness, but also completely fine tune contrast and colors if you shoot JPEG with the D5200. Like every other DSLR from the last decade.

I won't magically become identical to the E-M5, but you can come close.

All right, that explanation makes sense. I'm a little lost with the esoterics of sensor performance, since the fine distinctions that people are attempting to draw here don't seem to make much difference at all in the real world. YMMV.

I agree that the differences between cameras that are shown by DxOMark and other testing sites, often don't translate into large differences in real world use. I certainly wouldn't buy a new camera solely on the basis of a sensor test.
But a camera that has better noise performance and dynamic range typically gives you more leeway in post-processing, if that's your cup of tea.

And from a purely technological, as opposed to photographic/artistic, standpoint, it's interesting to measure, benchmark and analyze sensor performance, for those of us that are so inclined.

While I prefer to shoot RAW, jpg performance is as useful as anything to compare. I make use of JPGs at family events or other occasions where I might take a lot of shots at once, and be unlikely to want to print them large. JPGs have a lot of detail if you are only going to be using them for web use or small prints.

I agree with BobBarber - I see little advantage to the D5200 over a MFT camera for what I use it for. Even if it is better at an absolute level - the real question is if that translates into any difference in my photography. The ability to carry a small, lightweight kit with me trumps absolute image quality - but we all have to evaluate and weigh the trade-offs and make our own decisions...

While the performance of the D5200 is very good in low light compared to previous APSC, one has to weight the other body factors and overall weight with zoom lens as a carry around all day camera. Compare the EM5 with 45-150mm and primes/small zooms and compare to D5200 with 55-300mm and primes/small zooms. Huge difference in the carry along bag. I have the 55-300mm Nikon lens and 14-150mm Olympus (and previously 45-200mm Panasonic) and huge difference in these Apsc vs. m43 lens sizes. Despite smaller body for Nikon than a D7000, much more portable EM5 with variety of lens. Image quality of 2 cameras is debatable and close anyway.

And because there are more F-mount lenses?
Or because the D5200 is a lower-class camera?

Same argument for D7000..maybe the big pile of megapixels and (i presume) better dynamic range and video is in the D5200's favor. Also remember a lot of buyers will only buy the latest/newest..sad but true.

Pinpointed pixel peeping, I can do that too. If you look at the Martini bottle the D5200 dissapoints. I've never seen so much CA for a prime stopped down to f8. You can download the RAWS and see for almost any other combination of gear the Lightroom CA fix barely makes any difference...but boy it makes a huge difference here. Somewhat dissapointing since it is a studio shot stopped down to f8.
To be honest the narrow aperture modern sensor pixel peeping is so marginal in final IQ in many real-world situations. What is interesting however is given the poor CA I see at f8, I'm very sure both the Canon and Nikon 50mm's on APS-C will dissapoint at wide aperture compared to the Oly 45mm 1.8. And wide-aperture lens performances varies far more than narrow aperture studio shot tests. This is often the most glossed over performance facet of Nikon measurebaters is in fact wide-open lens performance or the belief software can fix all CA.

on the other hand the oly is a higher spec camera than the basic Nikon. Still the 24 m of the Nikon cannot rival 16 m of the oly.

Please note that exposure values. Eg., ISO1600, the Oly was 1/400s and F/6.3. The Nikon was 1/500s and F/8. That's a full stop difference. If you were shooting in low light, the Oly would either need longer shutter times or higher ISO's just to get the same exposure. Try comparing ISO3200 on the OMD with ISO1600 on the Nikon for a apples-to-apples comparison. Pretty clear the Nikon is doing better by about 1 stop. The difference in sensor would suggest 0.8 stop difference, so Nikon did a good job of getting even more than one should expect just due to sensor size.

I must agree Oly ISO is a joke. Comparing two camera noise at the same ISO is also a joke. DPReview ISO compare tool is also a joke. 4/3 sensor is better than APS-C and APS-C is better than full frame is also a joke.

JPEGs? That's what you guys are arguing about? Olympus used to have decent JPEGs. EM5 changed all that. Look at one at 100%. NR and artifacts make it hard to imagine anyone would use them over the ORF files. Download an X-Pro1 JPEG.

