tusoli5

4k in 8bit 4:2:0 h.264 while the competition had that nearly 2 years ago? By the time we'll get a Nikon DSLR with non-cropped 4k to internal 8bit 4:2:0 h.264 I wouldn't be surprised if the competition would all have external 10bit like the GH4 already has or at least way better codecs and/or higher bit rates internal.
If Sony would make the HDMI out 10bit they would kill the S-LOG problems with banding and have a crazy thing. To be honest I don't think Nikon will even try to compete for the video enthusiasts in the next years. Even the new cameras seem to only have zebra in regards to exposure/focus tools.

In terms of image? I have an NX1 and think it has some nice characteristics, but the poor highlight roll off and general lack of DR put it below the 1DC. As for Sony, the colours just don't cut it, for me.
For overall image and performance as a hybrid, I think the 1DC still reigns supreme.

Yup, one of the more exciting things about this announcement is what it bodes for the future of other Nikon DSLRs!! :-S

I totally agree, I own and loooove the D800 for its capabilities and these images didn't really shined for me. 
But remember that when the D4 was released it had been really quickly outdated by the D650/D800/ D5500 for video. 
This release gives me the same feeling. D4 and D500 for stills with great capabilities, and video renewal and we'll have to wait a bit more for a more convicing use of 4k in a next nikon dslr with same amazing still capabilites, maybe in Full-frame, maybe also in dx much cheaper.

Maybe im biased because of the NX1, but this doesnt look good at all.
The D500 seems outdated, and something that should cost 1000$ less, idk why they thought it would be ok to have 2.1x crop. The D5, as incredible as it will certainly be for stills, doesnt look good in video, I dont care for the 3 min limit, since I usually record short clips, but the image is simply not there, and judging from some clips and even the thumbnail, the DR looks quite terrible. The image lacks detail, looks muddy, and if anyone told me "This clip is from the D800" Id still think its a bad clip, since i've seen the D800 do much better.

The hw part that is evolving the most in these times are the ssd. Imo in three years from now we'll have some very fast ssd to be connected via super fast interfaces to video cameras. 500+ mb/sec will becone common and also big capacities, say 3+ tb for about 300€.
 
Of course those won't be sd cards...

I'd take 1080 raw over 8K any day of the week

I have already been down this road for several fashion shoots. It is absolute hell to edit.
One day of shooting, four days of editing.  The loss of time in post production is the single greatest reason why the concept does not in any practical way work.  Perhaps the on set photographer may lose their jobs eventually, but for the rest of us taking stills is a far more effective workflow than pulling from video.

How I think this would work, in still mode.... have x amount of seconds of 8K video (compressed raw) constantly rolling into the buffer, but allow the photographer to use the shutter to pick the moment with the shutter, storing the previous X seconds of video and the X seconds of video after the shutter press. This would give the photographer that passion to capture the moment, would create the excitement of the shutter that models feed off, but also give a short "video frame burst" to run through in post that would almost certainly contain the perfect moment (probably within a millisecond before or after the shutter click). It could also store a raw still image of the exact moment the shutter is pressed too.
That's a feature I would love.

Yes shutter speed is, but that part isn't hard to fix is it? If you're shooting stills with a video camera, just select the desired shutter speed. If you're not trying to deliver video, just stills, then it's not a problem. If you're trying to do both at the same time & you're using 1/50 for crazy fast action, it's likely the wrong choice anyway. 180 degrees shutter and 24p doesn't suit ALL.
I disagree with you I'm afraid Ebrahim as like I said in the article I think the technology or stills & video will merge.
Firstly it will be global shutter for both stills and video as standard soon and the mechanical shutter will be obsolete, as will a rolling shutter. We're nearly there... few more years.
It won't hurt image quality at all in 2018.
Look at the Sony F55 global vs F5 rolling for instance. We're almost already at a point where a global shutter's impact on low light is minimal. The sensor for that camera was developed before 2014.
Again it depends on the technology of 2018 not of 2014.
These are also all fixable in future.
The technology is merging, like I said in the article, but the technique is not really the same for going about capturing the content.
A7R II's on-sensor phase detect speed at it's best is easily a match for any professional DSLR!
Panasonic GX8 contrast detect system is as fast as a DSLR with many of Panasonic's lenses.
EVFs are coming that will blow your socks off. Seen the one in the Leica SL?
The mirror assembly is definitely going to be obsolete soon.
Nah it's not
Again an easy technological problem to fix in future.
Soon even REDs and Arris will have small batteries as their processing efficiency increases.
You see how thin laptops have become? That is all due to lower power consumption.
Same trend is seen across all semiconductor technology.
Who says what Panasonic, etc. will produce will be huge? Might be a GH4 style body!
Cost of storage is actually coming DOWN.

