We have recently published excerpts of our interview with Todd Rockoff . Today we publish the interview in full. Todd represents the HDcctv alliance and everything they are currently trying to achieve. With the growth of HD over coax comes responsibility and the HDcctv alliance are setting the standards and educating the world of CCTV why it is so important to ensure all things HD SDI should be interoperable. This approach makes sense and ensures the technology will grow with all involved having the common goal of a uniform technology and interoperability.
Question 7 is of particular importance for the HDcctv alliance.
However we started by asking about lenses, the interview is in depth but worth reading to find out everything you need to know about how the HDcctv alliance are forging the future of CCTV and HD in general across the world.
1 ) Todd when it comes to lenses there currently seems to be few
guidelines. What lenses are the correct lenses to use – will the
alliance regulate or at least stipulate the required lens
specification for use with HDcctv cameras?
TR: HDcctv is an electrical/mechanical interface standard for connecting devices within a secured site. An HDcctv interface transports an HDTV signal without any degradation. HDcctv compliance is a property of devices (such as cameras, repeaters, IP encoders, DVRs, and monitors) and not of their individual components. That approach allows manufacturers maximum flexibility in designing those products. As a result, the HDcctv standard does not directly constrain lens characteristics in any way.
HDcctv compliance guarantees 100% multi-vendor interoperability, but as with analogue CCTV cameras, manufacturers compete on video quality. The lens, the image sensor, and the image signal processor (ISP) all contribute to a camera’s video quality. The lens is an integrated component in dome cameras and board cameras.
The HDcctv Alliance plans eventually to define “HDcctv 1080” and “HDcctv 720” compatibility levels, with associated certification tests. The idea is that an “HDcctv 1080” compatible camera, for example, would measure a full 1920 x 1080 pixel array in each frame. If any camera component, including the lens, lacked sufficient resolution, then a 1080 or 720 compatibility certification could not be obtained.
For an HDcctv box camera today, it makes sense to choose a CS-mount lens of appropriate resolution. Be careful never to fit an old CCTV lens to an HDcctv camera! The result would be fuzzier video than necessary, failing to take advantage of the high-definition sensor and ISP. Similarly, to the extent that higher-resolution lenses are more expensive, there may be no need to fit a very high resolution lens to an HDcctv camera, because that would be overkill. On the other hand, some buyers find better results with slightly over-spec lenses. For example, sometimes a lens rated at 5 megapixels might give better results than a lens rated at 3 megapixels for a 1080p25 camera, whose native resolution works out to 2.2 megapixels.
I wish there were a more prescriptive approach to lens selection. Ultimately, just as buyers find their own comfort levels with camera price and performance, so should buyers explore price/performance tradeoffs in lenses for HDcctv cameras. One thing is for sure: product availability continues to grow, so there are plenty of alternative HD lenses from which to choose.
2) With HDcctv Recorders a lot of focus is spent on the quality of
locally recorded images and this obviously affects bandwidth and
storage – but there is a huge market for remote monitoring that
analogue covers right now, will recorder manufacturers consider
features such as throttled/reduced quality transmission for remote
monitoring?
TR: Certainly, HDcctv DVRs typically provide all the remote IP video capabilities of conventional CCTV DVRs. DVRs typically allow the operator to specify the tradeoffs among transmitted frame rate, resolution, and quality (amount of compression) to optimize the use of available off-site bandwidth. One of the appealing aspects of DVR-based architecture is that the DVR allows those bandwidth optimization tradeoffs for all cameras on the local site to be managed in one device; with MP IP cameras, by comparison, optimizing the use of available off-site bandwidth requires carefully configuring each IP camera, as well as possibly the NVR, in a harmonized manner.
3 ) Is it possible for MP resolution to increase in the future and if
not will HDcctv be able to compete with ever increasing IP camera
resolutions at decreasing prices?
TR: There are two ways in which the HDcctv standard is addressing higher-than-1080 resolution video.
