D5200 Anyone got it ???

Not possible (23.97p) via Hyperdeck so far.. nor 24p as far as i know? Am i wrong? 'Cos if i am, i'm running to next store and will get one. Any enlighting regarding this? Any progress?
 
squig said:
Ok I think I've got the 60i figured out. I was dropping the 60i H.264 mov into a 24p fcpx timeline and it looked like shi# I had to do a prores transcode first (duh) and now it's displaying as progressive in the timeline with minimal artifacts.

24p has very little noise @ 100, 400, and 800 ISO. 60i is good @ 1600 ISO and even higher. These are my lens cap test findings, I still have to do some more real world tests to confirm.
Click to expand...

In Vegas, all you have to do is set the project to progressive and the footage is automaticly progressive - FCPX seems a bit behind :grin:

In any case, it's strange that interlace mode would make that much of a difference. Guess Nikon doesn't know what it's doing re video...
 
BTW, I'd love to post some footage I shot today, but I don't quite get permission to do so. All I can say I'm blown away by the quality of this plasticy camera. It looks so much better than any previous DSLR.
I know this is a stupid post without footage, but I had to say it.
 
macgregor said:
BTW, I'd love to post some footage I shot today, but I don't quite get permission to do so. All I can say I'm blown away by the quality of this plasticy camera. It looks so much better than any previous DSLR.
I know this is a stupid post without footage, but I had to say it.
Click to expand...
Post when you get the time! ...and check out Lpowell's post on Nikon's beta tool! Cheers Buddy!
 
CoolColJ said:
In Vegas, all you have to do is set the project to progressive and the footage is automaticly progressive - FCPX seems a bit behind :grin:

In any case, it's strange that interlace mode would make that much of a difference. Guess Nikon doesn't know what it's doing re video...
Click to expand...

Okay, but do you know how Vegas is doing this? That's rub, there are lots of ways to convert interlaced footage to progressive footage. But if it PsF footage, there is only one correct way construct the progressive frames. And it's often time no trivial task figuring out if that method is indeed being used.
 
Michael Schrengohst said:
Better than a GH2 or GH3??
Click to expand...

With controlled lighting for sure, much better colours and dynamic range but the fixed pattern noise is its achilles heal so it's not so great for run and gun in the dark. I tested that last night, I had the 5D MKIII set to 1250 ISO and the D5200 set to 800 ISO, the 5D footage was clean but I had to add grain to mask the FPN on the D5200. I still trying to figure a way around the FPN, 60i is much cleaner but it slightly crops the image and is a bit soft in fcpx. If anyone has a ninja I'd really like to here how well it does the 60i to 24p pulldown. Surely Nikon can tune the sensor to get 24p as clean as 60i. Maybe they'll get it right with the D7100.
 
macgregor said:
BTW, the internal mics are terrible. Worst sound quality ever. Not that it really matters...
Click to expand...

Hehe I haven't noticed, I've been recording my voice so I remember all my test settings. I was running a test with the mic off and I said "mic off.... fuc#in' idiot". lol.

How have you been setting your white balance, grey card?
 
Regarding use with the Ninja2, this is an excerpt of a PM conversation I had with someone who has one and has been trying to get it to work on the D5200:

First of all, my experience with recording 23.98 frames with Ninja2 is a kind of workaround. And it does not seem so stable. When recording it seemed stable. But without recording the Ninja sometimes lost the setting for 3:2 pulldown removal and I had to repeat the workaround. This is one of the reasons for a posible firmware update.

The second point: If camera internal recording is set to 1080p24 and the HDMI out is interlaced 1080i59.96 I don't know what is in the interlaced HDMI signal. In best case it's the 24p with a 3:2 pulldown wrapped in i60. This would be nice. Like for example the FS100 does:http://atomos.active...inja-with-fs100

In that case, Ninja2 can do the 3:2 pulldown removal and you get back the native 24p from internal.

The other possibility is: The D5200 just ignores the internal setting to 24p and the HDMI out is the same as if internal setting is 30p. Could be possible because HDMI out also ignores PAL setting. In that case, Ninja2 with setting 3:2 pulldown removal (with my workaround) is also recording 24 p. BUT it is combining the "wrong" interlaced frames. And I think that could result in worse image quality. As if you would transcode 30p to 24p.
Click to expand...

And yes, I am still following this thread. Even though I have returned the camera, there is a lot to like about it and I'm still hoping that someone will discover the magic bullet for the FPN issue.
 
