@apefos Yes indeed the 6400 ISO clip looks really good! When denoised it looks even better!

After these great denoised results from D5200, Canon and Sony will need to rethink their marketing strategies and sensor design for their APS-C cameras.

Starting to look like the Nikon D5200 might be the latest happy accident like the GH2 was. higher quality than perhaps even Nikon meant to give us for a bargain price. Perhaps lighting can strike twice in the same low cost spot :) I look forward to more test videos.

@Craftwerk @noddydog

Try to download the Samuel Hurtado flaat profiles for Nikon cameras and install them in the D5200. This can increase the dynamic range a lot and turn this baby into almost a BMCC. Tell us if it works!

http://www.similaar.com/foto/flaat-picture-controls/index.html

I ordered one today, I'm going to be testing it up against the 5D MKIII as soon as it arrives. Hopefully I'll have the results up on vimeo by the end of the week.

The thing I'm most concerned about is the horizontal pattern noise, I'm only seeing it in some low light high ISO shots so I'm hoping some function that can be turned off is causing it. I'll be testing it with everything that can be turned off and the flattest profile that comes closest to prolost on the 5D. If I can get an adapter in time I'll shoot with Leica-r glass.

I don't expect much from the codec. Coupled with a hyperdeck shuttle you've got an uncompressed or prores camera for $1200! No shipping confirmation yet... bloody Chinese new year!

@squig That test up against the 5D MIII is great! Looking forward to the results. BTW, I've shot with different Picture styles and noticed that these horizontal streaks are gone with the Standard Picture style, seems only present sometimes in certain situations. But then again, when you have video footage with noise, you alway filter it out and so are these streaks. So no big deal for me anyway.

Samuel has given me some info so I can perform a dynamic range test on the D5200 and the MKIII.

I suspect there might be something else other than the picture profile that's causing the horizontal banding. On the MKIII the banding is much worse at particular ISOs especially when HTP is on, there could be some other camera function that accentuates it.

Flaat should work but the data I give to Samuel will give him the opportunity to create a flaat profile specifically for the D5200 if the cameras emulsion differs much from the D800.

@squig An external hdmi recorder can be great for this camera, but for the most cheap and portable lightweight gear a simple hack to 40Mbps would be perfect! Hope this happens someday. Also...

In the flaat for the D800 Samuel said he used the Portrait as the base to develop it. As Neutral is reported to give some horizontal lines, I believe if he use the Standard as a base to develop the Flaat for D5200 it will be better because the Standard do a better job to the highlights than the Portrait. In the japanese review stills you can see the Portrait test deliver more pronounced highlights and Standard is better, the difference is minor, but it is there, and for video each drop of improvement counts, see:

http://translate.googleusercontent.com/translate_c?depth=1&hl=pt-BR&ie=UTF8&prev=_t&rurl=translate.google.com.br&sl=auto&tl=en&u=http://dc.watch.impress.co.jp/img/dcw/docs/579/470/html/86.jpg.html&usg=ALkJrhicuEyb94mA7eyMSYqni-9HiYoKMg

http://translate.googleusercontent.com/translate_c?depth=1&hl=pt-BR&ie=UTF8&prev=_t&rurl=translate.google.com.br&sl=auto&tl=en&u=http://dc.watch.impress.co.jp/img/dcw/docs/579/470/html/90.jpg.html&usg=ALkJrhglsBw91BHbumhp70VLCSX9clT_Bg

http://translate.google.com.br/translate?sl=auto&tl=en&js=n&prev=_t&hl=pt-BR&ie=UTF-8&eotf=1&u=http%3A%2F%2Fdc.watch.impress.co.jp%2Fdocs%2Freview%2Fnewproduct%2F20121220_579470.html

Looking at those images neutral is showing more shadow detail and looks like it has the most latitude. There is some pattern noise in the shadows but crushing the blacks slightly gets rid of it without a big detail hit. Standard does have the least noise but I need to test all 3 to see which one has the most latitude with the least noise graded. My initial tests will be performed with the internal codec, if all goes well I'll do some more tests recording straight to prores in a couple of weeks.

@Craftwerk If you have time, do you think that you could upload a side-by-side lens cap-on ISO test like this?

There may be a sweet spot for ISO like there is on the GH2.

Oke guys, I've just uploaded the "Denoised" version of noddydog's video. Take a look for yourself... and be amazed!

looking good craftwerk, does the aa script help much? the nr works really well