In the near term, HDcctv 2.0, due to be completed in early 2013, provides for a 75Mb/s data stream to be transmitted from the camera in parallel with uncompressed HDTV signal. An HDcctv 2.0 camera could send higher-than-1080 video in this data stream in a compressed form, possibly as IP video.
In the longer term, future versions of the HDcctv standard may take advantage of improved cable driver / equalizer chip technology to send uncompressed video at higher pixel rates than HDcctv 1.0, which could translate to higher resolutions and/or higher frame rates.
4 ) Over the last 5 years the use of cat 5 and ballum technology has
meant less coax out in the field. We know plans are afoot for HDcctv
over cat 5 but when will this be viable and ready to sell to end
users?
TR: HDcctv CX provides for native transmission over 100m Cat-5e cable. The HDcctv CX standard is in preparation now, and we expect it to be completed by early 2013. Chips implementing the standard will be available within six months or less of the standard being completed, so HDcctv CX-compliant products should be available in the first half of 2014.
5) One argument for IP is PoE .Is there any development of technology
that would send power to the camera down the coax like the old line
fed systems?
TR: Yes, all of the current developments (HDcctv 2.0, HDcctv CX, and HDcctv XR) are being defined in anticipation of being able to send relatively high current up the cable in HDcctv 3.0. In the case of HDcctv CX, we expect to be able to leverage proven PoE+ solutions directly.
6) We hear of major developments with regard to chip technology that
will make HDcctv both better and cheaper. Can you outline exactly what
the benefits will be and when it will be available?
TR: Most of the leading semiconductor players in the surveillance industry, including two of the world’s top three ISP vendors, are implementing the features needed for HDcctv compliance in their current-generation chips. This industry development means that HDcctv compliance is becoming easier for device manufacturers to achieve.
Meanwhile, more than one HDcctv Alliance chip-maker Member has developed a multi-channel receiver chip for HDcctv-compliant DVRs. Multi-channel receiver chips are part of the answer as to why CCTV DVRs are so cost effective. As the multi-channel receiver chips become readily available to HDcctv DVR makers, HDcctv DVRs will be better able to compete head-to-head with CCTV DVRs with respect to make cost.
Currently, most DVRs implement HDMI outputs. HDcctv is better for surveillance than HDMI, because it is less expensive per port and carries signals over longer distances. For these reasons, and as costs continue to come down as the industry scales the learning curve, native HDcctv monitors are expected to become increasingly popular.
7) You have campaigned long and hard to convince the industry that
HDcctv compliance is the way forward. Few disagree but is there signs
of more members coming on board and do you think by not joining
manufacturers who are just selling HD SDI without standardizing risk
being marginalized.
TR: The surveillance market is moving toward HD video, whatever the transport technology. We have campaigned long and hard to convince the industry that HDcctv is a superior alternative to MP IP cameras in many cases, and the fact that an outright majority of the world’s CCTV manufacturers has already embraced HDcctv in some form validates the HDcctv value proposition, which seemed so controversial just four years ago: HDcctv is easy to design and manufacture while cost-effectively delivering typically excellent end-user experiences.
Installers who buy surveillance equipment through distribution expect 100% multi-vendor interoperability and assured electrical performance. These properties are essential to be able to mix and match products from disparate manufacturers. Since distribution accounts for more than 80% of all cameras sold worldwide, these properties are economically significant. HDcctv compliance certification is the only way a manufacturer can warrant 100% multi-vendor interoperability and certain electrical performance. The various interpretations of HD-SDI get partway there, delivering HD video in some way, but there is no compliance certification regime for any of the HD-SDI variants.
The steady growth of the HDcctv Alliance reflects an inevitability that most manufacturers will ultimately seek HDcctv compliance certification for their HDcctv products, because testing to the international standard is required for a manufacturer to be able to warrant interoperability and electrical performance, and these properties are essential for sales through distribution. HD-SDI was developed to meet the needs of broadcast television studios, rather than surveillance. The need for compliance certification becomes even more important for those manufacturers seeking to take advantage of the advanced, surveillance-specific capabilities of HDcctv 2.0/CX/XR.
