Key Decision – Camera Count

The first choice that a security dealer must make in selecting a DVR is the maximum number of cameras that the DVR will accept – typically, four, nine or 16. Bosch usually recommends that dealers buy DVRs with at least two extra channels to support future expansion, says Justin Lott, Bosch product marketing manager.

Most of today’s DVRs provide functionality that previously required a multiplexer, as well as a VCR. Like a multiplexer, today’s DVR can accept multiple analog cameras and prepare input from all of them for recording onto a single medium. Although DVRs that will accept a single input from a stand-alone multiplexer are available, Gary Perlin, Speco Technologies vice president of video, says multi-channel DVRs are much more popular because they provide more flexibility in recording options.

Unlike multiplexers – which in normal operation record all cameras continuously and at the same frame rate – DVRs may be able to record images from certain cameras only during certain hours or may increase the recorded frame rate during certain hours. Some DVRs have built-in motion detection and can begin recording or increase the recorded frame rate when they sense a change in the image viewed.

Key Decision – Frame Rate

An important decision in selecting a DVR is the maximum frame rate that the unit can record. The highest-end DVRs are capable of recording every camera simultaneously at the maximum rate of 30 frames per second (fps) specified by U.S. video standards. By recording at the maximum frame rate, dealers can help ensure that details are captured, even if a subject is in motion. For a 16-camera DVR, for example, the maximum possible recording rate is 480 (or 30 times 16) fps. But, because hard disk storage costs can be quite high for such a system, many systems record at lower frame rates except when motion is detected. Alternatively, they may use continuous recording at 30 fps for just one or two cameras in a system, such as a camera positioned directly above a cash register drawer.

To calculate the maximum frame rate needed, a dealer should first plan the recording parameters for each camera. If most cameras will not record continuously at 30 fps, the dealer may save money by purchasing a DVR with a lower maximum frame rate, such as a 16-camera system that records a maximum of 120 fps. Fully loaded, such a system would enable each camera to record continuously at an average of more than 7 fps, which exceeds the frame rate of many multiplexed VCR-based systems. The dealer should recognize, though, that if cameras are programmed to record at 30 fps when motion is detected, recorded frame rates for other cameras in the system will decrease during that time period.

Key Decision – Compression Format

A dealer’s choice of compression format will depend on a variety of factors, including price, recording parameters, and whether or not a high-security application is involved. The goal of compression is to minimize the number of bits required to transmit and store video images.

There are two types of compression. Full-picture or loss-less formats, such as wavelet and the JPEG/ MJPEG series, compress individual frames. Change only formats, such as the MPEG series, use certain frames as reference frames, recording and transmitting only the information that changes from one frame to the next (such as when a person steps in front of the camera). Depending on how cameras are used, change-only formats can help minimize hard-drive storage requirements in comparison with full-picture formats. But if a substantial amount of motion is recorded, change-only formats may operate at virtually full-picture rates, negating potential storage benefits, says Neil Heller, product line manager for digital products for Panasonic.

Some dealers may value predictability above all – and, if an application does not require particularly high image quality, a good choice for such dealers may be wavelet compression. “Wavelet compression uses a fixed file size so it’s easy to calculate storage needs. If clients need to archive 30 days, you can tell them ‘Here is your exact cost,’” explains John Sisk, western regional sales manager for Mitsubishi.

Most industry experts agree, however, that image quality with wavelet compression is not as good as with MPEG4, which is fast becoming the most popular MPEG format, or with JPEG or JPEG2000. In comparison with its predecessor, JPEG2000 – the latest version of JPEG – offers reduced file sizes without compromising image quality.

For the highest security installations, such as casino gaming tables, dealers often select the MPEG2 format, which offers image quality that is similar to that of a digital video disk (DVD). DVRs supporting MPEG2 are some of the highest priced, however, and storage requirements are higher than for other compression formats.

One additional consideration that may come into play in selecting a compression format is whether the system will be connected to a client’s computer network. If so, the dealer will want to select a compression format that transmits well over the available bandwidth. “MPEG4 will transmit better over a network,” Lott says. “You will get higher refresh rates than with JPEG.”

JPEG and other full-picture compression formats may have a different edge, however. Law enforcement agencies generally have greater success capturing still images when such formats are used, making them the preferred choice for clients who anticipate such a requirement.

Although the wide array of compression standards might appear to be plenty, many manufacturers also offer their own proprietary compression formats. These are often variations on existing standards and usually entail similar design considerations.

Once a dealer has determined which compression ratio to use, along with the number of cameras in a system and how they will be programmed, the dealer should consult with his selected vendor to determine the appropriate hard drive size for an installation. Because it can be difficult to precisely predict the hard-drive storage requirements, however, dealers may want to consider choosing a DVR that supports more than one compression format. Although this adds to the DVR’s cost, it can sometimes save on hard drive or other costs. A DVR that supports both change-only and full-picture compression can sometimes be tweaked to accommodate a specific size hard drive, Heller says.

Key Decision – networking

Security dealers and systems integrators must consider whether they plan to network a DVR in their installations.

“About 80 percent of today’s models are network-able,” estimates Gary Perlin, Speco Technologies vice president of video. “But only about half of them are on a network today.” Apparently many dealers are purchasing network-ready DVRs in anticipation of future needs.

Don Taylor, vice president of marketing for Dedicated Micros, offers advice to dealers who want to network a DVR. “Know the purpose of using the network,” Taylor says. “Will [the user] be retrieving video sporadically or continuously, as an example. Understand any bandwidth restrictions.”

In selecting a network DVR, determine whether the network will require static or dynamic IP addresses. A throttle – to limit the amount of bandwidth a video network can consume – also can be helpful. This is particularly important if images will be transmitted over a relatively low-bandwidth wide area connection and if they will share bandwidth with mission-critical corporate data transmissions.

Key Decision – integration

A main consideration in selecting a DVR is whether the client wants to integrate it with an intrusion detection or access control system. Most DVRs support connection of a motion detector or door contact to any of their cameras.

The downside is that the video and alarm systems still operate as two independent systems, but for many installations, that level of integration may be appropriate. DVRs that support full integration usually do so with only certain models of access control and intrusion detection systems.

The type of operating system that a DVR uses can be particularly important when the DVR will be networked or integrated. Some DVRs have embedded operating systems, usually based on Linux, while others are Windows-based which may be easier to integrate. But companies using embedded operating systems claim their products are more immune to viruses and provide a higher level of security. Matt Sailor, director of sales and marketing for Mace Security’s CCTV Division, notes that embedded operating systems are more likely to offer multiple levels of network access.

The future

Security dealers soon may have even more choices to consider when purchasing a DVR. DVRs that support connectivity of IP as well as analog cameras will be available soon, as will DVRs with internal DVD burners. And for the highest-security applications, some manufacturers are beginning to offer facial recognition, enabling the DVR to notify security personnel when an attached camera sees a face that appears to match one in a database of criminals or undesirables.