Not with H.265
PS - if Panasonic froze sensor R&D since 2011 that would explain why Sony made the GH3 sensor, but who made the GH4 sensor and what about the new Varicam S35?!

By 2018 computing power will be so fast and cheap that NLEs will cut through 8K H.265 like butter and the file sizes will be as small as 4K is now, thanks to H.265 but storage space will have probably tripped for the same price.
Indeed it is about what's creatively useful though, not just outright image quality.
I would prefer 240fps 4K and better low light performance than noisier 30fps 8K, although downscaling to 1080p or 4K would reduce noise, as we see with the current 8K camera, the A7R II's 42MP stills!!
Also I believe the gains coming from organic CMOS sensors in dynamic range will probably eclipse the excitement for resolution in the coming years.
So I hope Panasonic's R&D money can focus on more than just 8K.

Oh, please. The only thing that's kept me from buying a BMC (original cinema camera) is sensor size, and I'll eventually find a used one with a speedbooster, they show up every week. That's a camera that excels at pretty-much one thing: really pretty footage that can be graded like mad, for projects that don't need extensive keying but beautiful IQ on a "let's try to avoid a bunch of rentals" budget. So you're saying a "professional" would, what, go hunting for wide primes with no distortion vs. adapting lenses they already own? I've seen speed boosted Zeiss and Nikkor AIS 28's on that sensor and it's lovely.
I'm a "professional" - I make my living from pitching concepts, writing scripts, and shooting or animating commercial (IE, promotional vs. entertainment) projects for businesses, including some good sized brands. Sad to discover that owning a speedbooster for a specific purpose would make me an amateur.

Well, now that everyone bashed the NX1 for its codec, it is not for sale anymore...
It was really easy to see that it was only a matter of time to see premiere handeling H.265, but no, everyone turned down samsung offer without understanding its value, or even trying the NX1 (I am not talking about you Andrew, but much more about no brainers playing "I know it all" like its ever been through history of technical evolution). I hope next H.264 DSLRs will be understood for their real value thanks to this codec.
Anyway, lets hope that thanks to this native editing possibility blackmagic, nikon, canon, panasonic, red and more, much more companies, will handle this codec for what it was trully averagely made for, I mean 16 bits 422/444 and 120i/s in full 4k rec 2020 (at least 35% more color space than P3/dci).
It is indeed the real game changing fact about filming. This codec was made for a united new generation color space from shooting to viewing, processed to handle preview, color correction, theater/tv/Blu-ray UHD/internet from hd to 8k and more diffusion at the same time with more powerfull color we have ever used until now in digital cinema. So, no more upscalling down scalling, changing codec through the editing process to the final render for a bit of time... untill better capapbilities come to use...
At least, it could produce the cleanest 1080p high framerate 10bit 444 ever seen in rec2020 in low prices DSLRs !
Raw is still in DCi at max to this day, so try to imagine the use of new rec2020 processors in big cinema cameras. It gives you an idea why the whole cinema industry prices for sony FS,red and others put down their prices in 2015. New generation is on its way. 4k was a good thing, Rec2020 in 16bit will be amazing. 
Try not to destroy camera values by saying anymore that you prefer crappy H.264 possibilities over it.

Tbh there has been no official word from samsung or nikon, so all is speculation. But if true there is one thing you can count on,  you will be paying 3x the money for the nikon version of the NX1. Not sure if that's a win for anyone here. And don't expect Samsung's level of commitment as far as firmware upgrades go.  This is exactly why I bit the bullet and got the NX1. Unless my eyeballs are upgraded, the NX1 should be good enough for the next few years. At least until 8k is mainstream. 

This is great news if it's true.  The beautiful Nikon color science combined with incredible video recording abilities and a fast processor and hopefully a sensor with more dynamic range or smoother highlight roll off and you have an amazing camera.
Hell yea, it's a win-win for both companies.

