What is the difference between a DVR and an IP video server NVR. DVR or NVR - which DVR to choose? A good NVR now should have features such as

24.07.2006

Shawn Ciccarelli
Source: Hi-Tech Security Solutions

In the article, Shawn Ciccarelli, Network Product Manager at Reditron, articulates the difference between digital video recorders (DVRs) and network video recorders (NVRs) and what these differences mean to system designers and end users.

First came the TV camera and video monitor, soon followed by the video cassette recorder (VCR), which recorded one video stream onto a three-hour tape at 25 frames per second. Often, it could only be turned on with the help of an external device (for example, a packet switch).

Then the technology brought a video multiplexer onto the stage, which allowed recording several video streams at once on the same film, and dividing them into several separate images for viewing; a VCR appeared with the ability to record with the passage of certain periods of time (time-lapse VCR), thus allowing the use of three-hour cassettes to record more cameras, albeit at the expense of partial loss of information from them.

Digital Video Recorder (DVR)

The DVR is a convenient, albeit limited, replacement for the multiplexer-VCR pair and provides non-linear access to recorded material, usually selected by camera ID, time, and date. The level of quality of the recorded video signal is generally higher than that achievable in analog tape VCRs, but may be lower depending on the compression algorithm and the specific system configuration.

The main advantage of DVR is that more programmable parameters are available for each of the video streams (resolution, frame rate, options for turning on additional devices, recording on / off time, etc.), but DVR is useful only where it is possible to directly connect to it all available analog cameras.

Good modern DVRs have built-in UDP (CAT 5) network ports - thus, the device can be assigned an IP address and accessed via an Ethernet network.

However, there are many limitations here, not the least of which is that in the event of a device crash, you will most likely lose all your videos (or, for starters, they may not be made at all). Network video recorders (NVR) do not have this drawback, since they can be used in the "mirror" mode, but I will talk about this a little later.

In terms of reliability, if you're going to use a DVR -- make sure the model you choose to implement has an industry standard hard drive, not one specially designed by the manufacturer -- otherwise you'll fail much sooner than you think (most DVR crashes are caused by overload and overheating of hard drives). Ask the manufacturer what discs they use.

Like most things in this world, a DVR's actual performance, ease of use, and reliability depend on who made it, the features of the particular model, and how much you paid for it.

Network Video Recorder (NVR)

NVR heralded a new milestone in the development of video recording technology.

It is important to understand the difference between DVR and NVR as both types of DVRs are often referred to as "digital". Since the DVR digitally compresses the video signals it receives and records them on the hard drive, the term "digital" here refers to the compression and recording technologies, not to the video images being transmitted. Therefore, the DVR should be placed close to the cables carrying the analog video signal. An NVR, on the other hand, records digital images directly from an IP network. Thus, the most obvious difference between DVR and NVR is that while DVR records information from analog sources - cameras, NVR records video streams already formed at the camera level. That is, nowhere on the NVR you will find ports for connecting a video signal: its inputs and outputs contain digital IP data consisting of a compressed and encoded video signal. This signal is in most cases presented in MPEG-4 format. MPEG-4 compression technology has gained wide acceptance in the video surveillance industry, mainly in terms of efficiency.

A significant advantage of the NVR-based architecture is that NVRs can be located anywhere on the network - in the monitoring center near clusters of cameras, distributed at the ends of the network, assembled together in a single unit - anywhere, really. During use, their location is indifferent to the operator - he or she simply calls the desired video stream and, with the appropriate viewing rights, receives it on their screens.

NVRs record and playback simultaneously, and recordings stored on any of them can be remotely viewed simultaneously by multiple authorized operators distributed over the network, completely independently and without affecting each other.

The importance of independence from physical location, if necessary at a great distance from the cameras, should not be underestimated - system administrators are known to zealously observe load limits on their networks, however, by correctly calculating the distribution of data flows within the network and strategically placing NVRs , we can minimize the impact of video streams passing through the network on the load of network resources.

