Is shallow depth of field getting used to death?

What was once the Holy Grail, the shallow depth-of-field look that distinguished the “35 film look” from video, getting so overused that it will become passe?

Here’s a rant this week on APhotoEditor, which in turn reposts a piece from 2009 by Mike Jones:

Shallow focus and Rack-Focus is lazy. A ham-fisted and overtly slothful technique with little impetus other than to lead your viewer around by the nose, to force them to look exactly where you want them to look, when you want them to look there. As a tool, like all other cinematic tools at the filmmaker’s disposal, it can and may be very useful. But as a staple and default way to depict moving images it is as articulate as a house brick.

There are some great new camcorders coming out that are not playing to the shalllow-depth crowd. Those include the Canon XF100 and 105, which put out 1080p in a quality 4:2:2 50mbps stream, all with 1/3-inch sensors. There are some filmmakers who are already scoffing at these because they don’t have big sensors.

One obvious place a documentary filmmaker would want deep depth of field provided by such camcorders is in fact-moving action in which an EOS 5D would find focusing problems unless in the hands of a seasoned pro. But even in sit-down shooting setups, cameras with more depth allow for a better look at surroundings, rather than making them blur out. There are times when that can be good.

HD video is here to stay and fully accepted by film audiences, especially in documentary work, and there may be a counter-wave of people shooting to go against the trendy shallow-depth look. The reason: Because what was once a look associated with super-expensive 35mm film cameras is now a look assoiciated with relatively cheap DSLRs. When you see that indy film at the local festival with nothing but blurred background, you’re likely to say, “Oh, that was shot on a 5D” than to say “this must be a highly funded professional production.”

It can be funny how technology meant to do one thing can do another. We’ll see what happens next.

Jones, in his post from last year, notes that the technical challenge of filmmaking back in the day was actually “deep focus,” a look that only came at great expense – to shoot at small apertures with the slow film of that era, lighting had to come to a set by the truckload.

In film, the quest for deep focus led to sets blitzed with light; once everyone was getting that look, others came along and played against the grain: ‘The Godfather” used fewer lights, created shallow depth in many scenes, and was then quickly mimicked in the films of the 1970s. Around it goes. But when video cameras came out, the flat deep-focus look was associated with those who could only afford cheap equipment.

Jones said in his post,

In the 21st century I would attest that Shallow Focus and Rack Focus aesthetics have lost all meaning as useful creative problem solving techniques and instead have become banal, unimaginative staples of cinema. And it prompts us to ask loudly…. “What the hell happened to Deep Focus?”

DSLRs meet BBC broadcast standards (but only kind of), and the upshot of that

This is kind of like the joke where the man says to the doctor, “Doc, it hurts when I do this, what would you suggest?” and the doctor says, “Don’t do that.”

The BBC has exacting standards about the quality of footage it will broadcast, for good reason, but their technical demands also has created a proper British club of filmmakers who can afford expensive equipment. The hoi polloi with these “prosumer” cameras just haven’t cut it. The BBC has required 50mbps quality; HDV and H264, compressed down as they are to about 24Mbps, are on the outside looking in; even the great Sony EX1 at 35Mbps doesn’t technically cut it.

So now comes the news that the BBC is accepting EOS 5D footage, which provides a 38mbps bit rate, but also has well-documented flaws in regard to moire and aliasing.

The HD Magazine article quotes the producers of the drama “Road to Coronation Street” as saying,

“We shot a lot of tests which looked wonderful but the engineers at the BBC were saying that there was aliasing and moiré patterns that would make them fail the use of the camera for HD broadcast.

“However we’d done a lot of independent research and shot more tests and shown them to people. When the tests came back from the BBC they said there was aliasing but they suggested that we shoot more tests in a type of scenario that the production would encounter, like a scene with actors looking backwards and forwards with mid-shots and close ups.

“The head of technology at the BBC, Ian Potts, who was very supportive in our early days saying ‘You have these technical issues that would fail our broadcast tests but it’s very interesting what you’re doing and please do some more tests because we’d love to see what you do’. After seeing the four day’s worth of rushes he was so completely smitten with it and said it was some of the best material they’d ever had, ‘finish the film on it’. It went from nought to 60 in half a second.”

So, basically, it’s a “Don’t do this” scenario. That means avoiding such moire-inducing patterns as checked shirts and window screens, not setting long focus on straight-lined patterns, and keeping everything fairly tightly focused. (Other discussions such as this are common among DSLR users) There are many other technical considerations beyond that, but it’s the simple idea of being aware of the limitations of your equipment.We’ll see whether there’s a more general acceptance of DSLR footage out of this.