I really don't see why Nikon would keep the NX mount when they have a half century history with treasured Nikkor glass. In the 70s and 80s when Canon, Olympus, Pentax, etc changed their mounts, Nikon designed their cameras around their lenses.
Not to mention, think about how much of a boost their lens division will get by the onslaught of new customers. 

Well, now that everyone bashed the NX1 for its codec, it is not for sale anymore...
It was really easy to see that it was only a matter of time to see premiere handeling H.265, but no, everyone turned down samsung offer without understanding its value, or even trying the NX1 (I am not talking about you Andrew, but much more about no brainers playing "I know it all" like its ever been through history of technical evolution). I hope next H.264 DSLRs will be understood for their real value thanks to this codec.
Anyway, lets hope that thanks to this native editing possibility blackmagic, nikon, canon, panasonic, red and more, much more companies, will handle this codec for what it was trully averagely made for, I mean 16 bits 422/444 and 120i/s in full 4k rec 2020 (at least 35% more color space than P3/dci).
It is indeed the real game changing fact about filming. This codec was made for a united new generation color space from shooting to viewing, processed to handle preview, color correction, theater/tv/Blu-ray UHD/internet from hd to 8k and more diffusion at the same time with more powerfull color we have ever used until now in digital cinema. So, no more upscalling down scalling, changing codec through the editing process to the final render for a bit of time... untill better capapbilities come to use...
At least, it could produce the cleanest 1080p high framerate 10bit 444 ever seen in rec2020 in low prices DSLRs !
Raw is still in DCi at max to this day, so try to imagine the use of new rec2020 processors in big cinema cameras. It gives you an idea why the whole cinema industry prices for sony FS,red and others put down their prices in 2015. New generation is on its way. 4k was a good thing, Rec2020 in 16bit will be amazing. 
Try not to destroy camera values by saying anymore that you prefer crappy H.264 possibilities over it.

I agree with Dahlfors
 
there are several things to compare to talk about good or less good cameras.
 
First what is your style? Publicity, nature, macro, portraits, fashion, journalism, art and museum printing? All of that? This will help you know if you look more towards APSC or full frame, full frame with 12mp, 24mp, 36mp or more...
 
Comparing high iso, for example, on a 5DmkIII I am at 3200 iso when I am at 800-1000 iso maximum on a nikon D800 for the same results. So, yes, for me the Nikon D800 is good enough comparing to a 5DmkIII, yes it behaves worst at 3200 iso than the 5D but if you know these two cameras and practiced them a lot you see that you have to compare a bit differently than just in the specs because in this precise example, the 3200 iso should be considered as having the same results than around 1000 iso on a D800.
 
About quality for stills on the internet, I confirm what dahlfors says, it will always look shitty on a 1080p screen compared to paper, wich gives you full potential and depth about color, contrast, etc, simply because it has muuuuuuuch more resolution than any screen available today, even in 4k.
Professional paper-print meets around 70mm  celluloïd résolution (around 12K resolution If I am not wrong, yes we will get it one day, no you can't really measure the depth of your colors on the internet with compressed picture from crappy jpeg, yes that is why all professional are obsessed with calibration to get the more they can of what they see on screen and it justifies great gammut screens to be sooo important).
 
Then, about what reason drives the megapixel run? Really, have you ever achieved a 2meter by 3.5 meters on a museum wall with a 24mp camera? Yes you can but you make an upscale of your print and drain a lot of quality away from it. It will simply never looks exactly like what you thought you had on a smaller scale. So if you make publicity for biiig brands, you had the following solution before the D800E: Hasselblad, mamiya, 70mm and 24x36 celluloïd.  Nikon changed this with the D800, so yes there is a way to see this camera as a real need if it fits certain domains of your work or creativity. Size does matter with the size of your prints and also it need to be done very well, wich is the precise improvement with d800, a 36mp that helps to skip completely hasselblad, leica, mamiya, and use all the lenses that you work with everyday on your other projects. If I shoot sport models for an event and then get the opportunity to print it on the town hall of the biggest cities of the world I don't mind anymore, I stay confident with it I have the strongest 8K stills possible and may not need any upscale.
 
Knowing your applications, knowing you creativity, knowing your market and fitting their needs is what makes you choose even some cameras that might look like not so good in everything, precisely because you need them for what they do so well.
Furthermore, why does everyone suggest that when you buy one brand you don't buy the others? I need canon for certain things like softness, very special bokeh,  I need nikon for other things, like more crispy images, more mechanical works, and I need leica for others, like dramatic nature, etc... But from my D800, I stopped completely using Leica, Hasselblad, Mamiya for my stills. 
And I still like canon for the easy picture styles they put in their cameras, really it is a full color station in a tiny body.
 