What does it mean for a manufacturer to ship a proprietary interpretation of HD-SDI instead of embracing HDcctv, the global standard? It may be that the manufacturer is grappling with the technical issues that inevitably arise for the first products in a new product category and has not yet placed a high priority on compliance certification. Ultimately, given that compliance certification is valuable and not intrinsically expensive, persistently refusing to certify compliance might raise questions in some buyers’ minds about the manufacturer’s overall commitment to quality.
8) This year at Ifsec the bias towards HD over coax was obvious and
Samsung are one of the big names we see that are almost reluctantly
deciding they have to join the fray. Why the reluctance originally on
their part?. Will they be joining the alliance? And are there other
big name manufacturers waiting in the wings to launch HDcctv over
coax?
TR: I cannot comment on any non-Member’s plans. In general, the fundamental economic considerations discussed above motivate every manufacturer hoping to sell HDcctv products through distribution to seek HDcctv compliance certification at some point during the manufacturing engineering cycle.
You raise a fascinating question that has significant implications for our industry: Despite a few notable exceptions, why have the incumbents largely abandoned the HDcctv market opportunity to more nimble competitors, in favour of focusing only on MP IP cameras? There certainly are situations where IP cameras are convenient and cost effective, but not every situation. After some 15 years of trying and failing to gain significant market share, the one area where IP cameras have been beating CCTV cameras in recent years has been HD surveillance. When MP IP cameras were the only option for HD surveillance, it was easy to mistake the market migration to HD surveillance as the beginnings of a long-awaited mass migration to IP cameras, and many companies built their long-term strategies around the ultimate dominance of IP cameras. It can be difficult for a company, having made a huge investment developing IP cameras and educating installers to use IP cameras in all cases, to consider an alternative, no matter that HDcctv is a more practical path to HD surveillance in many situations. That is one reason why the incumbents have been hesitant to embrace HDcctv. Meanwhile, smaller and more agile companies are delivering HDcctv solutions to meet the market need for HD without the I.T. issues introduced by IP cameras.
Ultimately, surveillance product innovation is delivered in semiconductors. The HDcctv Alliance includes about 15 chip-maker Members collaborating on defining the standard and implementing the standard’s features in their chips. As those chips become ever more widely available, expect an ever-growing number of equipment brands to begin to take advantage of the capabilities in their product lines.
HDcctv technology is already in the process of disrupting the competitive landscape in the surveillance industry: those failing to offer HDcctv solutions are missing out of one of the most dynamic product categories ever to hit the industry.
9) Have you seen a marked increase in feature rich HDcctv DVRs as
manufacturers start to push for market share. If so, what features do
you think are bringing the most benefit to end users and what features
that we are yet to see should the manufacturers strive to introduce?
TR: HDcctv DVRs exhibit the same general architecture as CCTV DVRs. The fundamental difference is that instead of an analogue-to-digital converters on the inputs, HDcctv DVRs include all-digital, high-speed receivers on their inputs. This means that all of the features evolved over time for CCTV DVRs are typically already available in HDcctv DVRs. From the installer’s or operator’s point of view, an HDcctv DVR works just like a CCTV DVR, except with higher resolution values on the configuration pull-down menus. All of this means that HDcctv equipment is easy both to design and to deploy, with little-to-no learning curve for the seasoned CCTV professional.
One as-yet-unexplored product opportunity is multi-channel video analytics (VA) servers. Most VA solutions today operate on somewhat distorted IP video or in computationally limited cameras. A terrific opportunity exists to deliver a cost-effective, high-performance VA server that accepts multiple channels of HDcctv inputs.
Another important product area is installers’ tools. Specifically, HD field monitors are important for setting up cameras, and in-situ test equipment would be very important for the installer considering an HDcctv retrofit, to be able to answer the question, “Will this cable support HDcctv 1.0, or do I need HDcctv XR, or do I need to pull new cable?” prior to actually installing the HDcctv equipment. HDcctv Alliance Members are working on these tools and more; keep your eye on developments in this area.