In the general case, the NVR can be placed on a local area network (LAN) in close proximity (in the network sense, that is, not necessarily physically close) to the accumulation of cameras - thus, the main load falls on the local network, which can easily cope with such a load, thereby freeing up the resources of other parts of the network, perhaps more limited in capabilities. The system administrator can designate how much network bandwidth he is willing to devote to video data, and this will become a limit value that cannot be exceeded even in the case of the highest workload on the video components of the system.

Further, if a video recording is requested at some other point in the network (usually in the control center, but not at all necessary), the desired sequence can be easily called up by the operator, exported to his workplace and analyzed, viewed (which is not the same ) and appropriate actions can be taken in response to the information it contains.

Spreadsheet-based calculator applications are used to facilitate the calculation of required video data rates and disk storage capacities. Applications allow you to perform calculations for each of the cameras individually using parameters such as: type of observation location (busy street / inner corridor, etc.), camera functional parameters (constant panning, zooms in and out under the control of the operator, or a fixed camera for identification of personnel, etc.), video image resolution and refresh rate in frames per second, and when using video motion detectors, the frequency of movements in the frame and their nature.

A good NVR NVR should now have features such as:

hot swappable drives
support for Network Management Protocol (SNMP),
built-in diagnostics (so beloved by system administrators),
protection of files from deletion (accidental or intentional),
built-in software firewall to protect data from unauthorized access,
file export function, including applying electronic watermarking, digital signature at the frame level, as well as a network event registration file - to protect the recording from falsification,
synchronous recording and playback of video and sound,
monitoring the temperature of hard drives,
dual, fully interchangeable power supplies and network connections - the latter must ensure continuous long-term operation in the event of a failure of one of the power supplies or the network.

This technique is especially useful when consolidating multiple independent systems into a single managed space, or when streamlining or expanding a system—because it simplifies the system and reduces the cost of everything related to laying new cable lines.

Also, to reduce the need for disk space (and, accordingly, costs), the function of setting the frame rate depending on the activity in the surveillance area (ACF) is used.

This function is fundamentally based on the processing of a video signal when it is encoded by a television camera. When the camera is running, if there are no objects moving in its field of view, the video recorder switches to a low frame rate mode (usually 1 frame per second). And when motion is detected, the frame rate is automatically increased to a predetermined value, and this usually happens in as little as 100 microseconds (1/10 of a second). This function is most effective in places where there is no high activity, for example, in corridors and fire exits, or inside buildings that are empty at night; this saves up to 50 percent of the storage capacity that would have been used in the absence of ACF.

So what can we expect in the future?

The analysis software then searches for the requested events on its own, allowing the operator to focus on more specialized and urgent tasks. And these are not just user-friendliness features: they also help reduce overall network congestion.

This is just the tip of the iceberg - new developments include, for example, crowd detection (too many people in a small area), motion detection (a person or vehicle that moves, say, from left to right in the frame), abandoned object detection (suitcase left at the airport terminal), moving against the general flow (a person walking back along the customs corridor), detecting objects by shape signs (vehicles), tracking objects and detecting thefts (disappearance of an object from view).

It can be expected that the use of analytical programs will cause a significant increase in system performance when searching for recorded video in order to analyze events that have already occurred. And here you can not do without a network video recorder.

Currently, there are many types of video surveillance systems: from outdated analog to ultra-modern IP video cameras.

The most common are analog video cameras and modern digital systems for processing, recording and storing video images. Such digital systems are DVRs and NVR IP video servers.

Network video recorder (NVR) buy in China, in Urumqi
Digital Video Recorder (DVR) buy in China, in Urumqi

So what is the difference between a video recorder and a video server? What can be guided in choosing a particular video surveillance system?

The operation of a network video server and a DVR are very similar. Both devices receive and process the signal from analog video cameras for subsequent playback, recording and storage in digital format. The main difference between DVR and NVR lies in the very method of signal processing and transmission, as well as in their constructive solution.