On the D800, the dynamic range is so big that even with 36mp you can recover low light very easily and shoot more instinctivly, is there a test or internet talks about going easily instinctive with a camera? Not really, they speak about buttons positions etc but never can make you picture it fully. For my experience, D800 sensor and its 36mp really made the trick for me, I simply don't mind anymore. Would I get a better view in low light with less resolution? Yes. Is it my style of concieving photography? No. As I realised that my choice was made.
 
So there is a limit to testing without touching, there is a limit to knowing all without using.  You have to know what you want to do with it to get what you need. 
 
There is a limit to see pictures without looking at them printed, celluloïd lost the battle for the source but today paper still wins the print and from good to very good, you'll see a lot of differences in the final result at an exhibition, for example. ;-) 

I've worked with Alexa LogC, Canon Clog, Sony sLog in post and on set, the real difficulty is exposing correctly.
 
Because of the way a log gamma bends the response of the camera, it's easy to overexpose, even though exposure looks right on screen, so expose conservatively or you'll put the skin and the like on the severe end of the log curve.
 
In post, it's best to source an LUT for your chosen log as a starting point, and work from there. 
 
For Canon Log, the Abel Cine Clog to Wide DR Full Range is best. For Alexa Clog, Arri's website contains all the LUTs you need. For Slog you can get away with Cineon LUTs as a starting point.
 
To answer the specific situation there, you'd recover the scene as it was shot, with the extended highlight benefits of log. So once you apply your LUT your colour saturation returns, you have full black and white and it'll look more like it would had you shot in Rec709, except you'll have more highlight range to play with.
 
If you overexposed on set, the LUT will clip out the tops. With some LUTs you can just pull back the highlights and recover, be aware that some just slice ff your overs though, so experiment. 
 
In 8-bit, yes it's much harder to use log, because the colour space is very unforgiving. Canon are clever about how they do 8-bit in some way, so it's pretty good. Alexa is better of course. 

In mathematical terms Logarithmic scale is:
 
"... a scale of measurement that displays the value of a physical quantity using intervals corresponding to orders of magnitude, rather than a standardlinear scale. The function of the curve may include an exponent, which is the source of its curved nature."
 
Translation, it's a curve.
 
In digital camera terms it usually refers to a profile curve that is built into a camera's firmware that is going to select the bits to encode from the scene, to the eventual file, but that file will be recording 256 (or fewer) integers if it is an 8bit format. If you want to retain the typical full range (and have it be usable) of a camera's log profile, you will want to encode to 10bit. Most log profiles consider middle gray as 18%, not 50%, because our eyes see 50% as almost white, and require 10bits to cover the full range. Log profiles often look "flat" and can vary in their "flatness" based on the profile. The original source of the flatness was not desirable as much as necessary to reproduce a print and include more dynamic range, but it also can allow more details to be drawn out, assuming you have the color depth.
 
So basically, a log profile might not be desirable when shooting rec709 8bit, but this is the source of much (much) debate.
 
A LUT is a color lookup table and has more to do with post and changing the way you interpolate a file, although it can be used for lots of other conversion needs. 
 
Gamma log seems redundant to me, I'm confused by that one, that's kinda like "Scientology", i.e. the study of science. 

According to 4/3 Rumors, a reliable source says "Gamma Log" is coming in a firmware update.
 
http://www.43rumors.com/ft4-gh4-firmware-upgrade-will-add-gamma-log-and-dave-dugdale-leaves-canon-for-the-gh4/#disqus_thread

Yeah, with the hack you can grade your videos, because the compression artifacts almost dissapear. Before you couldn't do anything because the ugly blocks would show with any minor adjustment. I finally came up with a workflow which gives me exactly the look I like (you just have to film with white balance off a little into the blues and remove it later with a curve, suddenly the color problems seem to be gone). The only thing I miss is detail on far subjects, but that's a thing for 4K.
 
I will keep the d800 until a 4K 10bit internal camera comes around which also offers the d800 stills quality (unless used phase one backs go under 3000€, I want that iq280 so badly).
 
btw, great work tusoli.

Yeah Nikon are really going to give us oversampled 4K video on a DSLR aren't they??! What from? 8K!?