Pavel Mašek

There isnt a separate NR adjustment with the Mavic. It is lumped in with sharpening. Set sharpening to 0 or below and you get buckets of noise reduction and blotchy shadows. Setting sharpening to +1 gets rid of the noise reduction but does cause other problems. DJI resolved the problem with the Mavic Air.

Actually the Mavic Air is very good. Definitely softer than the Mavic Pro 1 but DJI footage is too sharp anyway. The only thing that bothers me on the Air is the noise even at base ISO. You must expose carefully. Here is a video I made with the Mavic Air. Most of the artifacts comes from the youtube compression (and from my flashing grading sometime).
 
Perhaps, as I said earlier in this thread I used the Phantom 3 Pro extensively and I never encountered the image issue I had with my Mavic Pro 1. Both have a 1/2.3" sensor that max out at 60 mpbs so it might be something else. I'm not exactly a pixel peeper but I got rid of the Mavic Pro 1 just a few days after purchase because the footage and "compression" issue was pure garbage.

Here's a quick little spot I shot for a local restaurant's facebook cover yesterday, very quick turnaround because those Sammy colors are so gorgeous! This is basically right out of camera, only a tiny bit of tweaking on a few shots. Normal gamma, color sliders at 1.99, sharpness -10, contrast -4, saturation -2, on both the NX1 and NX500 (the only difference being the master black level on the NX1 was at 15 and obviously the NX500 doesn't have this setting). While this isn't showcasing of increased dynamic range per say, I do really like the way it came out looking. 
password is jacks
 

short film with settings RGB 1.99 , sat-2, sharpness -10, contrast -2. MB at 10. With anamorphic lens Bolex 16/32/1.5x.
 

You are correct Pavel, as far as negative RGB is concerned highlight control is a problem. The histogram shows that the highlights are well within range but they are in reality clipping at a lower level than 255 more like 235/245 that's why everything above these levels is a flat grey.

Thank you all especially @Happy Daze  for investigation. I have concluded for myself that boosting RGB values is similar to highering ISO values but it makes sense that this way can help reduce noise reduction that is probably higher on ISO 1600 and above ( @ReinisK  video).
But I have test it and found out that details in shadows are higher with lower RGB values but downside is the highlights are clipped because of longer shutter speed (unfortunately it does not work as digital ND filter). So maybe RGB x1.0 is still best compromise. But I still do not understand it... 
NOTE: Images grabbed from video directly in NX1
0-255, ISO100, RGB x1.99 on the left; RGB x0.50 on the right (just longer exposure to compensate loss of brightness).
You can see more details on the ground and clipped highlights on the sky on the right (OK, I know it seems to be clipped on the left, but there completely no details on the right - just solid grey...)


 
0-255, ISO100, RGB x1.00 on the left; RGB x0.50 on the right (look on the bush and grass - they are more detailed). On the otherside white wall looks worse to me on the right (RGB x0.50) image. But maybe it is not so bad... I do not know...


So maybe lower RGB values means more details in dark and complex areas but as downside - DR is is lower.
 

What if rgb boost is actually boosting each channels iso somehow? ? Should it be tested also with negative values and try to shoot 6400iso or something like that... 
Example Red x1.99 is actually iso3200 x 1.99 = 6368iso for red channel...
I try to do my own tests also tomorrow. This is interesting.
 

I've done some tests today. The camera was set to UHD, 25p, Gamma DR, -5 contrast, -7 sharpness, 16-235, sRGB. I compared different ISOs with and without RGB set to max (1.99). 
 
So what I found out:
All RGB channels at 1.99 increase exposure by 1 stop (or very, very close to it);
Therefore ISO 1600 with RGB at 1.99, is the same exposure as ISO 3200 with RGB at 1;
It positively affects the inbuilt noise reduction. Seems like the NR really is triggered by ISO value and not the actual sensor gain;
It doesn't have any visible gains in highlights.
 
Here's the test. The password is - rgb .
Here you can download the edited test file, if you would like to take a closer look - https://failiem.lv/u/xvgh44ks
 

Yeah I'm excited to try this! After that debacle of bullshit fake-raw rumor thread, I'm glad there is at least something new to play with on these cameras. Nice work people!