As far as fools, go ahead an compare some real images from the D7000, D5200, K5, X-Pro1 with the EM5 images. ALL of these APS-C cameras have cleaner images at ALL ISOs, and better color depth. DR? Forget it. EM5 has about as much DR as a P&S like the RX100. It's a bit better IQ than the Panasonic GX1, nothing more. I have to laugh at all the FF vs EM-5 comparisons I read on the m43 forums. FF? m43 has not managed (nor wil it ever) to match APS-C performance, and m43 fanboys want to start threads with comparisons against the 5D3 and D800??? That's alternate reality stuff. Seeing what you want to see, DxOMark conspiracy theories, the whole nine years. I guess when you drop the kind of cash m43 users do on their gear they want to believe.

marike6

Your desperation is clear for all to see.

Marike, do you own a m43 camera? I own both a D7000 and an Olympus PEN, and I'm getting much more detailed images (edge-to-edge) out of the PEN.

err Yes. They have two studio setups, which are absolutely identical except for those watches. Or perhaps...

it is the same physical studio scene and maybe a few of the elements have (been accidently) moved slightly after some of those older studio test shots were taken, e.g. look at the staples and some of the feathers in the shots from older and newer cameras.

My money is on the light hitting the (same) watch face at a slightly different angle to before.

High ISO performance is actually outstanding.

As far as glass, like all F-mount cameras, the D5200 gives the user more lens choices than probably any other mount including EF mount.

And like other high resolution cameras, i.e., D800, D600, 5D3, NEX7 and A99, a mediocre lens on the D5200 will out resolve the same lens on a lower resolution camera. Defects and aberrations may be more apparent at 100%, but the resolving power of a given D5200/lens combo will be greater.

Utter nonsense. The kit lens is well able to show 24MP at a lot of settings.

The difference is that DXO down samples images to 8MP, which isn't far off the V1's native res, where the D5200 is downsampled to 1/3 resolution - at 1:1 pixel size you mightn't see that much of a difference, especially as the pixel size isn't that much different either, but downsampled there would be a much more striking difference.

It was like when DPR compared images from the A77 to the A55 - at 1:1 the A77 looked a lot noisier. But downsampled to the same 16MP they actually looked pretty similar.

Also, DxOMark analyzes the RAW file before demosaicing, whereas the "RAW images" you see in comparisons on DPR and other sites are, of course, JPEGs developed from RAW with default settings. These tell you as little (or much) about the capability of the sensor as the out-of-camera JPEGs do.
If the comparison tool shows little difference between D5200 and J1 (which is arguable, depending on what you look for in the images), it only means that the default ACR settings give similar results.

Which is more relevant to the photographer, DxOs numbers or DPRs comparison images, really depends on how much post-processing you intend to do.

I'm not sure which users those would be, one look at the 17 1.8 / EM5 gallery vs any good D7000 or X-Pro1 set of images
show that the EM5 simply doesn't match the better APS-C cameras, i.e. NEX7, K5, D5200, X-E1. The EM5 does not compare favorably with any of these cameras for IQ. Unless users like the grainier base ISO images, or weaker DR and color depth I can't imagine anyone would choose the m43 cameras for IQ.

I'm an m43 user, and IQ improves with each generation of cameras. But the smaller format will never equal the best of APS-C.
Sorry to be so blunt.

Agreed, marike6. I used to have both 4/3 and m4/3. Sold them because I still prefer the outputs from my Canon and Nikon. If one needs small mirrorless, NEX is a good start over m4/3.

"for my wife to use" rather sexist I n=must say. Just kidding.

All the best
Rocco Galatioto http://galatiotophoto.blogspot.com

Yeah right. I still laugh when i see a NEX. Like a giant milk bottle stuck on an iPhone, with similar optical properties.

By kewlguy (1 hour ago)
Agreed, marike6. I used to have both 4/3 and m4/3. Sold them because I still prefer the outputs from my Canon and Nikon. If one needs small mirrorless, NEX is a good start over m4/3.

some guys below said m43 is incomparable with APS-C cameras refering to some DP test shots of a recent Oly glass.
I am not sure what happened to those pictures but there are tons of test samples on the net to think otherwise.
Moreover, you guys can just use directly the studio comparison too with the d5200 samples and switch on the OMD and then set up the ISOat 6400 level-then you will see how the Nikon sensor is "better" than the Oly. You need to accept that there are some other cameras out there which are just better or at least as good as Nikon and some of them come from m43 community.