The DVR is a self-contained minicomputer with its own hard disk, processor and ADC. It receives an analog signal and processes (compresses) it itself, and then displays it on a monitor or sends it to an archive. The NVR is based on the operating system (Windows, Linux). The analog signal enters it already in digital and compressed form (signal processing and compression occurs through a special input/output board or immediately from an IP video camera).

If digital video recorders are connected to analog cameras with a rather expensive coaxial cable, network video recorders can conduct a signal through a regular Ethernet cable.

In addition, the difference between a video recorder and a video server is in their functional characteristics. DVR, despite the presence of network and other functions, primarily performs video surveillance, that is, displaying video information, transmitting it over a network and recording it in an archive.

The video server has access to intelligent functions, such as integration (the ability to combine with other video surveillance systems and fire alarm systems into a single integrated network to manage it from one point), face recognition, car number recognition, registration of cash transactions, etc.

Another difference between DVR and NVR is the number of cameras in the system. Digital video servers allow you to combine a limited number of video cameras in one centralized point, but if we talk about combining from 100 to 1000 video cameras, then only a network IP video server can do this.

But the level of reliability due to a limited set of functionality and the lack of a Windows operating system in the DVR is undeniably higher.

At present, since there are DVR, NVR and HVR, some customers feel confused about how to choose the right one. Today we will make it easier for you.

As the names of all these devices imply, they are all designed to record video from security cameras. The main difference between the two is what types of cameras they are designed to be fun to play with.

Digital Video Recorder (DVR)
The DVR is designed to work with standard analog security camera systems like , HD TVI cameras and HD CVI cameras. This does not mean that every DVR works with all of these cameras. These DVR categories are simply a line with the previously mentioned types of security cameras. Standard HD AHD cameras are designed to be paired with standard AHD DVRs just as HD-TVI and HD-CVI cameras are designed to be paired with the appropriate TVI DVR and CVI DVR.

The NVR is paired with an IP network security camera. There are two different types of network video recorder, but don't worry, they are both still used with IP cameras. The first type, and much more common, requires the IP camera to be connected to a router or switch. This would mean using the search feature on the NVR to "ping" the cameras and then adding them to the NVRs device list to start browsing. While this is not a very difficult step, it may be one you don't want to take.
In this case, the second option is to go with an NVR that is built in network or PoE (power over Ethernet) ports. This will allow you to connect your IP camera directly to the back of your NVR just like you would with a DVR. This will remove the step from having to manually add them to the device list, and it will remove all the worries of ensuring that they are properly connected to the network.

The hybrid recorder is rapidly growing in popularity due to its versatility. Hybrid Video Recorders (HVR) are compatible with standard analog signal and digital signal. Here you need to change the channel types as user needs, namely what we mean is that 8-channel HVR can be 4-channel analog signal and 4-channel digital signal, can supply 4pcs AHD / TVI / Analog / CVI cameras and 4pcs IP cameras.

Now it's not just that you get a lot for what you pay for that makes them so attractive to consumers. Since here the HVR can be connected perfectly acceptable AHD / CVI / TVI / analogue channels also to IP cameras.

It will only be a matter of replacing the cameras instead of the DVR and all the cables. Depending on the size of your system, this can save you thousands of dollars in upgrade costs.

So to sum it up like MVTEAM

It is important to understand the difference between DVR and NVR as both types of DVRs are often referred to as "digital". The most obvious difference between DVR and NVR is that while DVR records information from analog sources - cameras, NVR records video streams already formed at the camera level, at its inputs and outputs there are digital IP data, consisting of compressed and encoded video.

In the article, Shawn Ciccarelli, network product manager at Reditron, articulates the difference between digital video recorders (DVRs) and network video recorders (NVRs) and what these differences mean to system designers and end users.

