Sekhar

OUCH. What a bunch of losers we all are.

OUCH. What a bunch of losers we all are.

The nice thing about low quality YT 720p video launches is: If people like it, "Imagine how great it would be in full Q!" and if it's lousy, "Hey, it was YT 720p, what did you expect?" Sorry, couldn't help the cynicism.

Looks great, nicely shot too. Nice colors, e.g., love the group shot at 1:56. Indoors look good too, though they had extra lights so it's hard to tell. There's some moire at 1:10, not sure if that will be an issue (didn't see it anywhere else in the video).

I'm talking of saturation, not the hues. If I remember right, the early test you did had odd colors with one or two lenses, and I thought it was more an issue with lenses. In any case, I'd put fixing color casts and hues in a different category from saturation level. If you have a lens that's consistently giving a bad cast, you might want to try a custom hue setting just for that lens or create a look or LUT in your color correction tool to apply each time in post.
Also, it looks like you're relying a lot on the LCD display for effect of different settings. It's only approximate, so if you want experiment, I suggest you actually capture with different settings and check. May be that's what you're doing, but wanted to confirm that. I'm saying that because I'm not seeing any of the artifacts you're describing with my NX500. BTW, you're right, I guess I was thinking of NX1 when I said 16-235.

First, I would leave the saturation at 0. If you find it over-saturated you can always turn it down in post, but turning it down in the camera and then pushing them back up in post doesn't seem like a good idea. Next, sharpness -10 and contrast -5 work well, but based on my experience, with both NX1 and NX500 you really need to protect the highs more so than the lows. If you have critical highs, consider using 16-235 instead of 0-255, though in the footage you posted I only see the shirt, which isn't critical, so lowering the exposure should have sufficed. Finally, try working with gamma in post rather than offset or gain, and within specific regions (lows/mids/highs).
But as TheRenaissanceMan said, it really depends on what you want to communicate. Below is a test I did with your video, with three variations:
First part is just with color correction, it will give a good starting point and is more a technical stage. A standard way to correct for skin is to mask just the skin area and then adjust till you get the line on vector scope to align with the skin line. That is what I did here, with increase in overall saturation. Next, I reduced the saturation just for the greens (secondary color correction) to highlight the person, might be useful if you want to bring focus to the person rather than the background. Just to give an idea of the creative step that typically comes after color correction. Finally, I warmed up the footage to give an overall look. This usually applies for the entire video (i.e., the final cut with all the clips). https://vimeo.com/138867742

Congrats on landing the budget for the doc! What's your project like? The a7 pair will work better for low light, but NX1 will definitely give you better colors (e.g., for nature work).

Another telephoto capture, this time of the full moon today here in Pasadena, CA.

OK, I don't get it. Looks soft, contrasty, and over-saturated. Sadly, even the models look like addicts, usually they make up for the footage. Those of you wowed by this, could you explain in specific terms what you liked? I mean specific terms, not generic stuff like "pleasing," "filmic," or "natural."

I'm not so sure about that.  I just think they're not willing to be on the cutting edge of video specs, because they can't really devote serious resources and development to it.  However, they can adopt the technology a little bit behind the curve.  They seem about 2.5 years behind.  So, eventually, they'll get there.
Now, if your thing is to have the bestest and coolest video IQ from a consumer stills cam, then you'll always be drawn to those models offering it.
I, however, tend to think that even median video IQ from 2015 cameras looks "good enough" already.  If you're a competent shooter, you can make most cameras work for you.  
Here's an example:  Last week I did a quick documentary style corporate shoot for 20K.  I hired a guy that I know likes to trick out his GH4 and likes to run around with a bunch of grip stuff.  Now, I did that not because I knew he was going to get wonderful footage, (I know his capabilities and that was expected anyway) but I needed him there so the set looked "professional."  And it did.  Big lens, matte box, c-stands, silks, reflectors, dead cats, audio harnesses, etc.  Okay.  So that put the client at ease.  Their money was on display.  Meanwhile, while he was doing all the A-Cam interview stuff and some misc B-roll, I was running around with a little EM5II grabbing shots non-stop.  In the end, over 90% of the b-roll shots in the final edit are from the EM5II.  Final delivery is for the web @720.  Unless you're super-discriminating, you can't tell the difference in the edit between the EM5II footage and the GH4 footage.  As a casual viewer you're just not really going to see it.  
So this is how I view the whole camera IQ debate.  What I can do practically matters.  I think I've come to terms with the fact that many people will always chase the technical holy grail with IQ, but my focus with many of my jobs (and my philosophy on how to approach those jobs) lies elsewhere.
And let's not loose sight of the fact that in 5-7 years we're all going to have robust DR, great low light capability, and 4K.  Sure, there will ALWAYS be superior IQ devices, but even if you have that capability, what are you going to do with it?  I DO get excited to shoot with awesome cameras, love being blown away with great images when I rent pro camera gear, and I know they are an absolute necessity for certain productions.  But not for all.  Not for your cat videos --or cityscape videos set to emo music.  
At a certain point when are we all just arguing how many angels can dance on the head of a pin?  I mean, even this lowly Oly camera has better IQ than the greatest auteurs of the mid 20th century had access to during most of their careers, so I say the tools are great and wonderful, just go use them effectively and you'll be happy.
Finally, as a camera manufacturer, nobody can outdo Fuji when it comes to offering lousy video in their stills cams.  They set the bar so low that it's unacceptable even for me!