@kecajkerugo
The problem for your argument is none of the testing sites back up your claims at all. It's not even close.
You might have a look at D5200 sensor scores relative to the EM5.

DxOMark Sensor Ratings (click shortened link below)
http://tinyurl.com/b99gkhp

Better DR, high ISO, colors, DOF control, way better AF tracking, way more lenses at far cheaper prices. m43 cameras are smaller I'll give you that. But most men I know don't exactly have problems handling a 1 LB camera.

Like I said, I did use quite an extensive system of m4/3, including EM5. Sorry, the details are not up to the FF cameras with good lenses. It's reality - esp the low light details. Horses for courses, sure. I'm not saying m4/3 is bad, but it's not as good as the best APS-C system, still. NEX might be weird looking, but the sensor is better than the best m4/3 sensor currently available. The Oly defenders need to have some reality check, somehow.

You seem to have fallen for the trap of believing what you are told by a company who's interests you know nothing about, rather than your eyes. Remember the ridiculous delay from DXO re the EM5. They seemed panicked by the real results everyone else discovered for THEMSELVES.

@olypan So DxOMark conspiracy? lol.

@olypan, your comments say much about you than about dxomark. Anyway the reasons for delay in E-M5 results were that oly was playing tricks and that was tricking dxomark.

D5200 surely has marginally better noise performance than D5100, but its output is visibly softer than D5100 or T4i. It could be because of the AA filter or it could be because of Nikon's implementation of higher DR sensors where shadows appear lighter and hence lacking contrast and all colors appear pale. Am I missing something?

@aftab

First of all RAWs from all four cameras look similar in sharpness.

But realize that the D5200 is a 24 mp and the D5100/T4i is 16/18 mp. Resize the D5200 image to 18mp and you'll see even less subtle differences in sharpness.

But there is a big difference in DR, color depth and high ISO performance between the D5200 and T4i.

Have a look at DxOMark sensor ratings and on Select T4i, Select D5200 and click Compare Cameras.

The d7000 should be replaced quite soon. ... Perhaps just a bit longer wait.

You should try to handle both bodies. The first time I purchased the D7000, I actually went down to B&H to buy the D5100, but when I saw the large, bright 100% finder of the D7000, I just couldn't pass it up. A good VF makes the shooting experience so much nicer. D7000 also has two command dials (one right underneath the shutter button) to the D5200's one.

That said, I'm sure I could work with the D5200, but the IQ is so close between the two cameras I wouldn't let it be the deciding factor.

Those in the update/upgrade mode should wait a couple of months, the D7000+ has been already rumored for launch by Nikonrumors, usually riht. It will carry this sensor, very likely.

and how right they are

Not this again!!!

The more mp = more noise has been proven false now for several years.

In fact see who is leading ALL crop sensor cameras for Low-Light ISO score on DxOMark. Hint the result is related to this article.

http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Ratings

@RedFox88 : do you mean larger sensor?

There are no myths but some laws of physics that cannot be infringed.
DXOmark results are clear : bigger sensors are ALWAYS better than small ones for the SAME pixel count when build with then SAME technology.
Given the SAME SENSOR TECHNOLOGY, the bigger the photosite the more light (or photons) will be collected, resulting in a greater signal/noise ratio. And greater dynamic range too, because the photosite can collect a greater number of photons before being saturated. More pixels for same sensor size = more noise = less dynamic range !
Year after year technology improves so it is possible to put more pixels on the same sensor size. At the same time, improvements also allow to increase the signal/noise ratio for the same pixel size. The consequence is that a today 24MP APS-C sensor has a great chance to be less noisy than a 12MP APS-C sensor of yesterday. But if a 12MP APS-C sensor is build with today's technology, it will have a better signal/noise ratio then the 24MP one.

