With the FS7, Sony finally learns to cannibalize itself

At the International Broadcasting Conference in Amsterdam, broadcasting equipment companies announce new products, and often ship the new products they announced at NAB in April. For example, Panasonic showed near-production versions of its Varicam 35 that was announced at NAB, as did AJA with its CION camera. Sony, on the other hand, showed a new camera whose existence started to be rumored only a few weeks before IBC. The PXW-FS7 (referred to by most people as the FS7) is a Super 35 4K camera that fits in price between Sony's FS700 and F5, but is functionally superior to the F5 in many ways. It uses the full XAVC codec and records 10-bit 4:2:2 UHD 4K at up to 60 fps and 600Mbps (Digital Cinema 4K will be supported in a firmware upgrade scheduled for early 2015,) but it uses Sony's XQD flash media, which costs substantially less than the SxS Pro+ flash media used by the F5.

It's got built-in ND filters, and it natively accepts Sony's E-mount lenses; Sony announced a new professional power zoom 28-135mm F4 lens to go along with the FS7. An A-mount adapter is available, and of course, third-party adapters that connect a variety of mounts to A- or E-mounts will also work. It's got a standard grip control that puts many of the camera's most important controls on a hand grip. An optional extension unit enables the FS7 to record using Apple's ProRes 422 codec, outputs raw 12-bit 4K video that can be recorded by external Sony and Convergent Design recorders, and supports industry-standard batteries. The FS7 will be somewhat heavier that AJA's CION; the FS7 weighs 4.5kg without the extension unit that it needs to be functionally comparable to the CION, while the CION weighs 3.4kg. Both cameras are lightweights compared with Blackmagic Design's URSA, which weighs 7.4kg.

What makes the FS7 so worthy of discussion is that a number of observers have noted that it's in many ways a better camera than Sony's F5, for less money. The F5 sells for $16,490 (U.S.) at B&H, and that's without a viewfinder or lens. The FS7 will sell for $7,999 at Adorama ($10,499 with 28-135mm lens.) Introducing a new product that competes directly with another Sony product for less money was, until now, considered heresy. Sony took extraordinary pains to make sure that its products didn't directly compete with each other, except when the company was deliberately obsoleting an older product. In this case, however, Sony says that the FS7 will replace neither the FS700, which B&H sells for $7.699 and which the FS7 blows out of the water, nor the F5, which the FS7 compares very well to for about half the price.

Sony's no-competition policy dates back to when Sony was the undisputed technological and market leader in cameras. Any cannibalization of Sony's own products was seen as unnecessarily leaving money on the table. However, first Panasonic and then Canon showed that they could build cameras that could compete very well with Sony's offerings. Panasonic in particular was largely unconcerned if its cameras cannibalized its other models, and both Panasonic and Canon were happy to take sales away from Sony. Blackmagic Design showed that it can't yet design or build cameras to Canon's, Panasonic's or Sony's standards, but it introduced price competition into a business that hadn't seen much of it. That brought in AJA, which looks like it's learned from Blackmagic's mistakes and will combine high-end performance with aggressive pricing.

Sony's in a new world. It's now got competitors that are its technological equal and are willing to accept a lower gross margin on their sales. Sony has finally figured out that it's better to cannibalize yourself and keep the revenues, rather than let your competitors cannibalize you and take the revenues. Sony's going to let its customers tell it if the FS7 replaces either the FS700 or F5. The older products will stay in Sony's product lines until sales fall off sufficiently to make one or both unprofitable to continue to offer.

With the FS7, Sony is finally doing what many observers and customers hoped that it would do decades ago, Time will tell if the FS7 is a one-time fluke or the first product in a new strategic commitment.

For 4K cameras, price is A thing, but not THE thing

I'm back from NAB, where the overriding theme this year was 4K everything--cameras, monitors, editors, special effects, routers, switchers, etc. Perhaps the biggest battle was in cameras, where AJA Video entered the market for the first time and Blackmagic Design and Sony announced new 4K cameras. (JVC also announced its first 4K digital cinematography cameras, but gave no prices or availability dates.) 4K cameras have been a "thing" ever since the RED One, but $10,000 was the least that you could spend to buy one (Canon's 1D C) until Blackmagic shipped its 4K Production Camera late last year, priced at $2,995.

The floodgates have now opened:

• Panasonic's GH4: $1,699, or $3,299 bundled with its SDI/XLR interface dock
• Sony's A7S: $2,499.99, will ship in July
• Blackmagic's 4K Production Camera: $2,995
• Blackmagic's URSA EF: $5,995
• AJA's Cion: $8,995

None of these cameras cost more than a fraction of the price of an ARRI Alexa ($80,000+) or Amira ($40,000-$52,000 depending on enabled features,) RED EPIC-M Dragon ($50,000+), or Sony F55 ($29,000+) or F65 ($65,000+). You'd think that ARRI, RED and Sony would be shaking in their boots, but they're not. There are two reasons why the companies that make high-end cameras aren't necessarily threatened by the new inexpensive models:

1. There are many elements that determine whether or not a specific camera is appropriate for an application, and
2. You get what you pay for.

