Exploring Nikon D5200 HDMI output - review update

[hmcindie]

4:2:2 is not about detail in luminosity, but in chroma. As there is not much chroma information in this image, there is almost no chroma contrast/details/sharpness. We can't say for sure that it is 4:2:2.

 

The compression artefacts are 4:2:2 but just hoping that the camera is not wrapping 4:2:0 information in a 4:2:2 format.

 

Exactly. And almost every HDMI device tends to output a 4:2:2 signal, but is it actually 4:2:2 or 4:2:0 embedded in 4:2:2 is a different thing. 

[jgharding]

Just need to shoot a diagonal red line against a white and/or black background and compare internal with HDMI. That'll tell you one way or the other.

[QuickHitRecord]

@chauffeurdevan

@hmcindie

 

I am by no means an authority on this, but the examples shot by Andrew are consistent (at least visually) with what I have come to expect  of the difference between 4:2:0 and 4:2:2. Here's an example that I screen grabbed from a D800 with and without Ninja 2 external recorder that I downloaded from Vimeo a while back:

This is what I am seeing in Andrew's examples, especially when one looks at the lens center. Whether the D5200 is providing true 4:2:2 output rather than a re-wrapped 4:2:0 in a 4:2:2 signal, I have no way of knowing. But there is clearly more chroma detail in the first shot, wouldn't you agree?

[TwoLions]

@QuickHitRecord

Pretty sure what you're seeing isn't the result of the color space, it's the result of the massive overall bitrate increase going from H264 to ProRes.  Detail and clarity come from bitrate increase, color data change is more hard to discern.

[QuickHitRecord]

@QuickHitRecord

Pretty sure what you're seeing isn't the result of the color space, it's the result of the massive overall bitrate increase going from H264 to ProRes.  Detail and clarity come from bitrate increase, color data change is more hard to discern.

 

That makes sense. Maybe what I am focusing on is the macro-blocking and compression artifacts (and lack of it in the other image). I understand the basics of the science of color space, but I always thought that it was something that could be seen with the naked eye. Perhaps it's not.

[jgharding]

The differenc between  4:2:2  and 4:2:0 is only red channel resolution. That affects green though. The sub-sampling type represents the ratio between Y (luminance) Cb (blue) and Cr (red) resolution, crudely put.

 

Here's a great example:

 

[chauffeurdevan]

The differenc between  4:2:2  and 4:2:0 is only red channel resolution. The sub-sampling type represents the ratio between Y (luminance) Cb (blue) and Cr (red) resolution, crudely put.

 

 

Not exactly.

 

First, each Cb and Cr (could also be UV) channel represent the deviation from gray on an axis ( from wikipedia : C_B/P_B and C_R/P_R are deviations from grey on blue–yellow and red–cyan axes, whereas U and V are blue–luminance and red–luminance differences.) (The LAB colorspace is using the same principle)

 

So for example, the C_B/P_B channel representation in 8bit is at 0 - blue full saturated, 64 - blue half saturated, 128 - grey, 256 - yellow full saturated.

 

Second, in the 4-2-2 relationship, the 2-2 doesn't not represent the channels, but samples per row. The sampling formula always seems to be represented on by 4 pixel columns by 2 row. So in a 4-2-2 sampling, for a region of 4 pixel wide (4-x-x) by 2 (the second row - not related to the 2s in 4-2-2), you have on the first row, 2 samples of chroma (for both U and V, or Cb and Cr), and on the second row, 2 samples.

 

So, on a 4-2-0, on first row, 2 chroma samples of each channels, but no sampling on the second row (it is like have half resolution on both x and y axis of chroma versus luma, eg. 1920x1080 luma resolution, 960, 540 chroma resolution).

 

In 4-1-1, you have one sampling of chroma on each row (so for a 1920x1080 luma resolution, you get a chroma resolution of 480x1080).

 

in 3-1-1 sampling, like XDCAM EX and HDCam HD, you have for each 4 pixels of luma in source, only 3 samples of luma at the output, that is why those codec are 1440x1080 with a 1.33 pixel aspect ratio.