Testing the 5D’s dynamic range, and assessing the 60D

For technical geeks, the debate rages about the video quality of DSLRs. What seemed gained in sharpness and low-light performance seemed lost in such negative effects as shutter Jell-O, moire and aliasing.

But ProVideo Coalition has done a more definitive test on the Canon EOS 5D Mk II’s dynamic range, the ability to find detail over a number of f-stops.

The testing shows the $2,500 EOS 5D to cover 11 stops of dynamic range. In comparison, the $6,000 Sony EX1 has 10 stops, the $30,000 RED One is said to have 11.3 stops, the $150,000 Sony F23 has 13, and the $70,000 Arri Alexa about 13. Kodak claims 15 stops for its color negative film stock.

These studies only further what we know. That these DSLRs, despite their limitations, have been game-changers that have allowed more people to even think about doing film. Ten years ago, video cameras such as the Canon XL1 had only about 7 stops of dynamic range, which meant you had to light carefully and evenly. You would never have put a subject in front of a black background (such as in the clip at the bottom of this post, from Nathaniel Hansen’s “The Elders”) because the camera would have converted those blacks into noisy artifacts and blown out the whites on the other end. But the gap between video and film has grown so much smaller, newer filmmakers don’t even have to worry about such considerations (and may laugh at older films with very flat lighting).

Meanwhile, the EOS 60D has hit the market at $1,100, filling the gap between Canon’s $1,700 EOS 7D and the $900 T2i.

Philip Bloom posts about the 60D, and the Wired’s Gadget Lab calls the 60D “Frankencam,” cobbled together out of lower- and higher-end stuff:

It’s clearly the season for new camera gear, and today it’s Canon’s turn in the spotlight. Along with a few new lenses comes the EOS 60D, a “replacement” for the two-year-old 50D. Those looking to upgrade from their 50D should look elsewhere, though, perhaps to the 7D, as this new camera is more for consumers than enthusiastic amateurs.

The magnesium body of the 50D is now plastic, and the 60D uses SD-cards instead of Compact Flash. It also gets a slew of gimmicky image processing features (Toy Camera, anyone?) and the obligatory video capabilities.

One 60D feature that is getting praise is the articulating screen.

Coach from Nathaniel Hansen on Vimeo.

Nikon’s D3100 brings disappointment for serious video users

The Nikon D3100 is out, with 1080p, but the shortcomings of the camera are bringing about disappointment for Nikon fans hoping for their company to catch up with Canon.

Nikon has yet to introduce a 1080p camera with a full-frame sensor. The D3100 apparently has no audio input jack. The codec is questionable. And while the D3100 has a low price point, the sensor is 14 megapixels.

DPReview says the D3100’s screen is a “slight disappointment,” with lower resolution than hoped for.

EOSHD notes that the camera does not seem to be able to go fully manual for shooting video.

After the disappointment of Sony’s new DSLRs, the A55 and A580, this is not really what we need. According to HDSLR shooter Nino Leitner, Nikon’s German press office have said that the D3100 will not have the option for manual controls in video mode.

The general stance of Japanese DSLR manufacturers seems to be that customer research has told them their cameras need to be easy to use. At least the entry to mid-range consumer DSLRs.

Slashcam notes,

We have held consultations with Nikon’s press agency and wanted to know if the Nikon D3100 has fully manual video features: the unfortunately does not seem to be the case.

Canon develops a 120mp CMOS sensor

TOKYO, August 24, 2010—Canon Inc. announced today that it has successfully developed an APS-H-size^*1 CMOS image sensor that delivers an image resolution of approximately 120 megapixels (13,280 x 9,184 pixels), the world’s highest level^*2 of resolution for its size.

Compared with Canon’s highest-resolution commercial CMOS sensor of the same size, comprising approximately 16.1 million pixels, the newly developed sensor features a pixel count that, at approximately 120 million pixels, is nearly 7.5 times larger and offers a 2.4-fold improvement in resolution.^*3

With CMOS sensors, while high-speed readout for high pixel counts is achieved through parallel processing, an increase in parallel-processing signal counts can result in such problems as signal delays and minor deviations in timing. By modifying the method employed to control the readout circuit timing, Canon successfully achieved the high-speed readout of sensor signals. As a result, the new CMOS sensor makes possible a maximum output speed of approximately 9.5 frames per second, supporting the continuous shooting of ultra-high-resolution images.