Here are some (but far from all) of the elements of camera design that influence how the camera performs and what it's good (or not good) for:

• Imager size
• Imager resolution
• Color space (e.g., YUV or xvYCC)
• Color sampling (e.g., 4:2:0, 4:2:2 or 4:4:4)
• Bit depth (8-bit vs. 10-bit)
• Video output resolution (DCI 4K (4096 x 2160), UHD (3840 x 2160), 1080, 720)
• Video compression formats (e.g., AVCHD, H.264, ProRes, DNxHD, XAVC, XAVC S, AVC-Intra, AVC-Ultra)
• RAW storage and/or output
• Frame rates supported at specified resolution (e.g., 24, 25, 29.97, 30, 59.94, 60 or 120)
• Dynamic range in stops
• Native ISO speed
• Sensitivity (maximum usable ISO speed, NOT the camera's rated top ISO speed)
• Lens mount(s)
• Lens control (manual, automatic or limited automatic)
• Image stabilization (digital or optical, in the camera body or in the lens, or no stabilization)
• Viewfinder, display screen, both or none
• User interface design (e.g., touchscreen(s), menus, dedicated buttons and knobs)
• Storage capacity
• Storage media
• Internal or external storage
• Video interface(s) (HDMI or HD-SDI, with different HDMI versions and, for HD-SDI, maximum speeds)
• Audio connector(s)
• Balanced or unbalanced audio in
• Phantom power availability
• Run time on battery
• Removable or permanent battery
• External power voltage and connector
• Camera shape
• Camera weight
• Ruggedness
• Manufacturer and design maturity (how much experience does the manufacturer have in designing cameras, and how long has the manufacturer been making this particular camera)

There's an enormous number of elements to consider, and some elements work much better for certain applications than others. In some cases, buyers have a wealth of cameras to choose from, while in other cases, there may only be a handful that can do what they need.

Rather than salivating when you hear about a new low-priced camera with attractive features, ask yourself these questions:

1. What am I going to use it for?
2. What trade-offs am I willing to accept (for example, are you willing to live with less sensitivity in order to get a higher-quality compression format?)
3. How often will I use the camera (do you know that you'll be using it over and over on new projects, or do you have one project in mind and you don't know when you'll have the next one?)
4. How much can I afford to pay?

Answering the first two questions will allow you to compile a list of cameras that meet your needs. Answering the final two questions will tell you whether you should buy or rent the camera that you can afford. In some cases, you may decide to buy a less-expensive camera and use your remaining budget to buy lenses or mounting equipment. In other cases, you could rent a camera and use the savings elsewhere on your production, or rent a camera that you can't afford to buy that's superior to other choices for your application. In short, you should answer the four questions first, rather than starting with the price of the camera and instead hoping that it will meet your needs.

That was fast: Blackmagic Design drops price of Production Camera 4K to $2,995

Just three days after the formal announcement of Panasonic's 4K GH4, and before Panasonic has announced pricing or availability for its new camera, Blackmagic Design has cut $1,000 off the price of its Production Camera 4K, bringing the price down to $2,995 (U.S.). That should be below the price of the GH4 plus its dock. In addition, as I wrote yesterday, the GH4 will need an external video recorder to take full advantage of the camera's 4K output, while the Production Camera 4K has a built-in SSD dock, which makes Blackmagic's camera an even better deal. The Production Camera 4K also has a Canon EF/Zeiss ZE mount, compared to the GH4's Micro Four Thirds mount, which could be very important if you've already got an investment in Canon lenses.

I suspect that Blackmagic was planning the Production Camera 4K's price drop as a big announcement at NAB in April, but the company decided to make the announcement early after seeing Panasonic's moves. It's pretty amazing--in the space of four days, we've gone from no 4K cameras under $3,000 to two.

Panasonic's GH4 lowers the bar for 4K pricing, but are the compromises worth it?

Panasonic’s flagship GH series of Micro Four-Thirds ILCs has developed a strong reputation as cinematography cameras, starting with the GH2, which had its firmware hacked to enable much higher bit rates than the stock model. The GH3, introduced last year, took many of the capabilities added by third parties and built them into the base camera. Last month, at the Consumer Electronics Show, Panasonic showed a prototype of a new GH that looked almost identical to the GH3 but supported 4K video. As of this weekend, we’ve learned the specifications for the new camera, called GH4, but not the price or release date.

The GH4 16.05 megapixel CMOS imager doesn’t break any records for still imaging, but it does support 4K video at both standard resolutions: Ultra HD (3840x2160 @ 30p,) the broadcast/video 4K standard and the resolution of consumer Ultra HD displays, and Cinema 4K (4096x2160 @ 24p,) the baseline standard for theatrical production, post-production and exhibition. In 4K mode, it uses IBP compression with I and B frames at 100mbps, while in 2K mode, it supports All-I compression at 200mbps.

The GH4 has an interesting (and confusing) approach to how it handles storage of 4K video: When using an SD card, video is stored in 8-bit 4:2:0, and is output to the HDMI terminal as 8-bit 4:2:2. If you remove the SD card and use an external recorder, the HDMI output is 10-bit 4:2:2. You can also opt for a dock (Panasonic refers to it as an “interface unit”) that provides two XLR audio inputs with LED meters, four SDI outputs (two of which are 3G) and a 12VDC power socket. The SDI outputs can presumably drive an external recorder and monitor simultaneously at 10-bit 4:2:2. The dock won’t win any design awards—it just about doubles the size of the GH4—but it does add the interfaces that professional users need (or are forced to add with third-party hardware.)

Panasonic’s approach to storing 4K video means that an external recorder will be a necessity. In this regard, Blackmagic Design’s Production Camera 4K (the camera most likely to be compared to the GH4) has an advantage, because users can insert a SSD directly into the Blackmagic camera, eliminating the need for an external recorder.

I have to admit that I’m disappointed with how Panasonic chose to implement storage on the GH4. It would have been nice to be able to use the camera in a handheld mode without a lot of additional hardware, but in 4K mode, the SD card is only good for proxy recording in 4K mode. (To be clear, you can store 4K video on the SD card, but if you're serious enough about 4K to put up with all the other issues you'll need to deal with in post-production--massive storage, faster PCs, bigger monitors, etc.--8-bit 4:2:0 won't cut it.)

