Why are we talking about a $6000 camera on a site geared toward student multimedia journalists? Because the advances, the first real ones made by any video DSLR since the Canon 5D Mark II andNikon D90) introduced the feature in 2008, will eventually (I hope?) trickle down to cameras that mere mortals can afford.
So what’s new?
In a word – the headphone jack.
Why is a trivial little jack the feature I chose to start with? Because finally, someone put one of these in a DSLR.
Recording interviews (or anything, really) has been a mix of both fear and frustration for DSLR video shooters over the last few years. While the large sensors and manual controls have enabled some amazing new looks for news video, the audio components haven’t kept pace. With the 5D Mark II, a camera I used since the day it was released, it wasn’t until a firmware update, in March 2010 (a year and four months after its release) that you could change audio levels. I plugged my external microphones into the jack and hoped for the best each time – without being able to control your audio levels or monitor through headphones, you were (aurally speaking) flying blind. The D90 was even more constrained – it had no jack for external microphones, leaving the built-in one as your only option.
Since then, most cameras have taken care of the audio levels part. Current mid- to pro-range cameras give you full manual controls in both audio and exposure, but up until the D4, there still wasn’t a way to check and make sure of what you were getting audio-wise while you were shooting. If a microphone malfunctioned mid-shoot, you wouldn’t know until you played back the clip – usually long after your subject had gone home and a reshoot became impossible.
This led to many visual journalists devising custom DSLR rigs to hook up audio recorders as backups to the sound captured with the camera. This provided an extra layer of protection in case microphones failed, gave the journalist control over audio levels and allowed headphone monitoring – but at the cost of having to sync up the audio and video later in software. Not an ideal situation in a hurry. It also added bulk and expense to the whole endeavor. One of the upsides to DSLR video was that it was supposed to reduce the amount of gear to carry in the field – I remember days where I carried a bulky Panasonic DVX and a Canon DSLR to shoot with at the same time, and that was only 2006.
Including the audio solutions in the camera – audio levels and headphone monitoring – should cause these dreams of a lightweight field setup to become a reality.
The rest of the specs
There’s some other noteworthy inclusions that are a step up above the previous generation of video Nikons – audio meters on the screen (more help with audio monitoring is always a plus), more frame rates (1080p at 29.97 fps is something Canon has been doing for a while, so it was long overdue here and a nice addition to the 23.97 fps mode that was already slick on the D7000), and the ability to capture raw HD feed straight from the camera (now you can record those hour-long press conferences in one shot). There’s also another feature borrowed from Canon – a “crop mode” for video that uses a smaller area of the sensor to magnify your lenses by 2.7x. I’ve tried this on a Canon Rebel T2i and it is quite handy to extend the reach of even modest lenses. That said, such extreme magnifications are taxing on the optics of lenses – any minor imperfections of a lens are brought to light at such high magnification.
Where do we go from here?
While this kind of progress is good, there’s still more work to be done. I used to shoot primarily video in the field, and so here’s some more for the next generation of cameras, in no particular order:
1. One more audio input. I can live without professional XLR connections – most miniplug microphones (of various types – wireless, lavalier, stick, shotgun, etc.) I’ve used have all worked fine, but shooting a two person interview with one mic is difficult. This has been a longstanding feature of professional video cameras that is needed in this space as well.
2. Built-in neutral density filters. One of the great things about DSLR video is the shallow depth of field you can get with wide apertures (think f/2.8, f/1.4, etc.). That said, without screwing neutral density filters (think sunglasses for cameras) onto the front of lenses manually, there is no way to shoot that wide in broad daylight. Most video cameras have built-in filters you can control with a switch. How about one for DSLRs?
3. Smooth zooms. Video cameras have powered zoom buttons that enable smooth zooms. On a DSLR, twisting the rings on a lens introduces a lot of unsteadiness in the zoom, and it never looks as good as a real video camera. Panasonic has made some great strides here with their “X” series of lenses for the micro four-thirds system. I don’t own a Panasonic’s range-topping GH2 (I have a GF3), but the video quality on micro-four-thirds is quite good, and this could be Panasonic’s “ace-in-the-hole” to cracking the DSLR video market that Canon and Nikon dominate right now.
4. Swivel Screens. If Canon can do this on a 60D and Rebel T3i, and Nikon can do this on their D5100, surely this technology can make its way upstream, right? Who says that everything has to float downward – this is a great innovation for video shooters, let’s see it in more models.
Maybe some of this stuff can come around in the (at this point, fictional) Nikon D7100?
Samples
And of course, don’t just take my word for all of this – see some D4 sample video and a tech demo:
DSLRNewsshooter’s run-through of the camera’s features:

The real upshot of the 2.7 crop is that it doesn’t do any line skipping or resampling at all. It’s straight 1:1, pixel for pixel. This should cut way back on aliasing and moire that have been a huge weak point for high-res SLRs shooting video.
I’ll have to see it to believe it. Even my L lenses showed lots of purple fringing on the T2i when I used the crop mode. I love my Nikon 85 f/1.4, but even without magnifying anything there’s a lot of chromatic aberration. It would probably look worse at 2.7x. Still, to have that extra reach as an option is worth any compromises.
Enlighten me about line-skipping? I’ve read a little about it and understand how it causes moirĂ© patterns sometimes, but how critical is it really when you’re talking about video for Web use? It never really came up much in the shooting I did with the 5D Mark II or the D7000 now.
Sorry, just saw your reply.
Line skipping is exactly what it sounds like. The camera discards the data from every other line. It often results in more jagged curves and lines–aliasing–because you’re sampling the information from photosites that are farther apart. Take a camera like the C300, it has a 4K (I think) sensor, but only outputs 1080. It doesn’t skip lines the way the 5D does though, it resamples the image, which is way cleaner, and actually has some advantages in noise handling.
As far as how it affects web video, it depends. If you’re doing a lot of compression, it’ll often eliminate aliasing and CAs, especially if you’re resizing (resampling). If you’re uploading full-size to somewhere like Vimeo, the artifacts are more likely to make it into the final mix.
Does the 85 give you CAs in stills? I thought Nikon mostly eliminated that in the last generation image processor.