I'm doing some extreme tests, I like to know how far I can push a camera before the image starts to fall apart. This is a good test, my living room has green walls and a green lounge, mixed daylight and tungsten, and light coming from a 24" display in the kids faces. It doesn't get much worse! But hey it's the only way I can get 'em to keep still. I tested the D5200 (neutral, sharpening 3, everything else 0). The 5D MKIII (prolost). Both set to 24p @ 3200 ISO f/4 with a 35mm Leica Summicron-R on the D5200, and a 50mm Summicron on the 5D. Everything was underexposed by over a stop. Transcoded to ProRes 442HQ and graded in FCPX with filmconvert and magic bullet looks. Everything Post sharpened except the ungraded shots. I haven't posted a magic bullet looks grade of the 5D shots because quite frankly it looks like shi#. I've picked a winner from this particular test but I'd like to hear what everyone else thinks. If everything looks like crap it must be your monitor. :cheesy: Notice how there's no noticeable fixed pattern noise @ 3200 ISO. No grain was added. I white balanced the D5200 with a grey card and did a 3500k kelvin temp by eye on the 5D.

D5200 ungraded
D5200 ungraded.jpg

5D MKIII ungraded
5D prolost ungraded.jpg

Notice how the D5200 set to neutral with 0 contrast is flatter than the 5D set to neutral with contrast at -4. Saturation is 0 on the D5200 and -2 on the 5D. If you go any flatter on the D5200 things start to get really ugly, good luck trying to grade it. I've got more testing to do but so far this is my secret sauce.

D5200 magic bullet looks grade
D5200 magic bullet.jpg

D5200 filmconvert grade (Epic M LUT, FJ SuperX 400 stock)
D5200 Epic M LUT.jpg

5D MKIII filmconvert grade (Canon Standard LUT, FJ SuperX 400 stock)
5D Canon STD LUT.jpg

I'll run the same test in daylight at when the bloody rain stops! It's going to be really interesting to see how these cameras compare in a really well lit scene.

You can check out the 720p jpegs in my post here- http://www.personal-view.com/talks/discussion/5118/nikon-d5200-topic/p19
 
Last edited:
hard to tell from you test, faces are too far away..
could you try to make a "good" image from both cameras, exposing them correctly, trying to get a good skin tone, resolution, skin gradation?
at iso 100 and 3200!
thanks for doing that!
tree way color corrector is the best thing in CC. MB looks look like crap, (even in Philip Blum tests), totally non real, in your case unnatural magenta, imho.
 
BobbyMurcerFan said:
I have to say that in the examples above, the flesh tones on the 5DMIII look appreciably better.
Click to expand...

Really? It really depends on your monitoring. I have 2 monitors, one is a cheap Samsung 24" but I've calibrated it well for film output, the other is my 17" Macbook pro which has been a bastard to calibrate but I do know how far off it is. How are you viewing it?

@ Oleg I'll upload the video shortly. It wasn't meant to be a good skin tone test (obviously) but I will do one, I might have to tie the kids down though. I'll do it in daylight at 100 ISO tomorrow if the rain stops, the geen walls in my house would fuc# it up! I'll upload the ProRes files to my google drive. I didn't put much time into that magic bullet grade but the filmconvert file looks good to me.

Please keep in mind everyone this test was designed to break these cameras, the test was also skewed in favour of the 5D MKIII. To match DOF the MKIII should have been set to 6400 ISO @ f/5.6, and the filmconvert LUTs were made for the 5D, the LUT I used for the D5200 was made for a RED EPIC!
 
I've used three different monitors a very expensive HP, a cheaper HP and an HP laptop, and it's the same. The Nikon's skin tones get lost in the couch or other colors in the room. They don't pop or look different from inanimate objects. This isn't the case with the Canon's skin tones; they look different, distinct and alive. Now can we draw any conclusions about the cameras? IDK, since it sounds like you're using quite the amalgamation of LUT's, LOL! But Canon has been for quite some time now for having excellent skin tones.
 
I did post the 720p ungraded shots on the link above, so you can judge for yourself from the ungraded shots, but like I said this wasn't meant to be a skin tone test, there's 3 different light sources (one of them changing) and green walls! It's an extreme uncontrolled lighting high ISO noise and picture profile test.

I've always thought Nikon has the edge over Canon in skin tones, the Canons have a magenta push that produces umpa loompa skin.

You can download the video of my test here https://docs.google.com/file/d/0BzJ3L6nv6Fn0b2NEb3lMR0xERXc/edit?usp=sharing
Don't bother watching it online, it's heavily compressed. You can download the ProRes file from the file menu.
 
Last edited:
You must log in or register to reply here.
Back
Top