At CES, Panasonic representatives said that they expect the GH4 to be priced less than $2,000 in the U.S. That price won’t include the dock, which is likely to cost at least $1,000. So, Panasonic could get the GH4 plus dock to market at around $3,000—but I wouldn’t be surprised if the pair launches at closer to $4,000.

Update (March 10, 2014): Panasonic has revealed prices and availability dates for the GH4. The GH4 body's suggested retail price is $1,699.99 (U.S.). The dock (officially called the DMW-YAGH XLR/SDI Interface Unit) is priced at $1,999.99. If you buy both the GH4 and dock as a bundle, it's priced at $3,299.99. I’d love to see a third party develop a more elegant (and cheaper) dock, but there may not be a big enough market for it to make financial sense.

The GH4 is one of the least expensive ways of getting into 4K, if not the least expensive way. However, as we’ve learned from Blackmagic’s cameras, inexpensive means compromises, and the GH4 is no exception. It remains to be seen if the GH4’s compromises are ones that you can live with.

NAB 2013: The Wrap-up

I only had a day to spend at NAB last week, so I couldn't get to every booth, and undoubtedly missed some "gems" hidden around the show floor. However, I did get to see much of the South and Central Halls. Here's a summary of the products that impressed me (I've already written about Blackmagic Design's two new Cinema Cameras,) along with what wasn't there, and some thoughts about the future of the business:

The New

• Canon's XA20 and XA25: Canon's new small, light and low-cost ENG camcorders are the company's first models with 1080/60p capability. Both models have 20x zoom lenses, dual-band WiFi and dual SDHC/SDXC-compatible memory card slots. The XA25 adds dual XLR audio inputs and an HD-SDI output. The list price of the XA20 is $2,499 (US,) while the XA25 is priced at $2,999; street prices are $2,199 for the XA20 and $2,699 for the XA25. Both camcorders are expected to ship in late June.
• JVC's GY-HM650U ENG camcorder (street price $5,695) was launched at last year's NAB, and it's recently scored a number of high-profile, big-quantity sales to customers including the BBC. The 2.0 model introduced at this year's NAB (a firmware upgrade for camcorders already in use) adds a number of new features. The HM650U has three 1/3-inch CMOS sensors and a 23x zoom lens. It can simultaneously record to dual SDHC/SDXC-compatible memory cards, output video through its HD-SDI or HDMI connectors, and stream a webcast-appropriate version of the video over its built-in WiFi interface or a 4G LTE adapter.
• Perhaps the biggest hit of the show was Freefly System's Movi M10 camera stabilizer. Unlike stabilizers built around the Steadicam model, which uses a system of springs and joints (and requires a vest on larger models to handle the combined weight of the stabilizer and camera,) the Movi is an active hand-held design using direct-drive motors and accelerometers to keep the camera stable. The Movi weighs 3.5 pounds and is built using carbon fiber in order to keep its weight down. It can be operated in two modes: In "Monarch" mode, the cinematographer uses his or her movements to control the Movi, while in dual operator mode, one person holds and moves the Movi while another person wirelessly controls the camera's position using a tablet and RC control.
  
  Before NAB, a number of observers said that the Movi would be too heavy for long use. The maximum weight of camera, lens and accessories that the Movi can handle is 10 pounds, making the total maximum system weight 13.5 pounds or less. I saw men and women of various sizes handling the rig without problems. The Movi M10 model is priced at $15,000 and is expected to ship in Q3; the company plans to add a M5 model priced at $7,500 that can handle a maximum camera weight of 5 pounds. $15,000 is out of the range of most independent filmmakers, but the Movi will undoubtedly be available for rent.
• The low-cost UAV business got a big boost from the DJI Phantom, a fully-assembled quadricopter that includes a RC control, GPS navigation and camera mount for a GoPro camera, for under $700. The Phantom's maximum flight time is 10 to 15 minutes, and it has a maximum flight control range of 300 meters. DJI showed a prototype of a new Phantom model with a built-in video camera that can be remotely tilted. Neither the price nor the availability of the new model were announced at the show.
  
  The Phantom is about as foolproof as a radio-controlled quadricopter can get:
• It has a built-in autopilot that enables navigation to a specific latitude and longitude.
• The manual controls can be set to allow steering to be correct relative to the operator's position, no matter what position the Phantom is in.
• It can return to the operator automatically.
• If it flies beyond the range of the RC controller, the Phantom goes into hover mode, and if a good GPS signal is available, it will automatically return home.
• Matrox's new $995 Monarch HD live video encoder accepts video input from HDMI and outputs H.264 video at up to 20 Mbps in both RTMP and RTSP protocols, which means that it supports virtually any streaming server or service. It can simultaneously save the video in MP4 format at up to 30 Mbps on a removable SD card, USB hard disk or flash memory, or on network-attached storage. It has a simple web-based user interface, and can control up to three additional slave encoders for feeding to multiple streaming servers, services or CDNs. 

The Missing

• One thing that surprised me was the lack of new products from some of the leading broadcast equipment companies, especially Panasonic. For many years, Panasonic could be counted on to introduce new and exciting cameras, but this year, there was nothing really new. For example:
• The AG-AF100A, which pioneered the big-sensor low-cost cinema camera market, has only been lightly upgraded since it was announced in December 2010. Panasonic hasn't introduced any new cameras into this market (excluding the GH3, the follow-on to the company's "accidentally successful" GH2 digital camera that's gotten a wide following from budget-sensitive cinematographers.)
• Last year's "camera under glass," a professional 4K camcorder with an Android interface, disappeared this year and was replaced with a generic, consumer-looking 4K camcorder mockup that was first shown at CES in January.
• Sony, JVC and Canon didn't announce many new products. JVC's biggest news was a firmware upgrade, and Canon didn't announce anything new on the Cinema Camera front. It's possible that the companies are "catching their breath" after the last 18 months' explosion of new product introductions, but it's still disappointing to come to NAB and not see much new from the market leaders.