Here is an example of why I think DIS is such an important breakthrough. This is a video of close-ups of insect life on flowers, shot with a 90mm (35mm-equivalent) macro lens handheld. You cannot efficiently shoot dragonflies, butterflies and bees at work with tripods - these insects move around and you have to "capture" them opportunistically - you have to be able to move after them with camera and shoot while they are in place. Without stabilization, the close-up video would be unwatchable - way too jitter and shaky. Well, here is what you can get using the NX1 with DIS and OIS combined:

 

Well, here's a very simple shoot as an example.  I did this for a friend over the weekend on a moments notice. It was basically a 1 hour shoot. I grabbed shots as fast as possible. 
Sure, I could have placed a tripod in a lot of spots and had the subject wait for me to compose... Or I could have let her do her thing and I would just follow, and still get cool shots along the way. 
With the EM5II, I just did the latter  
Slider shots without a rig, static shots without a tripod...it's a very nice capability to have. 
Granted, this isn't groundbreaking stuff, but this particular camera does give me some worthwhile footage. 
I don't know what else to say.  It's fun for me.  That's all  
 

It's shooting at 5 fps and interpreting as 25 fps to deliver at 25 fps as a convenience. If it delivered at 5 fps you'd have had to interpret it as 25 ftps manually in the editor, but the result would be the same. The point is that it's capturing at 5 fps, not 25 fps: it's impossible to capture any faster than 5 fps because by the time it's time to capture the next frame it'd still be capturing the first one.

Thank you for pointing this out as it often gets ignored: if you switch the NX1 to HD res it's almost a global shutter cam.
When Cinema5d compared shutter speeds in the following chart they had the nerve to show the NX in UHD mode only on the far right - indicating it's the clear loser, while the A7R2 is given its HD crop speed, and the Amira winning even though it's only 2k. In HD the NX would be to the left of the chart. There is a lot of bias against this camera out there.
 

The main flaw is Samsung's crap lens lineup and lack of decent adapters. Outside of the two zooms there nothing great from a video standpoint because all those goofy pancakes suck when you want to manually focus. Sony is moving quickly to fill out its lens lineup and is announcing 7 more lenses this year, along with stellar entries from Zeiss - plus we already have a plethora of electronic adapters and speedboosters. Samsung? Nothing since the NX1 was announced. 

Is it just me or does anyone els feels like the "bouncing" feel is a dealbreaker ? 
This motion that goes up & down while you walk is kind of really weird and unnatural :/ 

I think you need to get an NX1 and try for yourself. The DIS is great in some situations. It's a free software addition from Samsung. It never was a selling point for the NX1. But in certain situations when you have to be handheld, or when you have particular lenses, it can come in very handy. I definitely prefer it over premiere warp stabilizer (espescially in terms of time management).

And yes there is some rolling shutter at 4K. If you really need long focal + handheld you might want to record in FULL HD since the NX1 is one of the best camera for rolling shutter management in Full HD. 

Also the detail resolving is so great with the NX1 that you dont feel a lost in quality with the DIS crop. Now what's great, is that if you don't like it, you don't have to use it. 

Yes, we know both capture the same info (nothing gets dropped at the top or bottom), the difference is in how they get ENCODED. I.e., which part (lows, mids, highs) gets preference in terms of space on the 8 bits. See comments on the page for the difference and effect on banding.
Here's my summary and recommendation:
0-255. Works for most situations, especially recommended when mids are more important than lows and highs. Uses the entire 8 bit range, so it's the best you're going to get for mids and overall image. Not good if you aren't planning post work. 16-235. Necessary when delivering out of the camera without post work. Also good on high contrast scenes where you want to preserve details in lows and highs at the cost of mids. 16-255. Good when you have critical info in the lows and can't lift the overall exposure because it will blow the highs.

Yes, we know both capture the same info (nothing gets dropped at the top or bottom), the difference is in how they get ENCODED. I.e., which part (lows, mids, highs) gets preference in terms of space on the 8 bits. See comments on the page for the difference and effect on banding.
Here's my summary and recommendation:
0-255. Works for most situations, especially recommended when mids are more important than lows and highs. Uses the entire 8 bit range, so it's the best you're going to get for mids and overall image. Not good if you aren't planning post work. 16-235. Necessary when delivering out of the camera without post work. Also good on high contrast scenes where you want to preserve details in lows and highs at the cost of mids. 16-255. Good when you have critical info in the lows and can't lift the overall exposure because it will blow the highs.

The new pilotfly H1+ is out with dual 32bit IMU. 

I think this is a game changer. Put your A7RII, RX100IV or D5500 on it and you get pro stabilized footage. Lots of creative possibilities (hyperlapse, fast movement while shooting in slowmo, ...)
What do you guys think?

Here more info:
https://www.divergentmedia.com/blog/samsung-nx1-h265-luminance-ranges/
 

Yeah, but trying to nail down the specifics, like what lens (Samsung vs. other, OIS vs. no OIS, 1.02 vs. 1.03, etc.), what situations (long vs. short), what kind of movements, etc. I will experiment myself and post, and hopefully we can put all the experiences together to come with a guide of sorts. Everyone with NX1: please test and post your experiences/footage!

over time, you develop an ear for this sort of thing. as a trailer editor, it's pretty much mandatory to be a good music editor. rather than relying on some plugin, looking at the waveform can help you make a picture or audio cut (or cuts) to the beat/percussion/etc. personally, i don't need to place markers because the waveform is right there.

I didn't watch Coreys tut, but I know one thing: You never cut precisely to the beat if you want the greatest emotional impact. You cut to the rhythm, and you do it intuitively. 
Don't believe me? Watch any good music video and tell me how many edits are precisely on the beat!

Shot handheld with DIS + OIS; NX 16-50mm S lens.
https://vimeo.com/136972283