KarimNassar

We don't have any white walls at home, but I think you can see what you are asking in the trim. The first is the NX1 shot in UHD transcoded to 422, the second is the VA recorded in 422. If you're looking for the blocks in the shadow areas, I've seen alot less with the VA so far. 

took the nx1 on a first test run this afternoon, not confortable with video grading yet and I noticed too much red and magenta in the skintones highlights so far.
colors "feel very thin" to me already when grading so I will not lower the saturation in camera, happy with the DR and details.
All available light, should receive some led panels soon that will make things more interesting. All the same lens, Nikon 20mm f 2.8D
4kdci, gamma C, maximum black level, minimum contrast, minimum sharpness. I will try to upload to youtube in 4k.

Allow me to explain further rimpamposh, because that video is explaining the situation in my opinion in a twisted and counter intuitive way IMO.
Lets look at the RGB waveform of this NX1 file:

If you look at the top of the waveform it seems the highlights are clipped. However pay attention to the bottom right Clamp Signal checkbox that is checked on and that I circled in green. That clamp signal checkbox, clamps the display of the highlight. It does not DISPLAY in the waveform scope the highlights above the 100 line. However the information is in the video file, it is simply not displayed because that checkbox is checked.
 
Lets have a look at what our waveform looks like once we uncheck it:

As you can see we have more highlight information visible in the waveform scope than what was previously displayed. 
And that is all that has changed, the clamped highlight information above 100 is now displayed in the waveform monitor only. In our video it is still out of range because it is above the 100 line, so in our video those highlights are still clipped. How do we recover that information and unclip the highlights in the video? 
This is where in the video he says to set the 235 number. You do not need to do specifically do that, and you can do it with a variety of tools. And that specific amount will not be adapted to every situation. All you want to do is bring the highlights down. That is all. In this instance I did it as following:

You can do it in different ways. Hope it helps.

Allow me to explain further rimpamposh, because that video is explaining the situation in my opinion in a twisted and counter intuitive way IMO.
Lets look at the RGB waveform of this NX1 file:

If you look at the top of the waveform it seems the highlights are clipped. However pay attention to the bottom right Clamp Signal checkbox that is checked on and that I circled in green. That clamp signal checkbox, clamps the display of the highlight. It does not DISPLAY in the waveform scope the highlights above the 100 line. However the information is in the video file, it is simply not displayed because that checkbox is checked.
 
Lets have a look at what our waveform looks like once we uncheck it:

As you can see we have more highlight information visible in the waveform scope than what was previously displayed. 
And that is all that has changed, the clamped highlight information above 100 is now displayed in the waveform monitor only. In our video it is still out of range because it is above the 100 line, so in our video those highlights are still clipped. How do we recover that information and unclip the highlights in the video? 
This is where in the video he says to set the 235 number. You do not need to do specifically do that, and you can do it with a variety of tools. And that specific amount will not be adapted to every situation. All you want to do is bring the highlights down. That is all. In this instance I did it as following:

You can do it in different ways. Hope it helps.

save file on computer and open in photoshop. That has only to do with the color shift and different color management of the different web browers. As long as the file has an attached color profile by opening it with photoshop you are ensuring it is displayed correctly.
it's not a conversion, all he is doing in that video is bringing the highlights down. If you want to understand what is going on, look at the bottom right of the video ( the red green and blue waveform) as he sets the output level to 235 the highlights shift down. The fact that he calls it "an error" and always sets the number 235 makes it unnecessarily confusing. All he is doing is bringing the highlights down to recover highlight information, you can do that with a variety of other tools as well. And always setting to 235 as if it is a magic number fix to that "error" is also unnecessary. Depending on the amount of highlight information that is above the displayed range it can be more or less.   

Allow me to explain further rimpamposh, because that video is explaining the situation in my opinion in a twisted and counter intuitive way IMO.
Lets look at the RGB waveform of this NX1 file:

If you look at the top of the waveform it seems the highlights are clipped. However pay attention to the bottom right Clamp Signal checkbox that is checked on and that I circled in green. That clamp signal checkbox, clamps the display of the highlight. It does not DISPLAY in the waveform scope the highlights above the 100 line. However the information is in the video file, it is simply not displayed because that checkbox is checked.
 