The Trends

• Video hardware and software pricing is looking more and more like computer pricing, where prices go down and capabilities go up each year. Here's a few examples:
• Adobe's Creative Cloud offers users everything in Creative Suite 6 for about $50 per month per user, and they can use the software on two PCs. That, combined with improvements in Adobe's software, is enabling Adobe to pick up lots of market share in video editing and post-production. In response, Avid has priced its new Media Composer 7 at $999, with the additional Symphony features priced at $1,499. $999 used to be the price of a competitive upgrade from Final Cut Pro or Premiere Pro to Media Composer, and only for limited periods; now, it's the list price of the software.
• Blackmagic Design has driven down prices in every market that it's entered, and competitors have had no choice but to respond. Prices for professional color correction systems have tumbled since Blackmagic acquired Da Vinci Systems, as have prices for video production switchers (except the very top-of-the-line models) since it acquired Echolab. The market for high-end video processing systems has always been small because of their high cost, but Blackmagic's acquisition of Teranex and subsequent rock-bottom pricing will dramatically increase the size of the market. The cinema camera market is already highly competitive, but Blackmagic is increasing options and decreasing prices for buyers.
• Canon, JVC, Panasonic and Sony are using their top-of-the-line consumer camcorders as the basis of their entry-level prosumer/professional camcorder lines, which increases production volumes, decreases costs and allows manufacturers to lower prices. In most cases, if you don't need XLR inputs or HD-SDI outputs, you can save a fair amount of money by buying the consumer models. However, even the prosumer/professional models are less expensive and more capable than comparable models from even a couple of years ago.
• DSLRs have dramatically decreased the cost of cinema cameras, and an entire ecosystem of lenses, rigs and accessories that are fairly priced in relation to DSLRs has emerged.

After more leaks than a sieve, the Panasonic GH3 is revealed

Perhaps it was the multiple leaks of specifications to photo websites, or Panasonic itself posting a promotional video on YouTube and then taking it down, or yesterday, Samy's Photo posting specifications and pictures, but today's announcement of Panasonic's new GH3 seems like an anti-climax. It shouldn't, since Digital Photography Review writes that only the Canon 5D Mark III has a higher 2K video bitrate than the the GH3, and the Canon DSLR is priced more than $2,000 (U.S.) higher. The GH2 was a firmware hacker's dream, with the video bitrate taken all the way to 176Mbps in AVCHD with All I-frames, while the maximum video ISO was increased from 3200 to 12,800. The problem was that the some GH2s became unstable when run at this insane bitrate (most users chose to use a more reasonable 44Mbps, which is still much faster than the maximum 28Mbps of AVCHD 2.0 at 1080p60.)

Panasonic has taken the hackers' improvements to heart, and has implemented a maximum bitrate of 50Mbps in 1080p60, or 72 to 80Mbps in All-I Frame at 1080/24p or 30p, both using H.264 compression. (All the frame and bit rates of AVCHD 2.0 are also supported.) Its maximum ISO, in both still and video mode, is now 12,800. This gives the GH3 virtually the same performance as the GH2 with hacked firmware, without requiring hacking or voiding the camera's warranty. In order to provide better performance while maintaining the camera's reliability and stability, the GH3 has a new three-core Venus 7 CPU.

The GH3 also supports timecode in H.264 and AVCHD modes, and it has a headphone jack for audio monitoring, in addition to a microphone jack and manual control over audio levels. The HDMI out can be configured with overlays on or off, so it can be used for monitoring and with an external recorder. (It's not clear whether the GH2's HDMI quirks, which made it unusable in many cases with external recorders, have been fixed in the GH3.)

The GH3 is no slouch as a still camera, either:

• 16 Megapixel sensor
• 1.7 million dot OLED viewfinder
• 614K dot 3" OLED touchscreen display
• Autofocus speed of .07 seconds
• 6 fps maximum continuous frame rate
• Memory card slot for SD, SDHC and SDXC cards
• A fully sealed magnesium alloy frame
• Built-in Wi-Fi

The U.S. price of the GH3, $1,300 for body only, is comparable to the price of the GH2 when it was first launched, but the GH3 is much more camera. The GH2 became the budget "go-to" DSLR-style camera for many cinematographers, even though its Micro Four-Thirds sensor is smaller than APS-C or full-frame. With its faster native bitrate, and a faster CPU that hackers may well be able to tune for even more outrageous performance, the GH3 is likely to supplant the GH2 as the bargain camera of choice for cinematographers.

When "something for everyone" may be too much

On the cover of the current issue of "TWICE" (This Week in Consumer Electronics,) there's an ad for Nikon's cameras with the tagline "There's a Nikon for Everyone." It got me thinking about something I noticed at the Sony and Panasonic booths at last week's NAB conference. These companies have so many different camcorders and cinema cameras that even the people selling them can't keep track of all of them. For example, when I was in the Sony booth, I couldn't find the 35mm cinema cameras (NEX-FS100, FS700, F3, etc.) I asked one of Sony's salespeople where they were, and she said that all of the company's cameras were on display in the huge circular "camera pit" at the center of the booth. I'd walked around the entire pit and hadn't seen the 35mm cameras, so I went around again but didn't find them. It turned out that the 35mm cameras were in a completely separate section of the booth.

There are so many products that they overlap each other in price and functionality. The same is also true for still cameras from Canon, Nikon, Sony and others, and smartphones from Samsung, HTC, LG, Motorola, Nokia, etc. Makers of notebook and desktop computers have the same problem--just look at the proliferation of models at HP, Dell and Acer. Manufacturers make so many models in order to avoid losing a sale, but they wind up confusing potential customers. Each of these products costs a significant amount of money to develop, manufacture and support. Resources that could be used to develop entirely new products are instead used to create minor product variations to fit into every conceivable price point.