First came the TV camera and video monitor, soon followed by the video cassette recorder (VCR), which recorded one video stream onto a three-hour tape at 25 frames per second. Often, it could only be turned on with the help of an external device (for example, a packet switch).
Then the technology brought a video multiplexer onto the stage, which allowed recording several video streams at once on the same film, and dividing them into several separate images for viewing; a VCR appeared with the ability to record with the passage of certain periods of time (time-lapse VCR), thus allowing the use of three-hour cassettes to record more cameras, albeit at the expense of partial loss of information from them.

Digital Video Recorder (DVR)

The rapid development of video image compression algorithms (JPEG, M-JPEG, Wavelet, MPEG-4, etc.), the growth in the speed of data processing by computers and the rapid drop in the cost of storing a unit of information gave impetus to the development of digital video recorders. The latter can be imagined as a combination of the functional similarity of a multiplexer with a computer disk for recording instead of a tape, placed in one case, which has several additional ports for connections.
The DVR is a convenient, albeit limited, replacement for the multiplexer-VCR pair and provides non-linear access to recorded material, usually selected by camera ID, time, and date. The level of quality of the recorded video signal is generally higher than that achievable in analog tape VCRs, but may be lower depending on the compression algorithm and the specific system configuration.
The main advantage of DVR is that more programmable parameters are available for each of the video streams (resolution, frame rate, options for turning on additional devices, recording on / off time, etc.), but DVR is useful only where it is possible to directly connect to it all available analog cameras.
Good modern DVRs have built-in UDP (CAT 5) network ports - thus, the device can be assigned an IP address and accessed via an Ethernet network.
However, there are many limitations here, not the least of which is that in the event of a device crash, you will most likely lose all your videos (or, for starters, they may not be made at all). Network video recorders (NVR) do not have this drawback, since they can be used in the "mirror" mode, but I will talk about this a little later.
In terms of reliability, if you're going to use a DVR -- make sure the model you choose to implement has an industry standard hard drive, not one specially designed by the manufacturer -- otherwise you'll fail much sooner than you think (most DVR crashes are caused by overload and overheating of hard drives). Ask the manufacturer what discs they use.

Like most things in this world, a DVR's actual performance, ease of use, and reliability depend on who made it, the features of the particular model, and how much you paid for it.

Network Video Recorder (NVR)

NVR his arrival heralded a new milestone in the development of video recording technology.
It is important to understand the difference between DVR and NVR as both types of DVRs are often referred to as "digital". Because the DVR digitally compresses the video signals it receives and records them on the hard drive, the term "digital" here refers to the compression and recording technologies, not to the video images being transmitted. Therefore, the DVR should be placed close to the cables carrying the analog video signal. An NVR, on the other hand, records digital images directly from an IP network.

Thus, the most obvious difference between DVR and NVR is that while DVR records information from analog sources - cameras, NVR records video streams already formed at the camera level.

That is, nowhere on the NVR you will find ports for connecting a video signal: its inputs and outputs contain digital IP data consisting of a compressed and encoded video signal. This signal is in most cases presented in MPEG-4 format. MPEG-4 compression technology has gained wide acceptance in the video surveillance industry, mainly in terms of efficiency.
A significant advantage of the NVR-based architecture is that NVRs can be located anywhere on the network - in the monitoring center near clusters of cameras, distributed at the ends of the network, assembled together in a single unit - anywhere, really. During use, their location is indifferent to the operator - he or she simply calls the desired video stream and, with the appropriate viewing rights, receives it on their screens.
NVRs record and playback simultaneously, and recordings stored on any of them can be remotely viewed simultaneously by multiple authorized operators distributed over the network, completely independently and without affecting each other.
The importance of independence from physical location, if necessary at a great distance from the cameras, should not be underestimated - system administrators are known to zealously observe load limits on their networks, however, by correctly calculating the distribution of data flows within the network and strategically placing NVRs , we can minimize the impact of video streams passing through the network on the load of network resources.
In the general case, the NVR can be placed on a local area network (LAN) in close proximity (in the network sense, that is, not necessarily physically close) to the accumulation of cameras - thus, the main load falls on the local network, which can easily cope with such a load, thereby freeing up the resources of other parts of the network, perhaps more limited in capabilities. The system administrator can designate how much network bandwidth he is willing to devote to video data, and this will become a limit value that cannot be exceeded even in the case of the highest workload on the video components of the system.
Further, if a video recording is requested at some other point in the network (usually in the control center, but not at all necessary), the desired sequence can be easily called up by the operator, exported to his workplace and analyzed, viewed (which is not the same ) and appropriate actions can be taken in response to the information it contains.
Spreadsheet-based calculator applications are used to facilitate the calculation of required video data rates and disk storage capacities. Applications allow you to perform calculations for each of the cameras individually using parameters such as: type of observation location (busy street / inner corridor, etc.), camera functional parameters (constant panning, zooms in and out under the control of the operator, or a fixed camera for identification of personnel, etc.), video image resolution and refresh rate in frames per second, and when using video motion detectors, the frequency of movements in the frame and their nature.