Lets have a look at what our waveform looks like once we uncheck it:

As you can see we have more highlight information visible in the waveform scope than what was previously displayed. 
And that is all that has changed, the clamped highlight information above 100 is now displayed in the waveform monitor only. In our video it is still out of range because it is above the 100 line, so in our video those highlights are still clipped. How do we recover that information and unclip the highlights in the video? 
This is where in the video he says to set the 235 number. You do not need to do specifically do that, and you can do it with a variety of tools. And that specific amount will not be adapted to every situation. All you want to do is bring the highlights down. That is all. In this instance I did it as following:

You can do it in different ways. Hope it helps.

I just tested and 16-235 appears to have more banding than the full range. 
I've read some people say it captures the same amount of shades than the full range and is just a read information for premiere or your editing software but it appears it is not accurate or there is some other issue.
Anyways don't see the point of using 16-235 in first place? Nothing wrong with the full range to begin with.
save file and open in Photoshop, some color shift with internet browsers.

I just tested and 16-235 appears to have more banding than the full range. 
I've read some people say it captures the same amount of shades than the full range and is just a read information for premiere or your editing software but it appears it is not accurate or there is some other issue.
Anyways don't see the point of using 16-235 in first place? Nothing wrong with the full range to begin with.
save file and open in Photoshop, some color shift with internet browsers.

Ok here are some grabs from the scene. In some cases the banding I would honestly say is "acceptable" though not ideal.. and maybe thats just me being a little generous.
The grade contains about 60% of neat video denoising, which is more aggressive than what I really ever use. I was hoping it would help mitigate some of the banding problems.. and it does to some limited extent.
Jpegs:

Tiffs for better viewing:
 NX1_Color_Test_Close.tif
 NX1_Color_Test_Close_Graded.tif
 NX1_Color_Test_SideL.tif
 NX1_Color_Test_SideL_Graded.tif
 NX1_Color_Test_Wide.tif
 NX1_Color_Test_Wide_Graded.tif
Camera settings as follows:
Gamma DR, Contrast -10, Sharpness -10, Saturation -4, MBL +10, range is 16-235
Shogun set to capture in ProRes HQ
Now I'm definitely a noob to this camera (thus the testing) so I must ask, is there some other setting somewhere that I need to make sure is adjusted properly? Like a picture profile or something? HDMI is set to 3840x2160 (24) clean out.
Tomorrow I'm going to try the test again. I'll shoot/edit quickly to test different settings. What values of above should I change? I am planning on dropping contrast to 0 (I actually did drop it to -5 to test one shot, and the banding was still prevalent) and adjust MBL closer to 0, which seems a shame because in my earlier tests (as posted in this thread) I was able to get a TON of usable detail out of the shadows... but the banding just might not be worth it.
Also maybe you can edit the tiffs to have less banding, if so, how?
Please share your thoughts.

if you could setup a shot on a tripod that introduces banding and macroblocking issues, such as for example a white wall next to a window that is a little underexposed with gradual fall off of the light on the wall, to compare the internal vs 10-Bit 4:2:2 shogun that would be great. I would assume it makes no difference since the output of the nx1 is not 10bit 4:2:2 but would be interesting to check and compare.

if you could setup a shot on a tripod that introduces banding and macroblocking issues, such as for example a white wall next to a window that is a little underexposed with gradual fall off of the light on the wall, to compare the internal vs 10-Bit 4:2:2 shogun that would be great. I would assume it makes no difference since the output of the nx1 is not 10bit 4:2:2 but would be interesting to check and compare.

This crippling of cameras capabilities on purpose is getting really ridiculous and insulting.
We know how much of a scam it really is since the 2009 mark II is 14 bit raw capable thanks to magic lantern,
- The gh2 capable of much higher quality compressions thanks to the hack.
- The sony F5 internal 4k capability with a simple line change edit in the config text file.
I'm under the impression we have come to a point in camera technology where the entry line cameras are fully capable of producing at least 4:2:2 color, possibly at 10bit depth and definitively with higher quality compressions. Problem is with this level of image quality the enormous price jump of the high end cameras would be hard to justify even with raw formats. They just have to mess up the cheaper ones somehow.
Since blackmagic strategy is to offer the most features at the best price to compete with the market, and they don't have such a crowded product line to have to apply this product segmentation garbage I believe they are a good comparison:
Ursa mini $4,995                               Sony fs5 $5,599.00
ProRes XQ 4:4:4  - 250 MB/s              XAVC Long 4:2:0  - 100Mb/s with macro blocking nightmare
It's 2016, you pay 5'600usd for macroblocked 4:2:0 8 bit. Come on now.