[see ya]

A couple of additional factors with Nikon DSLR h264 very similar to Canon is that chroma is normalised over the full 8bit range, +- 128 JFIF, the h264 stream is flagged full range so the decompressing codec must honour the flag and scale levels into 16-235 16-240 range for any conversion to RGB and also uses BT601 luma coeffs rather than BT709. Chroma placement is centered rather than to left. Nikon also uses BT470 transfer curve rather than BT709.

So VLC for example with HW acceleration on ignores full range flag and clips, QT ignores BT601 luma coeffs so pinks go to orange, blues to green.

All add to the mix with camera source comparisons and pixel peeping.

And does NLE when adding a BT601 luma coeff source like D5200 alongside a BT709 luma coeff source GH3 encoded out to typical BT709 h264 transfer the BT601 matrix to BT709? So resultant frame grab comparisons just done as if BT709 luma coeffs skewing color comparisons.

[Adl]

As far as I can tell, I don't see 4:2:2. In my opinion HDMI out is 4:2:0. For shure not a scientific comparison:

 

A small test-setting in front of greenscreen. I recorded intern and HDMI out to Ninja2. Three files to compare:

- HDMI out to Ninja2 ProRes 422 HQ 1080p29.97

- intern 1080p24, with 5DtoRGB transcoded to ProRes 422 (BT.601)

- intern 1080p24, ooc

 

All in timeline of Premiere 5 and took frame-grabs. A comparison:

I don't see an advantage from Ninja2 ProRes (HDMI) over 5DtoRGB ProRes (intern) in the frames and also not in chroma keying - just my 2 cents...

[zephyrnoid]

http://www.youtube.com/watch?feature=player_embedded&v=7JYZDnenaGc

[chauffeurdevan]

As far as I can tell, I don't see 4:2:2. In my opinion HDMI out is 4:2:0. For shure not a scientific comparison:

 

A small test-setting in front of greenscreen. I recorded intern and HDMI out to Ninja2. Three files to compare:

- HDMI out to Ninja2 ProRes 422 HQ 1080p29.97

- intern 1080p24, with 5DtoRGB transcoded to ProRes 422 (BT.601)

- intern 1080p24, ooc

 

Just opened your image in Afterfx and changed the channels to YUV.

 

Internal h264 recording is clearly 4-2-0. As for the Prores version, it is difficult to see as chroma change on each pixel. I would not be surprised if Apple is cheating by applying some chroma smoothness on their ProRes decoding.

[see ya]

AFAIK 4:2:0 is interpolated to interleaved 4:2:2 with QT. One of the plugin writers for Avisynth who created QTInput based on QT SDK simply couldn't get 4:2:0 out of it however hard he tried.

Adl can you post links to the three native files, personally I don't trust these sort of side by side comparisons via an NLE.

Just to mention regarding Canon DSLRs which also use JFIF spec like the Nikon DSLR's, internally within Canon cameras it is 4:2:2 raw sent to the h264 encoder. Magic Lantern have confirmed this and 4:2:2 raw frames can be saved to the memory card.

[Adl]

Here are links to the three native files:

http://85.214.28.227/briefkasten/D5200-HDMI-Ninja2-ProRes.mov

http://85.214.28.227/briefkasten/D5200-internal-5DtoRGB-ProRes.mov

http://85.214.28.227/briefkasten/D5200-internal-H264.mov

 

If you take frame-grabs from the internal footage, please take it from the middle of the file. The beginning and the end is a little bit blurred because of hitting rec button.

[QuickHitRecord]

Here are links to the three native files:

http://85.214.28.227/briefkasten/D5200-HDMI-Ninja2-ProRes.mov

http://85.214.28.227/briefkasten/D5200-internal-5DtoRGB-ProRes.mov

http://85.214.28.227/briefkasten/D5200-internal-H264.mov

 

If you take frame-grabs from the internal footage, please take it from the middle of the file. The beginning and the end is a little bit blurred because of hitting rec button.

 

Thanks for posting this. Maybe someone has a definitive way of finding out whether this is true 4:2:2 or not.

 

What are your impressions of this camera for video?