Apple is a great example of a better approach to the problem. At any one time, Apple has a single line of smartphones, tablets, and notebook, all-in-one, mini and full-sized desktop computers, each of which is refreshed once a year. Apple continues to sell a single version of the previous year's tablet and smartphone (two years in the case of phones) at lower prices. Each computer line has four or five models, which vary by display size and processor. When a new computer line is launched, the previous line is discontinued. It covers all the price points, yet it's simple for consumers to understand and for Apple to sell. It also works well with Apple's strategy of making product announcements into newsworthy events.

Sony lost $6.4 billion last year; Panasonic lost $10.2 billion. They no longer have the money to invest in endless product proliferation--which might explain the relatively paltry number of new products shown by Panasonic at NAB. They, and companies like Canon, Nikon, Samsung, etc., would be well advised to focus on fewer, better products that are clearly differentiated from competitors and from each other.

My year-end waste of time: Predictions for 2012

I've decided to participate in one of the most potentially embarrassing annual blogging rituals: Predictions for the coming year. So, for what it's worth, here are my predictions for 2012, in no particular order:

eBooks and Publishing

• Both the European Commission's Directorate for Competition Law and the U.S. Justice Department will file suit against Apple and five of the "Big 6" trade publishers (Lagadere's Hachette publishing group, News Corporation's Harper Collins, Holtzbrinck's Macmillan, Pearson's Penguin Group and CBS' Simon & Schuster) for eBook price-fixing under the agency pricing model. Bertelsmann's Random House most likely won't be charged, because it joined in agency pricing long after the other five publishers. All the companies charged will strongly deny any conspiracy to fix prices, but they'll all eventually agree to a consent decree (and the European equivalent) before the cases go to court. The settlement will require Apple and the publishers to make cash payments for consumer damages, and the agency model will be discarded. eBook distribution will go back to the wholesale model.
• There's also a possibility that the U.S. government and European Union will use the antitrust litigation as a lever to force the Big 6 to make their eBooks available to libraries on commercially reasonable terms. Currently, only Harper Collins and Penguin make their titles available for library lending, and both companies impose significant restrictions.
• eBook sales in early 2012 will follow the same pattern as the last few years--there will be a huge burst of sales in January and February as millions of consumers who received eReaders and tablets as holiday gifts stock up on titles. However, the year-to-year growth rate in eBook sales will drop, due both to the increased share of eBooks as a percentage of all book sales and higher prices from the Big 6 publishers.
• Even though the growth of eBook sales will slow, print sales will continue to decline. Independent booksellers in the U.S. won't pick up the slack from the closure of Borders, nor will they make big strides in increasing their overall share of U.S. book sales.
• The Big 6 publishers' pricing policies will continue to encourage sales growth for smaller publishers and self-publishing authors, as consumers experiment with less-expensive titles and find that many of them are just as good as titles from the top publishers.
• While the number of titles from medium, small and self-publishers continues to grow, the Big 6 will continue to cut back on the number of titles that they release, focusing even more on pre-sold authors and titles, series and backlist titles that are reissued with a variety of value-adds.
• The "eSingle revolution" (short eBooks, no more than 50,000 words and typically 30,000 words or less) will grow, with more conventional book publishers offering titles. In addition, more media companies from other fields (magazines, broadcasting, cable and the web) will enter the eBook market with eSingles, either by themselves or in partnership with established book publishers.
• $99 will become the top-end price for dedicated eReaders sold in the U.S.; someone (probably Amazon) will go to $49-$59 for an entry-level model. The ad-supported/no-ads issue will become moot, as consumers show that they're perfectly happy with a cheaper, ad-supported eReader.
• The tablet market in 2012 will look very much the same as the market at the end of 2011: Apple will continue to dominate the high end of the market, with two lines of tablets: A new "iPad 3" (although I'm not sure that'll be its name) at the current iPad 2 prices, and the existing iPad 2, possibly with fewer storage and broadband options, at $100 or so below its current prices (for example, $399 for a 16GB model). At the low-end, a variety of tablets will compete in the $149 to $249 range, led (at least for the first few months) by Amazon. I wouldn't at all be surprised to see Barnes & Noble drop prices of both the Nook Color and Tablet by $50, to $149 and $199 respectively.

Cameras & Camcorders

• We're almost certain to see new cinema camera models from Canon in 2012. The prototype cinema camera based on the EOS body will be launched, as well as at least one new model in the C3XX range, with improved electronics including auto-focus, auto-aperture and auto white balance and 10-bit log output. The new EOS model could be announced as early as NAB in April, and the new C3XX model is likely to be shown at IBC in September.
• Panasonic's AG-AF100/101 is getting a little "long in the tooth", so I expect a refresh of the model in time for NAB in April. I also expect the GH3 to be announced in the first half of the year.
• Given all of Sony's 2011 EVIL, DSLR and camcorder announcements, I don't expect any big announcements from Sony in 2012.
• AVCHD 2.0 (also called AVC Progressive) will become ubiquitous on all new cameras and camcorders supporting AVCHD.

Motion Pictures

• We'll see major consolidation at the U.S. movie studios, like what we've already seen at Paramount, with even deeper cuts. Studios will become even more conservative about which titles they greenlight for production, continuing to focus on remakes, series and pre-sold titles (very much like the big publishers). This risk minimization strategy will lead to even more boxoffice and home video revenue declines.
• Online movie rental services such as Netflix and Amazon will continue to increase their share of home video revenues, but what could have been a huge win for Netflix will be a much more competitive market, due to Netflix's self-inflicted wounds from 2011.
• Studios will rethink the value of 3D given audiences' rejection of the format, and will put more effort into using 3D well on a smaller number of "event" titles. That means that 2D-to-3D conversion, which has never worked well, will go away. Studios will have to come to grips with the fact that 3D, like Blu-Ray before it, will not be their financial savior. Even well-done 3D won't save movies that audiences don't want to see.
• UltraViolet, the "online digital locker" system supported by most of the major studios, will fail to get significant market share, although the studios won't give up on it in 2012. Consumers will find it too hard to use, not worth the effort and not a compelling reason to go back to buying DVDs and Blu-Ray discs.
• With an handful of exceptions, independent films will reach audiences through VOD and online streaming services, not through theatrical exhibition or sales of physical media.

