IP video: Where's the tipping point?

For more than a decade, anyone who was anyone predicted it was coming: exponential growth in IP video. Old analog companies were either rotting by the wayside or acquiring the new technology to stay alive. But even after the large acquired the small, it still took time to adapt to the 'new way.' They said analog/IP sales would be 50/50 by 2005; then it was 2006, 2007, 2008. Each delay was accompanied by an array of plausible excuses from the very people who claimed with overwhelming confidence that it was coming. Among the reasons stated: a weak economy, smarter DVRs rising to meet the challenge, hybrid recorders, high-quality encoders, incredibly reliable analog cameras, a decline in new sites, integrators slow to learn the necessary skills and over-selling by manufacturers. Early adopters found that bleeding-edge IP technology was not without a downside, often needing such frequent patch updates that an installation would turn into a series of upgrades. Integrators complained of overly complex technology and low margins from countless maintenance and site visits.

The reality is that different markets adopt new technology at different rates for any number of reasons. The same is true for video surveillance technology. In the last few years, I've not encountered any casinos replacing older DVRs with newer DVRs, nor airports. When they upgrade their video surveillance, they are usually opting for IP video instead. The transportation market, too, has embraced encoders instead of DVRs and occasional IP cameras for new sites and forklift upgrades. Utilities, education and a whole swathe of military and civilian government entities are all moving quickly to encoders and IP cameras.

But why are some others voting for no change? A big chunk of the market has reached the conclusion, at least for the time being, that analog cameras (coupled with a DVR and a PC running some free viewing software) is good enough for them. It's relatively inexpensive (compared to pure IP video) and it's mature technology which means it's reliable, compatible, easy to install and requires fewer repair visits. DVRs also offer features normally associated only with high-end systems, such as H.264 compression and field-replaceable RAID storage. Customers may not achieve the video quality equivalent of HD/megapixel, but they get video quality, retention and reliability that are simply put, "good enough." Moreover, for integrators and installers, DVR systems are easy to quote and project success is almost always predictable. But there is a downside. Competition is ruthless, including do-it-yourself all-in-a-box systems from the local big box or other Home Depot and even from Amazon.com.

Hard to predict who wants what

On the other hand, IP video advocates would argue that HD and megapixel IP cameras take video quality to a new level, blowing past the NTSC limitations. This is true, and I am always surprised by the number of DVR installations that elect to economize on storage and bandwidth at the expense of video quality. I have seen this in installations where I would have least expected it.

Still, different market segments place a fundamentally different value on their video and what they expect out of it. To illustrate, consider these two very different types of customers.

At one extreme is the small installation, with a DVR gathering dust. It hasn't been touched in months, possibly years. The system's video quality is about the equivalent of a time multiplexed VHS tape-a handful of frames per second at CIF resolution. Even if the DVR failed, it would probably go unnoticed. You can well imagine that extolling the virtues of conventional IP video to such a customer would likely fall on deaf ears.

At the other extreme are the command and control centers that do round-the-clock monitoring and are equipped to the hilt, with walls of cameras tightly integrated into access control and other systems and video recorded on fault-tolerant RAID storage devices using redundant network video recorders (NVRs). No expense is spared. Picture complete redundancy built into the system from top to bottom, along with sophisticated alerting capabilities for NVR failures, camera tampering or other malfunctions. Usually in this type of setup, the image quality can be adjusted for live or recorded video, with higher resolution images for real-time video (30 frames per second at 4CIF or with megapixel IP cameras).

So given these two extremes, where is the tipping point to IP video? The answer depends on the end-user application and market. For some, the tipping point has come and gone. For others, it's not even on the radar. It's meaningless to argue whether IP video is universally superior or inferior to analog cameras and DVRs because customer needs are not universal. The tipping point is not so much a tipping point, but more a swinging pendulum.

So where is IP video becoming the de facto standard? From my observation, some key markets include airports, mass transit, utilities, gaming and education. Three-letter agencies are also quick to adopt IP video for its quality, reliability and flexibility. Other commonalities with IP video systems are that they are typically integrated to access control systems and involve multiple recording and monitoring locations, as well as higher camera counts. While DVRs can accommodate these environments to a certain extent, you'd be hard pressed to find the same world-class video management capabilities in DVR-based systems as you would in systems comprised of NVRs, encoders and IP cameras. Some other things that have also helped to spur IP adoption in these markets are turnkey solutions (where the video management software is pre-installed on carefully pre-configured server hardware for optimum performance); dropping costs of servers and storage; and increasing IP competency and IP networking skills for system integrators and installers. (Don't forget, the Internet boom only started in 1999).

There are two other factors that I think will eventually help swing the IP pendulum for late adopters and those are the release of the ONVIF and PSIA interoperability specifications and the proliferation of compliant edge devices and head-end software. These trends will go a long way toward eliminating incompatibility and strengthening the concept of plug and play, which will in turn increase designers' and installers' confidence in IP video.

And, at the end of the day, for integrators and installers, the tipping point to IP is all about that "C" word-"confidence." Some integrators will source video management software, workstations, servers and storage from a variety of conventional IT distributors. But for most, there is still the sense that anything that reduces risk and gives them a single phone number to call if something goes wrong is the way to go. Going with the proverbial 'one throat to choke' (turnkey) approach-(where for example the NVR software is pre-configured on NVR servers with storage included)-can be easier and can eliminate the possibility of "blame storming" when multiple vendors are involved.

But perhaps the biggest driving factor for integrators/installers to beef up their IP video capabilities might just turn out to be the very thing they don't know. I'll frequently hear integrators/installers in a particular region report that there's not much demand for IP video in their territory, yet I am fully aware, down to the exact cent, how much their competitor down the road is generating, at least with my product lines. Eventually when they become more aware of the opportunities they're missing out on, the IP pendulum will swing. Or, they will find themselves progressively marginalized into low-end projects. True, analog cameras and self-contained recorders like DVRs will be around for many years yet, but will become increasingly commoditized. Even if they haven't yet jumped on the IP video bandwagon, forward-thinking integrators and installers need to adopt a long-term strategy that sets them on a path to gaining real-world IP video experience. One way to do this is to partner with a technology provider that can help build their confidence and support them in the transition.

Dr. Bob Banerjee is senior director of Training and Development for NICE Systems Security Division, Rutherford, N.J. Banerjee develops programs and initiatives to educate, train and support NICE's extensive network of security system integrators and dealers and provide thought leadership for NICE's security industry outreach efforts. He holds a Ph.D. in Artificial Intelligence from the Advanced Research Center at the University of Bristol, England and can be contacted at Bob.Banerjee@nice.com.