Yes you are right. As I said in an earlier post the highlights are actually blown/clipping but at a lower level than 255, even though the scopes appear fine. I believe that the method of lowering the color settings and increasing exposure is failed, BUT in reverse: take color settings to +1.99 and reduce exposure (-1, -1.30) then this proves to have benefits, which you can read from the other posts in this thread.
I normally adjust colour as well to be different form standard as I found some colours were just over the top, so sure you can vary them slightly for the same benefit. But I have found that the colours improve when on full +1.99 and I find the colour variance is now not necessary. 
It may be April Fools day, but think of this as an Easter present, much better than an egg.

That is exactly what happens. You get 1 extra stop which reduces the ISO required for the same exposure and the noise reduces accordingly.

Thus is definitely worth looking at. I've done a few tests in even lighting. Can't say anything about the DR improvement, but it certainly increases exposure by about 1 stop. 
Someone asked for a boosted max iso hack. So there you go :D
Also, maybe there's a chance that it would have a positive effect on noise reduction. Like if the NR kicks in at 2000iso, you could use 1000iso instead for thr same exposure and not get the NR. That's just a thought for now.

I have lately been experimenting with video capture settings on the NX1 and I have been exploring a method of increasing dynamic range which I don't think has yet been touched upon.
It's really no big deal and I am not making any great claims but it may be of interest to some so I produced a quick video How-To.
Take is as fun, I have done very little real-world testing and it may be of little use, but over to you.
 

No boy has been mistreated in this forum.
We are a community, our mission is to share information.
We have asked for a shred of evidence to support the declarations.
He has lost more time writing excuses and posting screenshots than just uploading an original dng file and silencing everyone.
This is a very suspicious behavior in the light of his statements.
Furthermore, the tests provided, analyzed by expert photographers, have increased our doubts.
Who is publicly accused of lying defends himself by showing evidence to his defense and humiliating those who accused him of lying, in fact this was his reaction "I'll show you the video on Vimeo" statement never contradicted
Remember that the video on Vimeo does not constitute any proof, if this is his long awaited demonstration on April 1, it will still be a big disappointment for everyone.
Nobody asked to have the availability of the hack, just a simple test that would not compromise his work.
I think he is a prankster or in search of fame, or he wants to get a good sale, but this remains my opinion. nobody can feel offended by this.
I like everyone I hope that the raw video is true but I keep my feet on the ground because all the facts and the lack of evidence deny it.
I can not believe him by divine act of faith, and I discourage everyone from buying an NX1 camera if he believes he is shooting in raw.
Today, for the good of the community, I want to end this discussion to dispel doubts and misunderstandings once and for all.
Break the silence, today show everyone that you have developed a hack to shoot in RAW, even if still under development, share your information and I will send you for free an NXL adapter that is worth 429.00 euros, if you do not, you will prove that I am right.
Luca Butera.

Regarding untouched footage - I am not sure why it should be problem. You can always claim to Samsung layers that it was all just joke and footage was captured by eg. RED camera and EXIF of DNG files has been edited   However I can understand your worries in case you will release that hack for public usage. Needles to say that Magic Lantern team thinks that tuning own cameras is completely fine in same way as you can tune your own car or whatever. Moreover there were 2 hacks for NX1 in past and I think both guys who developed them have not faced any legal actions. 
I am sorry to say that but frame you have provided looks like heavily compressed and oversharpened H265. I really hope that real RAW sample is much better than this. 

Regarding untouched footage - I am not sure why it should be problem. You can always claim to Samsung layers that it was all just joke and footage was captured by eg. RED camera and EXIF of DNG files has been edited   However I can understand your worries in case you will release that hack for public usage. Needles to say that Magic Lantern team thinks that tuning own cameras is completely fine in same way as you can tune your own car or whatever. Moreover there were 2 hacks for NX1 in past and I think both guys who developed them have not faced any legal actions. 
I am sorry to say that but frame you have provided looks like heavily compressed and oversharpened H265. I really hope that real RAW sample is much better than this. 

Regarding untouched footage - I am not sure why it should be problem. You can always claim to Samsung layers that it was all just joke and footage was captured by eg. RED camera and EXIF of DNG files has been edited   However I can understand your worries in case you will release that hack for public usage. Needles to say that Magic Lantern team thinks that tuning own cameras is completely fine in same way as you can tune your own car or whatever. Moreover there were 2 hacks for NX1 in past and I think both guys who developed them have not faced any legal actions. 
I am sorry to say that but frame you have provided looks like heavily compressed and oversharpened H265. I really hope that real RAW sample is much better than this. 