10) HDcctv cameras come in many shapes and sizes but is there
companies out there that have developed HDcctv PTZ domes. If so who?
TR: Several HDcctv Alliance Members specialize in PTZ domes, but only EverFocus has so far certified a compliant speed dome. The full list of certified compliant products is maintained at http://www.highdefcctv.org/compliant-product-finder .
11) IMS predictions have shown exponential growth in the sale of
HDcctv solutions across the world but where do you expect to see most
sales to end users taking place.
TR: The appeal of HDcctv solutions is very broad based, and it is not inherently limited to any specific geographies or vertical market segments. HDcctv cameras are being deployed everywhere from small retail shops and petrol stations to large, enterprise-connected warehouses, from schools to hotels, from freeways to public plazas, right around the world. As more installers and their customers become aware that HDcctv is an appealing alternative for HD surveillance, and as prices continue to become more attractive, this adoption trend will only accelerate.
We have observed that HDcctv solutions tend to be adopted relatively slowly in those markets where the all-IP-cameras proposition has gained significant mindshare. When a tender document specifies that a surveillance system must contain only IP cameras, the system designer does not have the freedom to choose the most cost-effective solution for the application. Ethernet and HDcctv are just local-site video transport technologies, rather than risk mitigation strategies; selection of local-site video transport technology is appropriately an engineering decision. Of course, IP cameras are the right choice for some cameras, but certainly not for all cameras in all projects. A buyer specifying only one type of camera may spend too much, or end up with a system that is not quite fit for purpose. HDcctv adoption has so far been fastest in those markets and applications wherein surveillance system designers are free to select the most cost-effective technologies to address the end buyers’ security concerns on a case by case basis.
12) Recently the BBC ran an article on HD CCTV in the UK and the
threat to privacy this represented. It certainly bought the subject to
the fore here in the UK but what is your thoughts on this article and
is HD a threat or a benefit.
TR: The HDcctv Alliance does not take positions on matters of public policy. Suffice to say that whether or not surveillance is appropriate in a given situation is independent from the local-site video transport technology used. In general, clear surveillance is better surveillance. Clearest video is the value proposition for HD surveillance, and the reason that the worldwide market is migrating to HD video. HDcctv is an appealing way, but not the only way, to deliver HD surveillance.
13) The argument for HD IP verses HDcctv seems to have died down
recently with an acceptance in the market place that each technology
fits well into the right application can you expand on where you think
HDcctv is best suited and why?
TR: I thought you might bring up the old “MP IP vs HDcctv” argument 🙂 The HDcctv Alliance began making the case for HDcctv cameras at a time when chips implementing HDcctv interfaces were just coming out, and only a few HDcctv products containing those chips were initially available. There was a lot of resistance to the very idea of HDcctv back then. As we were saying, “Not every camera should be an IP camera,” many heard us saying, “Every camera should not be an IP camera,” which of course is not true. Now that a broad, growing diversity of HDcctv products is available, everyone can see for themselves that, as you suggest, each technology fits better in certain applications.
As always, it is not possible to characterize all surveillance applications, because of the highly fragmented nature of this market and this industry. Ultimately, each project presents its own challenges and its own design considerations. Here are some examples of situations wherein either MP IP cameras or HDcctv cameras tend to be preferable:
• HDcctv cameras deliver the best possible live view experiences, because the video signals are not compressed prior to live view. IP cameras compress video prior to Ethernet transmission, and compression degrades video quality by removing information from the signal.
• MP IP cameras deliver the highest possible resolution video streams, going beyond the 1080p maximum resolution of HDcctv cameras. HDcctv will not go beyond 1080p until HDcctv 2.0 cameras are available next year.
