Samsung NX1 - which is 4K video and which is the 28MP raw still? Can you tell?

[Vesku]

I made GH4 100% sensor readout test with different modes:

 

http://***URL removed***/forums/post/54849358

 

1. 4k center crop

2. 4k 2x dig zoom downsampled to fullhd

3. ETC fullhd

4. best JPG center crop 1920x1080

5. RAW (ACR) center crop 1920x1080

[BookMark]

i got it right, so what's the prize? :P 

 

joking aside, it's definitely an interesting offering. i read that it can do full 6.5K full sensor readout, but in real world application, can that be used, say some magic-lantern-type hack ? 

[neverending]

A is 1.6MB video frame source

B is 2.0 MB raw processing

[astrotripper]

First post fail.  Sorry about that. lol

 

:D  I'm perfectly fine with being totally wrong on this one. It's just impressive, that a single video frame can be that good.

[IronFilm]

I'm simply amazed that such a simple post generated such a long thread with so many posts discussion A or B! wow

[ND64]

Damn.. I cant believe A is video!
Well done Samsung

[NOC40]

 Matt James Smith's original comment on the colour info was all you needed folks.... sharpness<>RAW

[John Wendell]

First, a disclaimer, my monitoer is crummy. At first I thought "A" was the video because there seemed to be less color detail. But when I pixel peeped, it was clear that "A" was MUCH sharper. Yet "B" seems to have more detailed color information. It remindes me of the difference between sRGB and Adobe RGB. I don't understand color space that well but when I export an image from Lightroom as sRGB and then Adobe RGB, this is the type of difference I see where the sRGB has more saturated colors but obviously less color detail. Now I think it is a trick question. I'm supposing Andrew deliberately blew focus on "B" to confuse us. So I'm sticking with my first guess. "A" is video and "B" is from RAW but out of focus.

[DBounce]

So what did I win?

[Andrew Reid]

So what did I win?

 

A sense of pride in your existence :)

[BenjaminJ]

I think it doesn't have so much to do with the color space, but with tone mapping or camera profiles. The same RAW file will give different colors in different RAW developing applications for this reason (because they use different camera profiles).

 

Regarding the worse sharpness of image B: it's because of the CA of the lens -- different colors get different magnification and this reduces sharpness, but this can be corrected in software quite effectively by scaling the different color channels. The NX1 does this in-camera very effectively, as you can see in image A.

 

First, a disclaimer, my monitoer is crummy. At first I thought "A" was the video because there seemed to be less color detail. But when I pixel peeped, it was clear that "A" was MUCH sharper. Yet "B" seems to have more detailed color information. It remindes me of the difference between sRGB and Adobe RGB. I don't understand color space that well but when I export an image from Lightroom as sRGB and then Adobe RGB, this is the type of difference I see where the sRGB has more saturated colors but obviously less color detail. Now I think it is a trick question. I'm supposing Andrew deliberately blew focus on "B" to confuse us. So I'm sticking with my first guess. "A" is video and "B" is from RAW but out of focus.

[Andrew Reid]

Turns out Adobe Camera Raw does have a setting for the NX1 + Samsung 30mm F2.0, I just have to turn it on to remove the fringing wide open. Quite a nice lens for 250 euros if you need the AF and small size. It weighs practically zero.

[majoraxis]

I'm very interested in the AF performance of the 30 F2.0 for video Samsung prime lenses.  From the videos I have watched, the Samsung zoom lens with Ultra-Precise Stepping Motor (UPSM) seem to provide the best tracking AF performance. It's claimed that UPSM improves upon the conventional Stepping Motor (SM), offering three times the precision in its ability to focus on subjects...

 

I hope someone will do a test that covers the AF performance in video for at least a zoom and a prime for the Samsung lenses.

[majoraxis]

Here's a youtube video comparing the AF of the Samsung NX1with the 45 f1.8 prime as again the GH4 with the 12-35.... both at 2.8 - the DOF is shallower on the NX1 because of the crop factor.  Looks like the AF on the 45 is pretty good - it did not expect it to perform as well.  Looking forward to more comparisons, but this looks pretty hopeful to me regarding how this particular prime as it performs with AF.

 

[vas]

Here's a youtube video comparing the AF of the Samsung NX1with the 45 f1.8 prime as again the GH4 with the 12-35.... both at 2.8 - the DOF is shallower on the NX1 because of the crop factor.  Looks like the AF on the 45 is pretty good - it did not expect it to perform as well.  Looking forward to more comparisons, but this looks pretty hopeful to me regarding how this particular prime as it performs with AF.

 

The AF is pretty damn amazing on the NX1 in video mode.

[majoraxis]

Yes!

 

NX1's AF is pretty amazing. A few software tweaks for rack focus style transitions and it will be ideal.

 

The fact that it can do great AF in 120 fps is a big deal as shown in the last half of the video.

 

I would like to see 240 fps at 720p with contentious auto focus.  Hopefully it will be possible with a firmware upgrade as well - we'll see.

 

 

 

 

[Joe Person]

A is raw--blurrier, perhaps due to clicking shutter button?

B is video--because A is raw.