D800

  • Thread starter Thread starter Guest
  • Start date Start date
GRADED FULL.jpgSo I decided to hop on the grade the D800 footage bandwagon. I have to say, I pushed it around pretty hard with Colorista and I was rather impressed. Here is a grab of my results.
 
Bruce/Geronimo: Thanks for all your work. What picture control were you using on the eyebrow shot? What happens if you turn down the sharpening?
 
Last edited:
I've just watched Philip Bloom's updated 5D3 review, and perhaps what PB has learned with the 5D3 applies to the D4/D800 as well.

The arithmetic for downsampling to 1920x1080 is much better in the 5D3 than the 5D2. Rather than decimate (line-skip), the pixels are first low-pass filtered to eliminate (well, greatly reduce) aliasing and moire artifacts. The result is clean, but looks 'soft' compared to the unfiltered 5D2 line-skip results. PB then graded and sharpened the 5D3 material in post, and sees that all the 'missing' detail and sharpness is entirely recoverable (and it's real, not just artifacts).

PB's recommendations: Turn in-camera sharpening all the way down, and perform all sharpening in post.

I wonder if the same advice might be useful for the D4/D800. The underlying principles are cetainly the same for all of these cameras.

-- Bob Elkind
 
Last edited:
PappasArts said:
Definitely odd, since the video posted here, as well frame grabs don't exhibit the same. http://www.dvxuser.com/V6/showthrea...1080P-video-tests&p=1986112841#post1986112841
Click to expand...

The video and screenshots posted there don't show any eyebrow, and she is mostly wearing sunglasses. Lighting setup is also different.

However, I'm not sure why a difference that would be odd. Just because aliasing shows up on one house roof, for example, doesn't mean that it will appear in all house roofs.
 
I've seen aliasing in just about every D800 video. I love the colours, love the dynamic range, rolling shutter looks like it's better than the MKIII too. I'm going with the MKIII for now but I'm keeping a close eye on the Mosaic filter developments...I could flip flop on this at some point.
 
bufo55 said:
Bruce/Geronimo: Thanks for all your work. What picture control were you using on the eyebrow shot? What happens if you turn down the sharpening?
Click to expand...

Happy to do it. :)

All of my tests were done with Neutral picture control. Will play with sharpening at my next opportunity.
 
eteam said:
I've just watched Philip Bloom's updated 5D3 review, and perhaps what PB has learned with the 5D3 applies to the D4/D800 as well.

The arithmetic for downsampling to 1920x1080 is much better in the 5D3 than the 5D2. Rather than decimate (line-skip), the pixels are first low-pass filtered to eliminate aliasing and moire artifacts. The result is clean, but looks 'soft' compared to the unfiltered 5D2 line-skip results. PB then graded and sharpened the 5D3 material in post, and sees that all the 'missing' detail and sharpness is entirely recoverable (and it's real, not just artifacts).

PB's recommendations: Turn in-camera sharpening all the way down, and perform all sharpening in post.

I wonder if the same advice might be useful for the D4/D800. The underlying principles are cetainly the same for all of these cameras.

-- Bob Elkind
Click to expand...

I think it goes double for the D800, since it has more artifacts. In camera sharpening should probably be turned all the way down.

One other unknown variable (besides pictures styles, sharpening etc) in a lot of the low light videos popping up is the noise reduction settings. A few have stated that the high iso noise is much more film like with noise reduction turned off. I'd like to see some comparisons between the different noise reduction settings.
 
Geronimo said:
The video and screenshots posted there don't show any eyebrow, and she is mostly wearing sunglasses. Lighting setup is also different.

However, I'm not sure why a difference that would be odd. Just because aliasing shows up on one house roof, for example, doesn't mean that it will appear in all house roofs.
Click to expand...

The odd part is with the 5DMII and my personal mileage, it was consistent for me in the aliasing issues. The D800 does it too, just it hides much better in my opinion compared to the 5DMII. As for the hair, I was referring to the very minimal aliasing in her hair. Aliasing is there, just not as bad as we had on the 5DMII for 3 1/2 years with the same type of shots. Of course I wish it wasn't there at all. Maybe that OLPF might get designed for the D800, and change this. No matter what, these tools are still quite nice.

This was a screen capture from the Vimeo page ( no down loadable file on this one ) of one of the fabric shots in the footage.