This crippling of cameras capabilities on purpose is getting really ridiculous and insulting.
We know how much of a scam it really is since the 2009 mark II is 14 bit raw capable thanks to magic lantern,
- The gh2 capable of much higher quality compressions thanks to the hack.
- The sony F5 internal 4k capability with a simple line change edit in the config text file.
I'm under the impression we have come to a point in camera technology where the entry line cameras are fully capable of producing at least 4:2:2 color, possibly at 10bit depth and definitively with higher quality compressions. Problem is with this level of image quality the enormous price jump of the high end cameras would be hard to justify even with raw formats. They just have to mess up the cheaper ones somehow.
Since blackmagic strategy is to offer the most features at the best price to compete with the market, and they don't have such a crowded product line to have to apply this product segmentation garbage I believe they are a good comparison:
Ursa mini $4,995                               Sony fs5 $5,599.00
ProRes XQ 4:4:4  - 250 MB/s              XAVC Long 4:2:0  - 100Mb/s with macro blocking nightmare
It's 2016, you pay 5'600usd for macroblocked 4:2:0 8 bit. Come on now.

ursa mini weight is 5lbs, red weapon is 3.3lbs and it is said that the Raven will be lighter
weapons bare body is much smaller than the ursa mini bare body, much more compact and better suited for movi type stabilizers.
image quality wise I would be surprised if the new 4.6k ursa sensor performs better than the red dragon sensor but wait and see.
but the price of a working red raven package will likely be at least twice that of a ursa mini 4.6k though.

ursa mini weight is 5lbs, red weapon is 3.3lbs and it is said that the Raven will be lighter
weapons bare body is much smaller than the ursa mini bare body, much more compact and better suited for movi type stabilizers.
image quality wise I would be surprised if the new 4.6k ursa sensor performs better than the red dragon sensor but wait and see.
but the price of a working red raven package will likely be at least twice that of a ursa mini 4.6k though.

big news today:
- form factor is weapon sized, smaller than scarlet dragon
- lightest of all their cameras ever made
- redcode at 5:1 compression
- dragon sensor "a bit shy of Academy 35mm and APS-C. All your APS-C / S35 lenses should cover."
- canon ef mount with electronic control
- sub 10k price
now is sub 10k for the body with ssd module included, I would assume so but we will find out friday.
 

 

It's all in what you do with it...

So it's named Raven.
still no specs nor price for now.

So it's named Raven.
still no specs nor price for now.

the reason there is no difference is because you are using a grayscale image so there is only luma information in your image so nothing gets subsampled.
It is the foundation of how chroma subsampling works:
"Chroma subsampling is the practice of encoding images by implementing less resolution for chroma information than for luma information, taking advantage of the human visual system's lower acuity for color differences than for luminance"
only color information gets discarded when using chroma subsampling, in your images there is none.
4:2:0 4:2:2 4:4:4
^ first number is the luma component, as you can see there is no reduction no matter the chroma subsampling.

I would LOVE for the Ursa mini 4.6k to be a great camera and I would buy it if that is the case.
However I keep reading over and over again complaints about corrupt files, camera crashes, fixed pattern noise unusable images, and overall unreliability of the bm cameras that I have zero expectations.
Now we are talking red, industry proven cameras with amazing R3D format, yes the accessories will be expensive, but they might actually deliver a reliable workhouse in the price range that blackmagic has over hyped and failed over and over again due to severe issues on all models.
Like you guys said it will come down to what an actual working package will cost, but this is seriously interesting news.

8 bit jpeg off the highest dynamic range dslr sensor available, thus why they look good.
And I did say give us "better dynamic range" not only higher color sampling and more bit depth.
I'd take a high dynamic range 1080p image over a poor DR 4k one anyday. but to each his own.

yeah well if they could get into a dynamic range and color depth war that would be great thanks.
can't wait to get my compressed 8k 4:2:0 8 bit image. so exciting.

yeah well if they could get into a dynamic range and color depth war that would be great thanks.
can't wait to get my compressed 8k 4:2:0 8 bit image. so exciting.

8 bit jpeg off the highest dynamic range dslr sensor available, thus why they look good.
And I did say give us "better dynamic range" not only higher color sampling and more bit depth.
I'd take a high dynamic range 1080p image over a poor DR 4k one anyday. but to each his own.