[Adl]

Did some further tests with greenscreen. Gave my best to produce messy hairdo :)

Compared HDMI out Ninja2 ProRes with internal transcoded to ProRes with 5DtoRGB

Sharpness in camera completely dialed down, no grading and post sharpening. So might seem a bit soft, but detail is there.

Only used ultrakey for chroma key with exact the same settings (standard), no further tweaking, just for testing.

100% crops:

Now it seems as if there would be more chroma information in the HDMI out. Could be a evidence for 4:2:2 ...

[hmcindie]

I looked at those original files in After Effects. It looks like they are both originating from 4:2:0 material. H264 version looks worse but color wise they are quite equal. Except in h.264 looks like there is some chroma smoothing but I guess that's normal.

 

Screencaps of red channel:

 

h264:

http://talvi.com/asiakas_ftp/temp_mikko/demo_h264.jpg

 

prores:

http://talvi.com/asiakas_ftp/temp_mikko/demo_prores.jpg

[chauffeurdevan]

Here are links to the three native files:

http://85.214.28.227/briefkasten/D5200-HDMI-Ninja2-ProRes.mov

http://85.214.28.227/briefkasten/D5200-internal-5DtoRGB-ProRes.mov

http://85.214.28.227/briefkasten/D5200-internal-H264.mov

 

If you take frame-grabs from the internal footage, please take it from the middle of the file. The beginning and the end is a little bit blurred because of hitting rec button.

 

Thanks, I looked at it at home.

 

Had to convert with ffmpeg the ProRes to non-compressed 4-4-4 as the ProRes decoding in AfterEffects smoothed the chroma.

 

It really seems like 4-2-2. Sharp block of 2x1 pixels in the U and V channels. It didn't look like an upscaled 4-2-0 to 4-2-2.

 

Too bad I didn't have time to get some grab.

 

So unless another expert from the forum can't prove otherwise, the hdmi out of the d5200 is 4-2-2. This is really nice.

[zephyrnoid]

I don't even understand why the question is so difficult to answer. If the 'capture format' is known that defines the outer limit of the color space, 4:4:4 being 100% uncompressed. Minute that you apply compression you can bet that color compression is going to be one of the losses, thus 4:2:2, 4:2:0 etc.
Unless your dealing with fully uncompressed color space there's not much point in talking about Chroma Key.
Let's focus on the overall look to the eye and the vectorscopes for grading in post.
Based on the side-by-side with the GH3 the D5200 footage looked visibly better in most shots with less obvious macroblocking in the deepest values.
We've been through this exact discussion before, when the Nikon D800 was first released.
Bottom line is we're looking for RAW out of HDMI.

[TwoLions]

So that means that when we hear "Uncompressed 4:2:2," like in the 5200, or "Uncompressed 4:2:0," like for the GH3, there is then a level of compression to get it to this color space.  I think I'm following this.

That's just the color information.

And it means we're really not getting Uncompressed, we're not getting all the pixel data, and of course we're not b/c the camera has to downscale the image from the 4k or 5k or whatever size image to HD size, so that's a level of compression.

So does anyone know what other kinds of compression, like to the detail of the image, happens before image usually gets to an HDMI port?

If I'm asking a stupid question that was answered last April/May when the D800 was released, I apoogize and can you point me to that forum page so I can read up.

[chauffeurdevan]

So that means that when we hear "Uncompressed 4:2:2," like in the 5200, or "Uncompressed 4:2:0," like for the GH3, there is then a level of compression to get it to this color space.  I think I'm following this.

That's just the color information.

 

I don't even understand why the question is so difficult to answer. If the 'capture format' is known that defines the outer limit of the color space, 4:4:4 being 100% uncompressed. Minute that you apply compression you can bet that color compression is going to be one of the losses, thus 4:2:2, 4:2:0 etc.
Unless your dealing with fully uncompressed color space there's not much point in talking about Chroma Key.
Let's focus on the overall look to the eye and the vectorscopes for grading in post.

 

There is a lot of mixed-up here. Chroma sampling has nothing to do with uncompressed, lossless..... You can have a lossy 4-4-4- compression codec and a lossless 4-2-0 codec.