A good NVR NVR should now have features such as:

* disks with a possibility of "hot" replacement,
* support for Network Administration Protocol (SNMP),
* built-in diagnostics (so loved by system administrators),
* protection of files from deletion (accidental or intentional),
* built-in software firewall to protect data from unauthorized access,
* file export function, including applying electronic watermarking, digital signature at the frame level, as well as a network event registration file - to protect the recording from falsification,
* synchronous recording and playback of video and sound,
* monitoring the temperature of hard drives,
* dual, fully interchangeable power supplies and network connections - the latter must ensure continuous long-term operation in the event of a failure of one of the power supplies or the network.

The technique of "mirror servers" is now often used to duplicate the recording of video streams to additional NVRs located in different parts of the network, which gives a high level of protection from network failures; if one of the parts fails, the other one takes over as a backup. You can place as many NVRs as you want in your system: adding another one is just a matter of connection and configuration. No additional cabling is required.
This technique is especially useful when consolidating multiple independent systems into a single managed space, or when streamlining or expanding a system—because it simplifies the system and reduces the cost of everything related to laying new cable lines.
Also, to reduce the need for disk space (and, accordingly, costs), the function of setting the frame rate depending on the activity in the surveillance area (ACF) is used.
This function is fundamentally based on the processing of a video signal when it is encoded by a television camera. When the camera is running, if there are no objects moving in its field of view, the video recorder switches to a low frame rate mode (usually 1 frame per second). And when motion is detected, the frame rate is automatically increased to a predetermined value, and this usually happens in as little as 100 microseconds (1/10 of a second). This function is most effective in places where there is no high activity, for example, in corridors and fire exits, or inside buildings that are empty at night; this saves up to 50 percent of the storage capacity that would have been used in the absence of ACF.

So what can we expect in the future?

Many tools are already available to help the operator identify and display the desired events from the video stream. For example, the software that controls the video surveillance system and alarms analyzes the movement in the frame and, at the operator's command, displays a set of thumbnail images (icons) of scenes containing a similar movement of objects on the screen. By clicking on the icon, you thereby cause the playback of the corresponding video fragment. The system is able to analyze 24 hours of video footage and display a set of icons in just a few seconds. Changing the search parameters allows the operator to "comb" huge amounts of recorded material quickly and efficiently.
The analysis software then searches for the requested events on its own, allowing the operator to focus on more specialized and urgent tasks. And these are not just user-friendliness features: they also help reduce overall network congestion.
This is just the tip of the iceberg - new developments include, for example, crowd detection (too many people in a small area), motion detection (a person or vehicle that moves, say, from left to right in the frame), abandoned object detection (suitcase left at the airport terminal), moving against the general flow (a person walking back along the customs corridor), detecting objects by shape signs (vehicles), tracking objects and detecting thefts (disappearance of an object from view).
It can be expected that the use of analytical programs will cause a significant increase in system performance when searching for recorded video in order to analyze events that have already occurred. And here you can not do without a network video recorder.