DXOMark Tests: Nikon D5200 Edges out the D7100?

[cgw]

This came as a mild surprise, along with the D7000 still delivering slightly better dynamic range than the D7100. More of a surprise is the price differential between the D5200 and D7100, as well as the price difference(nearly 50%)between the D7100 and D7000. The D600 cruddy sensor issues, subsequent problems dealing with it, and pricing on what's really a D7000 with an FX sensor aren't helping Nikon. I'm no Nikon hater but this sort of thing tests customer loyalty.

Nikon D7100 gets DxOMark tested | Nikon Rumors

Funny but Fuji took its X-Trans sensor from the X-Pro1 and plunked into the slightly stripped down X-E1 and charged a good deal less for it.

 

[jake337]

You are paying for the extra features in the body. It's not all about the sensor. Really though, there is not much difference in image quality between, alomost, any DSLR camera at base ISO levels. If there is a difference, it is negligible.

in-body focus motor
metering with AIS, AI lenses
faster, better focusing
more focus points
larger, brighter viewfinder
build quality, weathersealing
Button layout(no digging through menus)
1/8000 shutterspeed
7fps vs 4fps
dual SD slots
Triggering off camera flash(without radio triggers)
etc, etc, etc


It's not all about the sensor. If any of the items above are must haves then you will pick the D7100.

 

[bike4fun12]

I completely agree with, Jake337 on this one. It's not just about the image quality. yes, image quality is important, but the D7100 provides a lot of features that many people look for.

 

[KmH]

Overall scores of 80, 81, 82, 83, and 84 are indistinguishable from each other in real life.

The point to take away is that the Nikon D3200, D5200, D7000, D7100, and Pentax K 5 IIs deliver virtually identical overall Raw image quality.

The D7100's Advanced Multi-CAM 3500DX auto focus module alone makes it worth the difference in price.

 

[cgw]

Overall scores of 80, 81, 82, 83, and 84 are indistinguishable from each other in real life.

The point to take away is that the Nikon D3200, D5200, D7000, D7100, and Pentax K 5 IIs deliver virtually identical overall Raw image quality.

Thanks to Sony!

 

[KmH]

Sony makes image sensors using Nikon photo-lithography steppers. Stepper - Wikipedia, the free encyclopedia

So, Sony thanks Nikon.

Stepper makers:

• ASML
• Ultratech
• Nikon, Precision division
• Canon

 

[TheLost]

Isn't this the same debate we had with the D7000 vs. D5100?

 

[KmH]

Nikon Rumors as usual is well short of being helpful.

 

[cgw]

Nikon Rumors as usual is well short of being helpful.

Nikon Rumors is just the messenger. Why shoot 'em? Inconvenient data, I guess.

 

[TCampbell]

I am not a big believer in DxO Mark. You can look at DxO for their datapoint, but you really should read other reviews on the camera. This would be like buying a car based on who makes the best fuel-injectors and ignoring all other attributes -- and to make matters worse... not even knowing how they test or score the fuel-injectors.

I came to this skepticism when I started to see results from DxO that didn't make sense compared to (a) other reviews and (b) looking at actual test shots (which DxO doesn't show) as well as real world results. At first I just thought it was one datapoint, or possibly just the variability of sensors coming off the line (two adjacent sensors being manufactured are not really identical... even if they're meant to be identical. One will always be fractionally different than another.)

DxO says they test only the sensor. You can see from their full reports that they look at several aspects (dynamic range, ISO/noise, etc.) but they aren't very detailed or specific as to how they conduct a test. For example... there are Nikons and Sonys with the same sensor. Sony has a mirror which doesn't move. That mirror cuts about 1/3rd of the light from reaching the sensor. It stands to reason that when looking at the cameras "as a whole" a Sony which loses 1/3rd of the light should NOT be able to get the same quality image as the Nikon which gets ALL of the light when taking a shot in the same room. Sony would need to boost the ISO by 1/3rd of a stop to compensate and this will generate more noise.

So what does DxO do in this case? We don't know. They're not specific. We don't know if they remove the mirror from the Sony body. We don't know if they tweak the score to put the cameras on an "level playing field" so that they can truly just compare the "sensors" and ignore impact caused by features of the "body".

An article recently revealed that if two cameras have differing resolutions, DxO resamples the image to give both sensors the same resolution. The problem here is that there are many algorithms to resample and you can't resample without doing some pixel averaging and pixel averaging naturally removes noise. It's something that really needs to be disclosed (e.g. camera a scored better than camera b after resampling, but but camera b scored better prior to resampling.) At least then the reader can make a judgement as to how important this might be.