Rifles vs. shotguns: The GoPro advantage

The rule over the years for camera and camcorder manufacturers has been to make a model for every need and every price point. Canon, Nikon, Sony and Panasonic sell everything from inexpensive point & shoots to DSLRs. All but Nikon do the same with camcorders--prices run from around $100 for YouTube-focused models to upwards of $100,000 for digital cinema cameras.

The "model for every purpose and every pocket" approach means that, as a manufacturer, you won't miss a sale because you don't have a model that a customer can afford or can use, but it has some significant downsides. One is that it's expensive to develop new camera designs, both in terms of money and time. Canon's new C300 digital cinema camera took two years to develop, and that was considered a "fast track" project that required adapting the electronics from an older camcorder design in order to meet its deadline. In addition, as development budgets get strained, it's necessary to "milk" designs by releasing cameras that are minor variations on each other. Not to pick on Canon again, but the T2i, 60D and T3i DSLRs are very similar to each other, with minor differences in areas such as LCD mountings and video settings.

There's another side-effect of having so many models--features are deliberately left out of some lower-priced models in order to avoid cannibalizing sales of more-expensive ones. Sony is famous for this; for example, a big reason that the FS100 only has a HDMI output instead of HD-SDI is to avoid cannibalizing sales of the F3 camcorder. There's no technical reason why the FS100 can't have HD-SDI--the less-expensive Panasonic AF-100 has it, and it was introduced a year before the FS100.

Some companies practice another approach--build a limited number of models (or even a single model) of camera or camcorder, with a very specific target market or application. That brings us to GoPro, a camcorder company based in Half Moon Bay, California. GoPro only sells two models: The HD Hero and the new HD Hero2. Physically, the two cameras are almost identical to each other, but the Hero2 has improved electronics and optics. There's about $60-$70 difference between the two models, and none of them sell for more than $300. According to company founder Nick Woodman, GoPro initially built ruggedized cameras for use by surfers and skiers, but they were designed to be used by two people--one to surf or ski, and the other to shoot the action. Woodman's revelation, and the core principle behind everything that GoPro sells, is that athletes want to take video or still pictures of themselves in the act, or from their point of view. That meant that GoPro's cameras needed to not only be ruggedized--they had to be tiny, operate automatically, and be mountable just about anywhere.

GoPro sells a suite of mounting kits that allow its cameras to be mounted anywhere from the exterior of a race car to a surfboard. The company has a library of incredible footage shot underwater, on skydivers, mountain bikes, snow skis, skateboards, even as the payload for a weather balloon at the edge of space. It also has accessories to make the cameras easier to aim, extend their battery lives, transmit their video via Wi-Fi and gang two cameras together for 3D video. Yet all of it is based on the same camera design, for the same fundamental application.



I was amazed by how crowded the GoPro booth was at the NAB conference last April. This is a under-$300 camera, yet broadcast professionals were packed into the booth. GoPro's cameras are used for shooting the contestants' points of view on reality game shows, for recording experiments on Discovery's "Mythbusters", and for use almost anywhere danger is involved. Two thoughts went through my mind:

• Someone is going to buy Woodman Labs, the parent of GoPro, and
• Surely one of the big Japanese camera or camcorder makers will jump into the market.

I certainly hope that Woodman Labs isn't sold--the scariest example of what could happen is what happened when Cisco acquired Flip Digital. Before the acquisition, Flip was the leader in the market for inexpensive, simple-to-use camcorders. Earlier this year, due both to competition from smartphones and mismanagement, Cisco shut down Flip completely. Whenever a big company buys a small, focused company, it's usually the small company that suffers. As for the second possibility, a Japanese competitor could try to copy GoPro's ideas, but they'll stumble on their need to be all things to all people. To build a viable competitor, you need to understand GoPro's markets and applications as well as GoPro does, and that's hard when you're also trying to build cameras for every possible market and application.

Had GoPro tried to enter the general-purpose camera or camcorder markets, it would have had its head handed to it. Instead, it dominates the point-of-view market, which it can effectively defend. There's a lesson there, not just for other small companies but for the big camera makers as well. It may be time to focus on a few markets instead of trying to compete in all of them.

Atomos' little solution to a big problem

From the standpoint of getting high-quality video and audio out of a camcorder, either HDMI or HD-SDI will work, but HDMI is a consumer connection--the connector doesn't lock, and it's designed for casual use. SDI, on the other hand, is a professional connection--the connectors lock, and it's designed for much heavier usage than HDMI. There have been converters that input HDMI and output HD-SDI and vice versa, but they've tended to be big and line-powered--not something that you'd want to carry with you wherever you go.

At IBC, Atomos, the Australian video recorder folks, introduced a new product, Connect, that solves the problems with existing converters. The Connect is available in two versions: HDMI to HD-SDI, and HD-SDI to HDMI. Each one is priced at 249 Euros/$349 (U.S.), and is only 42.5mm W x 29 mm H by 72.5mm L. They're powered by a Sony battery and fit on Sony, Canon and Panasonic battery plates, so they can be mounted easily on almost any device. They can also be ganged together, so that multiple converters can share a single external battery.