Canon’s newly developed CMOS sensor also incorporates a Full HD (1,920 x 1,080 pixels) video output capability. The sensor can output Full HD video from any approximately one-sixtieth-sized section of its total surface area.

Images captured with Canon’s newly developed approximately 120-megapixel CMOS image sensor, even when cropped or digitally magnified, maintain higher levels of definition and clarity than ever before. Additionally, the sensor enables image confirmation across a wide image area, with Full HD video viewing of a select portion of the overall frame.

Through the further development of CMOS image sensors, Canon will break new ground in the world of image expression, targeting new still images that largely surpass those made possible with film, and video movies that capitalize on the unique merits of SLR cameras, namely their high mobility and the expressive power offered through interchangeable lenses.

*1	The imaging area of the newly developed sensor measures approx. 29.2 x 20.2 mm.
*2	As of August 20, 2010. Based on a Canon study.
*3	Canon’s highest-resolution commercial CMOS sensor, employed in the company’s EOS-1Ds Mark III and EOS 5D Mark II digital SLR cameras, is equivalent to the full-frame size of the 35 mm film format and incorporates approximately 21.1 million pixels. In 2007, the company successfully developed an APS-H-size sensor with approximately 50 million pixels.

Study: Video quality matters less than story quality (Thank you!)

Back in the days when Hi-8 was about the best video quality an average person could get, we longed for a day when our work wouldn’t look so amateurish that we wouldn’t be taken seriously. When we got the $6,000 Canon XL1s in our hands, it was still clearly video, but at least it was starting to look on a level with those $100,000 camcorders the TV station used. Some of us still shot on 16mm film, hoping we’d get something to justify the cost.

In the last decade we’ve gone from 480i to 1080p and beyond. The ability to “look like film” is nearly a given, and $100 Flip cameras shoot at a quality surpassing the XL1s. Because it’s been easier to make big leaps in video quality than in the actual quality of our work, it can be easy to focus on that.

But a rather reassuring study comes from the Rice University Department of Psychology. Published in the journal Human Factors by Rice professor Philip Kortum, the study in essence says that if the content is something the subject wants, the perception of video quality is far less important.

Physorg.com reports that,

Using four studies, Kortum, along with co-author Marc Sullivan of AT&TLabs, showed 100 study participants 180 movie clips encoded at nine different levels, from 550 kilobits per second up to DVD quality. Participants viewed the two-minute clips and then were asked about the video quality of the clips and desirability of the movie content.

Kortum found a strong correlation between the desirability of movie content and subjective ratings of video quality.

“At first we were really surprised by the data,” Kortum said. “We were seeing that low- quality movies were being rated higher in quality than some of the high-quality videos. But after we started analyzing the data, we determined what was driving this was the actual desirability of the content.

We recently posted here on Stefan Sargent’s “Four Times Better Rule,” and it seems all the more sensible given the study. Instead of scrimping for that 1080p camera to replace your 720p, put the money into finding really interesting subjects, because the audience is less likely to notice. Back in December, when we interviewed filmmaker Robert Greene, he noted that he’d shot his film on standard definition, and that it ended up not only premiering at Full Frame, but also getting onto iTunes.

The Kortum study says this:

“If you’re at home watching and enjoying a movie, we found that you’re probably not going to notice or even concern yourself with how many pixels the video is or if the data is being compressed,” Kortum said. “This strong relationship holds across a wide range of encoding levels and movie content when that content is viewed under longer and more naturalistic viewing conditions.”

Redrock’s wireless follow-focus system

The end of DSLR filmmaking? Not so fast…

ProVideo Coalition’s Clint Milby recently assessed the new Sony NEX-VG10 camcorder’s claim as “the new sheriff in town” for lower-cost, high-quality video and came away with some reservations. While the camcorder brings to a more traditional package large APS-C sensor size and interchangeable lenses, it’s also not quite there.

Milby notes the Sony’s main drawbacks as 30fps only, video saved to an interlaced container, no XLR audio input, and no adapter system to fit lenses other than Sony’s. But, he says,

The NEX-VG10 makes a giant leap forward to bringing features to the consumer that were only available to professionals, but the camera still lacks some of the most meaty parts of the HDSLR. However, as Panasonic prepares to release its interchangeable lens camcorder in the Fall of 2010, we should consider this device a sign of things to come.

Cinematographer vs. Producer

Just one for fun that takes on the DSLR craze. Warning: foul language ahead.

Colorista II with Stu

01 – Getting Started with Magic Bullet Colorista II from Red Giant Software on Vimeo.

Next Page »