@Arikhan Congratulations for your project with the NX1 Raw hack !!!  I think we are all looking forward to it.
But I think the same @Pavel Mašek , your frame is encouraging, but not very convincing. I see it very compressed, macroblock or miniblock?? and oversharpened. However, I trust you. I am very excited about the final result.

@kidzrevil + @lucabutera
Here is a screenshot from an early release. 
UHD | 25fps | 1/50 | Manual WB (blue hour like human eye sees it) | Zeiss O 28mm 1.4 at F7.1 | edit in post: slight WB correction | File edited with Premiere CC | Workflow recommended by @Mattias Burling --> here <--
 

Regarding untouched footage - I am not sure why it should be problem. You can always claim to Samsung layers that it was all just joke and footage was captured by eg. RED camera and EXIF of DNG files has been edited   However I can understand your worries in case you will release that hack for public usage. Needles to say that Magic Lantern team thinks that tuning own cameras is completely fine in same way as you can tune your own car or whatever. Moreover there were 2 hacks for NX1 in past and I think both guys who developed them have not faced any legal actions. 
I am sorry to say that but frame you have provided looks like heavily compressed and oversharpened H265. I really hope that real RAW sample is much better than this. 

If it really works then I bet that he solved it that way:
I think he basically wrote some app (NX1 is running on Linux-like OS Tizen) that is 'snapping screenshots' (maybe directly to DNG) from life view feed in video mode (he mentioned AF and IS works) - so it probably the similar thing as external recorders do. 
So image is already resized to 4K and minim. aperture is 1/30s.  That is why he mentioned same behaviour in low-light  (I think it is caused by way how NX1 resamples image to 4K). I would not expect REAL 6,5K 14bit CR2 RAWs but 'only' un-compressed 4K 10bit DNGs. I think it is only one possibility because otherwise would be bitrate huge as already mentioned here.
 

http://bokehmarket.com/blog/Best-Budget-Alternatives-Sony-A7II-Samsung-NX1-Fuji-XT-2-A7RII

Not sure that Tizen is "Linux-like" -- I think it actually is version of Linux.
 
 
 
This sounds like the old Magic Lantern HDR video in which alternating H264 frames have different ISOs  (WARNING!  STROBING/FLASHING FRAMES):
 
The trick is that you shoot with the ML HDR video feature at a higher frame rate (50fps or 60fps) and then combine every two consecutive images to get HDR video at half the frame rate (25fps or 30fps).  As mentioned in the video, one can encounter motion/ghosting problems with this method.
 
I wonder if the Magic Lantern HDR option works their 60fps raw video feature on the 5d III (although I am not sure if this HDR method makes sense when shooting raw).
 

@Pavel Mašek
It seems to be a misconception out there...You can't increase DR endlessly - this would be a perpetuum mobile and so...impossible. At the end of the day, output DR is limited by the maximum DR-capabilities of the sensor.
By stacking together you can only get more and more near to the DR limits given by the sensor - and get more maleability in post by gathering useful information to work with in highlights, mids and shadows. So, that's a kind of optimizing and NOT increasing. Only sensor manufacturers can increase the DR capabilties of a sensor...

Guys, a community always can have some more useful ideas than a single person. The problem is the technical implementation and the workflow after getting the results you want. One more problem could be the legal aspect...
As the processor of the NX1 is very powerful, I invented a workflow called IDOB - Image Data Optimization Blocks. It can combine 2 RAW images and output a single DNG file (16 bit) with a convenient amount of data (much less than a single RAW) but with a tremendous amount of DR. And it fits in a modern workflow with Premiere or Davinci.....
This ground principle is not my invention: When putting (stacking) more images together (even shot under same conditions = similar EV) you will get a single superiour image with great DR and tremendous possibilities to work in post.
Just try it: Go with your camera and shoot in quite dark conditions JPEGs in burst mode (even without EV change). Stack them in post to a single image together. Do some post work on it and compare this result with a edited single picture...By far more details and DR...Have fun...