Atomos' Connect converters can add a HD-SDI interface to any camcorder, make DSLRs work with professional video recorders and switchers, and enable lower-cost HDMI monitors to be used with HD-SDI devices. The company expects to begin shipping the devices worldwide by the end of 2011.

Episode 3 of The Feldman File Videoblog

Episode 3 of The Feldman File videoblog has been posted below--if you can't view it here, you can watch it on YouTube. This week's episode covers:

• Sony U.K.'s announcement of the PMW-F3 "budget" professional camcorder with Super 35mm sensor
• Boxee's introduction of the Boxee Box, the forthcoming version of Boxee's software and new content partnerships
• Kno's announcement of prices and an availability date for its single- and dual-screen eBook readers for the higher education market

Panasonic's new GF2: Full HD in a compact camera

Panasonic's long-rumored GF2 has finally been announced, and Digital Photography Review has had a few days to test a pre-production model. Panasonic will ship the GF2 in the U.S. in January, and will release U.S. pricing in December.

The GF2 is slightly smaller and lighter than its predecessor, the GF1. (It's still slightly bigger overall than Sony's NEX-5, but when using its pancake lens, it's a bit thinner.) It uses the same 12.1 Megapixel imager as the previous model. In order to reduce its size, Panasonic had to remove some of the physical controls that were on the GF1 and replace them with menu options.

Perhaps the biggest change in the GF2 is that it now has Full HD video, rather than the AVCHD Lite in the GF1. That means that the GF2 has 1080/60i and 720/60p video (the 60i mode reads the image data off the sensor in 30p mode).  However, the GF2's AVCHD bitrate maxes out at 17Mbps, not the 24Mbps maximum that Panasonic's GH2 and some other DSLRs can achieve. ISO sensitivity has been increased to 6400 from the 3200 maximum of the GF1. The GF2 also has a built-in stereo microphone, but as with most DSLRs, you'll be better off capturing audio with an external recorder. It fully supports Panasonic's new 3D lens, which will turn it into a tiny, albeit limited, 3D camera.

Digital Photography Review didn't do image and video quality testing, so that will have to wait for another review. The fact that the GF2 only goes to 17Mbps AVCHD means that its video output almost certainly won't stand up to that of the GH2. There's no word about whether the GF2 provides a live output on its HDMI interface while recording, what the resolution of the live output is, and what (if anything) is superimposed on it by the camera.

The GF2 is clearly a consumer-oriented camera, a step up from point-and-shoots rather than a smaller replacement for a DSLR. It's unlikely to have many cinematography applications, especially if it's priced only a few hundred dollars less than the GH2.

Panasonic's AG-AF100: A long answer to a short question

Panasonic showed a "70% complete" version of its new AG-AF100 Micro Four Thirds camcorder at DV Expo last week. The AG-AF100 will ship in late December at a price of $4,995 (U.S.). Dan Chung of DSLR News Shooter interviewed Panasonic's product manager for the camcorder, Jan Crittenden Livingston, and asked her why a DSLR user should switch to the AG-AF100. She went into a long rundown of all the features that the new camcorder has that a DSLR doesn't.

Having spent most of my career as a product manager, I sympathize with Jan's situation, but at the end of the day, the right answer is to focus on use cases rather than "speeds and feeds". If you're primarily a still photographer and want to capture a little bit of video as well, go with a DSLR. Any good DSLR from Panasonic, Canon, Nikon or Sony will do a far better job of shooting stills than the AG-AF100.

However, if your primary interest is shooting video, you'll have to add at least $1,000 of equipment to any DSLR in order to fix its inherent problems with video. You'll need to add a mounting system to make it handle more like a video camera, a viewfinder (either a magnifier/loupe attached to the LCD or a separate electronic viewfinder) and an external audio recorder. Even after that, you still won't get all the features that you'll get in a camcorder purpose-built for video, such as HD-SDI video out, timecode sync in/out and long record times on internal media (in the AG-AF100's case, up to 12 hours of continuous shooting on two 64GB SDXC cards).

If your primary interest is still photography or your budget is extremely strapped, buy a DSLR and the cheapest add-on hardware you can find, but if you can afford it, the AG-AF100 and future camcorders like it will be a better solution for full-time video use.

A new model for video acquisition and live production

Earlier this week I attended an event at Morningstar's headquarters in downtown Chicago. To get there, I walked past the studios of WBBM-TV, CBS' owned-and-operated station in Chicago. Like many of the new television studios built in major cities in recent years, WBBM's news studio is surrounded by windows and visible to the street from two sides, so that pedestrians can watch the shows as they're produced.

I was passing by in mid-afternoon, so there was no production going on, but I could take a look at the equipment. There were three high-end studio cameras on robotic pedestals (only one person is needed to operate all three cameras) plus a smaller camera on a fixed pedestal looking out the window for street shots. I looked up the list price for a typical high-end studio camera from Sony that's like the ones used at WBBM, and it costs $100,000 (U.S.).

By comparison, Panasonic can equip a HD webcasting studio with three automated cameras, a camera control unit and switcher for $23,000. One person can simultaneously switch the show and operate the cameras. Add in everything else you need--tripods with heads and dollies, a computer for graphics, an audio mixer, wireless microphones, LED lights, a live streaming encoder and software--and you'll be around $40-45,000. under $50,000 (I priced it out, and a complete system is priced around $47.000.) So for about half the cost of a single high-end studio camera, you can have a fully-functional live HD production facility (that, incidentally, is small and light enough to fit in the back of a minivan.)