Continued,

So does it worth having a 12MP APS-C sensor with today's technology rather than a 24MP APS-C sensor for a better signal/noise ratio if you do not need 24MP ?
IMHO, I think it does not. 24MP is more versatile. If you need high-res, you have it. If you do not need it, you can downsize your image. In-camera if you are too lazy/busy or want to save memory card space, or in post-processing. If the downsize algorithm is good a 12MP picture downsized from 24MP will not be much noisy than the full-res picture from the 12MP sensor : combining pixels averages the noise. Plus it will be SHARPER.
The reason is the anti-alias filter cut-off frequency is higher on the 24MP pixel sensor... So it's more or less like having a 12MP sensor without an anti-alias filter.

Never mind it's not worth it.

Let's bury this hoary myth deep, for pete's sake.

I couldn't figure out his point either.

As far as Toshiba, I wish they made a 1" sensor that Nikon could use for the N1, to get performance up to the RX100 level in some regards. The Aptina sensor has very good IQ, the Sony 1" has excellent IQ.

Nikon fanboys shouted all these years the advantage of lifting shadows like crazy in post, and argued that this is the single most important thing and Nikon beats them all. Now that the D5200 is rubish at that, I hear no word. Talking about blind fanboyism

Maybe it's because the D5200 is not rubbish at lifting the shadows and as DX sensor it is not expected to be as good as full frame sensors in that regard anyway?

Re: lifting shadows in LR. DxOMark DR scores:

D5200 13.9 EV (Same as D7000)
D5100 13.6 EV

If people want to argue that if you lift shadows some absurd amount like 8 EV as someone on the DX Forum was suggesting produces banding in the D5200, I'd submit that ANY sensor, including the D800 will have negative side effects to such a dramatic EV push in LR.

I love the 16 mp Exmor but all things being equal I'll take the high resolution, DR and low-light any day.

Because it is known
that it is a Sony sensor outsourced by Sony to Toshiba in a long term relationship, probably because of capacity shortness due to the flood twintigjarig years ago.

@bluevaping : I have checked the techradar review of D5200. The SNR figure of 650D RAW does not match with DXO website data.On techradar, 650D is better than D5200. On DXO, the 650D is very bad.

" Yet, on DxOMark the D5200 manages 1284 ISO score to the K-5IIs's 1208 ISO Low-Light (Sports) Score. "

"D5200 13.9 EV (Same as D7000)
D5100 13.6 EV"

Don't read too much into the minor differences like that, they are all within margin of error of DxOMark (obviously, k-5, d5100 and d7000 use the same sensor, as well as Sony A57, A65, NEX-5N, NEX-C3 etc etc).

Vodan, Tchrader is not very good at sensor testing, and they use DXO software for testing.

starting at ISO 800 RAW I start to see noticeable better IQ from Pentax K-30. the smaller Mpix seems to handle the noise and retain the detail surprisingly better than d5200.

yeah k30 has the best apsc low noise iso out there.

plus it has anti shake inbody. with nikon and canon you have to buy a lens with that built in = blurry edges.

Re lifting shadows, for any dcent realistic lifting the D5200 is on pair with the D7000, the champion there for APS-C files. Only for extreme 5+stop compensation you may see slight lines, not really banding like one sees in Canon files. Checking RAW values, only when they reach below 5, in all values, you get issues, That's in the last 2 bits, where all the noise is. The D7000 will show slightly less issues there. The estimate is that you get 12EV of DR, RAW, very clean. Find another sensor that does much better. Most will do much worse.

People stepping up from a compact camera with 12-16mp expect their SLR to have even more, so Nikon has low end SLR's with 24 mp sensors, just what the customers believe they need.

Aside from more ability to crop in post, downsize a D5200 ISO 6400 image to 16 mp and compare the same image with the D5100, and get back to us.

As far a the Canon 650D, go to RAW and go up the ISO scale. The T4i/650D sensor is not even in the running.

I just compared them in DXO Mark. The D5200 sensor is better than 5100 in all categories, and you get extra resolution.

I have to agree with you on the D5200 vs. D5100. If you're really looking closely, the D5200 is slightly better than the D5100 but not by much. Certainly not worth an upgrade from an image quality standpoint. But might be worthwhile for the performance boost.