Fabrics at end.
https://vimeo.com/39274584

The girl talking.
https://vimeo.com/39358549

Pappas
http://pappasarts.wordpress.com/
http://twitter.com/pappasarts
 
Last edited:
With the mention of people adding USM to the 5Dmk3 video to counter the soft images, I decided I'd give it a shot with the D4 portion of a couple of the D4/D800 comparisons.
http://vimeo.com/39397906
http://vimeo.com/39398244

My goal for those two was to see if I could make the D4 video (soft but lacks aliasing and noise) look as good as, or better than, the D800 footage. I think it turned out pretty well.

Here I added USM to both the D4 and D800 footage.

http://vimeo.com/39399277

Pretty identical imo.
 
Last edited:
I am asking myself if it is not a problem with the lens like on the gh2. I mean the gh2 does not alias or moire (It can as any camera, but should be in very very rare case) but put one of those Panasonic lens and the result is very video harsh looking image. It has been seen that there was lot of processing that was done when the gh2 detect one of those lens, to sharpen the image and do lens correction like vignetting etc. It has also been tested that the render of the image was nicer by just masking the Pany lens so that the camera did not know what lens it was. When you put some more vintage nikon or Canon glass the image becomes very organic and nice. It might not be the case for the Nikon D800 but would be nice to see if softer lens and even not nikon ones could decrease the aliasing/moire.
 
Danielvilliers said:
It might not be the case for the Nikon D800 but would be nice to see if softer lens and even not nikon ones could decrease the aliasing/moire.
Click to expand...

Interesting thought, glass of course can play role like a OLPF in some ways.

Here's a frame grab of some trees. Download the 1080P for best quality.


https://vimeo.com/39390343

And this D800 Moire test. Download the 1080P for best quality.

https://vimeo.com/39392212


Pappas
http://pappasarts.wordpress.com/
http://twitter.com/pappasarts
 
eteam said:
I've just watched Philip Bloom's updated 5D3 review, and perhaps what PB has learned with the 5D3 applies to the D4/D800 as well.

The arithmetic for downsampling to 1920x1080 is much better in the 5D3 than the 5D2. Rather than decimate (line-skip), the pixels are first low-pass filtered to eliminate (well, greatly reduce) aliasing and moire artifacts. The result is clean, but looks 'soft' compared to the unfiltered 5D2 line-skip results. PB then graded and sharpened the 5D3 material in post, and sees that all the 'missing' detail and sharpness is entirely recoverable (and it's real, not just artifacts).

PB's recommendations: Turn in-camera sharpening all the way down, and perform all sharpening in post.

I wonder if the same advice might be useful for the D4/D800. The underlying principles are cetainly the same for all of these cameras.

-- Bob Elkind
Click to expand...

USM simply cannot bring out detail which is not there. And detail cannot "somehow be encoded in the blur". If its blurred, its blurred. It makes an illusion of a sharp image, but these are two different things.

Ive been experimenting with video for only a year, but i've had my fair share of still work. And USM just makes images look fake. But there are people who like it and sharpen the hell out of their pictures.

Compare a wall sized print from 6mp with sharpening and a 16mp print with 0 sharpening. Same thing with video... Mk3 vs C300.

Another analogy (taking it to the extreme) is applying DOF in post... yikes.
 
Last edited:
latyshev said:
USM simply cannot bring out detail which is not there. And detail cannot "somehow be encoded in the blur". If its blurred, its blurred. It makes an illusion of a sharp image, but these are two different things.

Ive been experimenting with video for only a year, but i've had my fair share of still work. And USM just makes images look fake. But there are people who like it and sharpen the hell out of their pictures.

Compare a wall sized print from 6mp with sharpening and a 16mp print with 0 sharpening. Same thing with video... Mk3 vs C300.

Another analogy is applying DOF in post... yikes.
Click to expand...

True, sharpening can't create detail that is not there to begin with. However it is a visual trick of enhancing edge contrast, and as such sharpening in post in still photography is usually needed to some extent. How much and how depends on many factors from how the photo was shot (camera make and settings, raw vs jpg etc.). All jpgs coming out of the camera will have digital sharpening applied to some extent. Raws are usually softer out of the camera, and resizing will make them even softer. Finally your final format will require different sharpening, print often requires more sharpening than the computer screen etc.

When it comes to DSLR video with all its limitations, the sharpening algorithms are often very crude due to data processing restraints, so shooting soft and post sharpening it is often the better way to go.

So in short, no amount of sharpening is gone turn our DSLRs into a C300, since the resolution just simply isn't there, but it still an important consideration when working with digital images.
 
Last edited:
You must log in or register to reply here.
Back
Top