• HDcctv transmission is intrinsically reliable, whereas Ethernet packets may be lost in transmission. The result of lost packets is dropped frames and/or repeated frames in the transmitted video.
• MP IP cameras can cover very large spaces via trees of Ethernet switches, whereas implementing home runs for HDcctv cameras becomes expensive for such large spaces.
• HDcctv cameras conveniently re-use legacy CCTV power & cabling. With HDcctv CX, even legacy installations fitted with Cat-5e can be retrofitted to transmit uncompressed HDTV signals on site. By contrast, MP IP cameras require LAN infrastructure.
• MP IP cameras enable WiFi implementations, which is ideal when power is available at the camera mount but there is no physical cable. By contrast, all HDcctv cameras now and in the near future require a physical cable over which to transmit the video.
• The skill set needed to commission an HDcctv camera is the same as the CCTV skill set, whereas MP IP cameras draw on a different, I.T.-based, skill set.
To the surprise of some in our industry, there are some circumstances which do not really intrinsically favour either type of camera:
• Off-site communications is nearly always accomplished with IP networks. However, DVRs and IP encoders are very cost-effective “on-ramps” to IP networks. Integration with IP networks certainly does not favour IP cameras.
• Video analytics (VA) performs best when applied to uncompressed input video. Because of the degradation of video for Ethernet transmission, many IP cameras perform VA in the camera. However, IP cameras are not inherently better at VA than alternative solutions. For example, HDcctv provides for metadata transmission, so an HDcctv camera can also implement VA. Moreover, a multi-channel VA server, receiving uncompressed HDTV signals from a collection of HDcctv cameras on a local site, can be a very cost-effective VA solution.
14) since 2009 HDcctv has seen a massive rise in awareness amongst the
public but are you happy enough people understand the potential and
what can be done to further raise awareness about HDcctv?
TR: I appreciate your enthusiasm, but I think the majority of our work on mainstream awareness lies before us. Technology enthusiasts and early adopters know about HDcctv and are using HDcctv products appropriately. And it is clear that most manufacturers are generally aware of HDcctv. However, mainstream awareness remains elusive. Indeed, many manufacturers who have produced HDcctv prototypes, as HD-SDI variants, do not themselves yet grasp the strategic implications of the HDcctv standard.
The HDcctv Alliance plans to continue to expand our education and outreach efforts, drawing on the coordinated resources of all of our Member companies, delivering seminars and publishing articles in forums around the world.
15) If we fast forward 5 years to 2017 what do you think commentators
will be saying about the HDcctv alliance and HDcctv over coax/cat5 in
general?
TR: HDcctv is the only open, comprehensive electrical interface standard for HD surveillance video. By definition, while manufacturers differentiate on product interfaces (as is the case with the proprietary variants of HD-SDI today), interoperability remains elusive and market growth is impeded. In order for the HD surveillance market to reach its maximum potential, HDcctv will be well on its way to becoming the predominant interface technology by 2017, supplanting NTSC/PAL.
An inevitable consequence of the HDcctv Alliance successfully driving global adoption of a powerful standard is that HDcctv itself will fade to become a tick box on the product specs rather than a highlighted feature. Just as has become the case with analogue CCTV products, HDcctv product manufacturers will compete on reliability, usability, image quality, performance features, form factor, customer service, manufacturing quality, price, and so forth. Universal compliance to the HDcctv standard ensures that all manufacturers’ products interoperate, enabling customers to mix and match with confidence. The more successful the HDcctv Alliance, the less people will be talking about HDcctv per se.
16) Can we expect anymore announcements or surprises in the field of
HD over coax and are you able to give anything away to us and our
readers today?
TR: HDcctv technology is dynamic: the HDcctv standard continues to develop, and chip makers are putting new chips in the hands of equipment makers that implement advanced capabilities in smaller form factors at even more attractive price points. The web page http://www.highdefcctv.org/hdcctv-specification is kept up to date. Surveillance system designers are well advised to be informed about HDcctv developments, because the tradeoffs outlined above will continue to evolve.