Treat DxO as a datapoint... if you see a DxO 'score' for one sensor which is wildly different (e.g. the other sensor is "twice as good") then that's probably meaningful. If the 'score' shows one sensor is a few pennies' worth out of the dollar better, that's probably not very meaningful as there are too many factors that can influence that score and DxO doesn't spell out WHY they score one better than the other. DxO explains in rough terms (without being too specific) what they do... but they never disclose the score accorded to each attribute they test. So you never know how the score is weighted and/or if you'd agree with it.

 

[KmH]

DxO says they test only the sensor. You can see from their full reports that they look at several aspects (dynamic range, ISO/noise, etc.) but they aren't very detailed or specific as to how they conduct a test. .

DxOMark - What is DxOMark?

....To ensure highest precision and reliability, measurements are performed at DxO Labs’ dedicated testing laboratories, where conditions are controlled as in standard metrology labs. Read more...

...To ensure that measurements for all lenses and cameras can be reliably compared, we have developed detailed protocols that are systematically repeated for all equipment we test. These protocols are detailed on the site so that anyone interested in performing measurements can do so. Read more....

....Finally, DxOMark has no ties to or interests of any sort with camera or lens manufacturers, which means that we are completely independent from them.

 

[orb9220]

Yep see this all the time. Especially with newbies or Sensor Only considerer's when evaluating cameras.
And obsessing over technical details blinds you to other real world shooting needs that has nothing to do with sensor performance.

With little thought given or considered about other real world features and uses.
Like more dedicated controls,superior AF stage and more AF points. In body motor and flash commander.etc...etc..
The list goes on...

And for daily real world shooting would take a D300,D90 or even a D200 over the stripped down entry cameras.
With the missing everyday features and controls that allow me to get the shot.
Occasionally I need higher iso but is minimal for me.

Now if I shot a lot of night,low light clubs and nightlife.
Then would give more serious weight to the newer sensors as a major point for consideration.

And the majority seems obsess about the high iso performance.
Then I see all or the majority of their images are shot at 200 iso anyways.
So I guess I just don't get it.
.

 

[cgw]

"And the majority seems obsess about the high iso performance.
Then I see all or the majority of their images are shot at 200 iso anyways.
So I guess I just don't get it."


So buy a D3200? Thom Hogan(4/23) has quite a bit to say on this and the DX-vs-FX debate:

Thom Hogan's Nikon Camera, DSLR, Lens, Flash, and Book site

 

[goodguy]

Yep see this all the time. Especially with newbies or Sensor Only considerer's when evaluating cameras.
And the majority seems obsess about the high iso performance.
Then I see all or the majority of their images are shot at 200 iso anyways.
So I guess I just don't get it.
.

Interesting, I never made any survey about the ISO people use, I do know what I use and 3200ISO is a setting I use quite a lot on my D7K.

Overall the D7100 is better, it offers so much more then the D5200 and if the D5200 has few small advantages in some areas then that great but you really need to look at the big picture.
If you cant afford the D7100 then the D5200 is a good camera but in the crop sensor world as a whole the D7100 is simply better.

BTW the D600 blows both of them out of the water and in today prices if you are considering the D7100 its well worth going for a refurbish D600!

 

[Ballistics]

Yep see this all the time. Especially with newbies or Sensor Only considerer's when evaluating cameras.
And obsessing over technical details blinds you to other real world shooting needs that has nothing to do with sensor performance.

With little thought given or considered about other real world features and uses.
Like more dedicated controls,superior AF stage and more AF points. In body motor and flash commander.etc...etc..
The list goes on...

And for daily real world shooting would take a D300,D90 or even a D200 over the stripped down entry cameras.
With the missing everyday features and controls that allow me to get the shot.
Occasionally I need higher iso but is minimal for me.

Now if I shot a lot of night,low light clubs and nightlife.
Then would give more serious weight to the newer sensors as a major point for consideration.

And the majority seems obsess about the high iso performance.
Then I see all or the majority of their images are shot at 200 iso anyways.
So I guess I just don't get it.
.

Because those that can't shoot above low ISOs don't?

I never shot @ ISO 3200 with my D7000 because noise reduction looked like crap. I just did 3 shoots @ 3200/6400 the entire time with my D800 and it cleans up very well.

Now if I shot a lot of night,low light clubs and nightlife.
Then would give more serious weight to the newer sensors as a major point for consideration.

You'd be surprised what you decide to shoot when you finally have the ability to do so. You may not need high ISO capability,
but some people want to have that option there, in the event you're in the position that you do need it and don't have it.