WBBM, of course, is an over-the-air broadcaster that's also carried on cable, satellite and IPTV systems all over the greater Chicago area, so even though its news programs are struggling in the ratings, it gets a huge audience in comparison with what most people could attract through webcasting. However, the webcasting operation can run at a tiny fraction of WBBM's budget, so:

• It doesn't need to generate a lot of revenues to be profitable, and
• Once it reaches profitability, it will have a much higher operating margin than the broadcaster

A couple of years ago, I visited Current TV's headquarters in San Francisco and saw how they were using computers and data-based video workflows to build an automated cable network at a fraction of the cost of legacy networks. Just as Current changed the model for how to build a cable network, these new systems from Panasonic and other vendors are changing the model for video acquisition and live production.

Panasonic to intro Micro Four Thirds 3D lens by year-end

According to Engadget, Panasonic is working on a 3D lens for its Micro Four Thirds cameras (G1 and G2), and plans to ship it by the end of the year. The 3D lens will use the Lumix G-series mount. It will operate on the same principle as the snap-on 3D lens just announced for the HDC-SDT750 camcorder; the right and left eye images will be split and recorded together on one imager. The focal length and aperture range of the lens hasn't yet been disclosed, and the lens shown by Engadget is still a prototype.

Panasonic is clearly very serious about 3D for both camcorders and DSLRs, and it'll be interesting to see how other manufacturers respond.

Panasonic's new consumer 3D (maybe) camcorder

Panasonic has formally announced the HDC-SDT750, a consumer-targeted 3D camcorder. Instead of creating a true full HD 3D camcorder, which would have required two imagers, Panasonic uses a snap-on 3D lens that splits the right- and left-eye images and sends each one to half of the imager, resulting in a 960 x 1080 image for each eye. Both images are captured in one 1920x1080 frame. (When the 3D lens is removed, the camcorder can be used for conventional 1080P recording.) About the only thing you'll be able to do on Day One will be to run the video from the camcorder out to a 3D HDTV for direct viewing, but I suspect that Adobe and Apple will build support for Panasonic's format into future versions of Premiere Pro and Final Cut Studio. The HDC-SDT750 will ship in October, priced at $1,399 in the U.S.

Sony: What's that word for "copycat" in Japanese?

Not long ago, pictures of a 3D camcorder from Sony that's nearly a copy of Panasonic's AG-3DA1 leaked out, and today, Engadget is covering a preannouncement from Sony about a new camcorder that uses the same sensor and interchangeable lenses as its forthcoming NEX3 and NEX5 digital cameras, as well as its Alpha DSLRs. This new product is largely a clone of Panasonic's AG-AF100 camcorder that's based on that company's Micro Four-Thirds DSLRs. Neither of Sony's products was shown or even rumored at the NAB conference just a few weeks ago, so it's clear that Sony is playing "fast follower".

Many years ago, Matsushita Electric, the original name for Panasonic, was accused of copying designs from Sony and other companies. In fact, Matsushita was sometimes derisively referred to as "mohousha", or copycat. Now, Sony seems to be the "mohousha", copying designs and ideas from Panasonic, Nintendo and others.

Sony to clone Panasonic's 3D Camcorder

Engadget has published a picture of a prototype 3D camcorder from Sony that's based on its XDCAM PMW-EX3. It looks very much like the dual lens from Panasonic's AG-3DA1 3D camcorder has been transplanted on the front of the EX3. We're likely to see this camcorder "in the flesh" at IBC in Amsterdam in September, if not before then.

All the big announcements at NAB in 35 seconds!

Google announced its new Search Stories feature today--based on the idea behind the company's 2010 Super Bowl ad. I used it to showcase the new products at NAB 2010 that I think will have the most impact:

Panasonic may be making the biggest news at NAB

A few years ago, I was at NAB when Panasonic announced its P2 flash memory camcorders and media. They were models under glass, not even working prototypes. The P2 media was going to be very expensive, capacities were low, and the whole idea of using flash for professional video production was untested. Sony had recently introduced its XDCAM optical disc technology, and the company laughed at Panasonic for moving to P2. Flash forward a few years, and flash memory has become the standard for most video production; even Sony has jumped in with both feet. In a few years, I think that we'll look back and see that Panasonic is doing the same thing this week.

The first big news is the $21,000 AG-3DA1 3D camcorder. There are a number of things that make the AG-3DA1 interesting: First, it's a one-piece 3D AVCHD camcorder that records in 1080p and 720p at frame rates up to 60fps, depending on resolution. It can adjust and track convergence points and horizontal/vertical alignment from within the camera, without using a separate computer. It uses dual SDXC cards for storage and weighs just over 6 pounds (2.8 kg). In short, it's a professional 3D camcorder that you can take out and start shooting with immediately. 3D production is much harder than shooting in 2D, but the AG-3DA1 could make 3D both easier and faster.

The second major announcement is the (estimated) $6,000 AG-AF100, which I wrote about earlier this week. The AG-AF100 takes the guts of Panasonic's Micro Four Thirds digital still cameras and puts them in a camcorder body. The biggest single complaint that cinematographers have with DSLRs is that their ergonomics are designed for photography, not video, so they have to kludge together eyepieces, external monitors and mounting hardware to make DSLRs handle like camcorders. The AG-AF100 offers the sensor size and interchangeable lenses of DSLRs together with the ergonomics and features of a professional camcorder.

The AG-AF100 is a harbinger of what's to come. Consider what would happen if Canon put a 7D, 5D Mark II or 1D Mark IV, or if Nikon put a D3s, in a camcorder body: High resolution, big sensors, multiple frame rates and interchangeable lenses (including some new killer prime lenses from Zeiss.) If you've been following the DSLR shootout that Zacuto's been doing, you know that the new DSLRs can do things that motion picture cinematographers have dreamt about for years. Put that capability into a familiar package,  and you redefine the professional camcorder business.

The AG-3DA1 and AG-AF100 are both game-changers. Even if they don't go on to be big sellers, they're establishing new directions for professional camcorder design that will fully play out over the next few years.