IP Video Tips

How CPU's and processor speeds affect system design

• By Jeff Stout
• Jun 01, 2012

All NVR and hybrid recorders are not created equally. Every manufacturer has different CPU and throughput speed specifications. These two factors are key in determining the types and numbers of cameras that can be connected to the recording device.

If your customer requires higher megapixel or HD-resolution cameras, the CPU and throughput speed will limit the number of cameras and/or the frame rates. In turn, a customer requiring a greater number of cameras for an installation will sacrifice resolution and frame rates due to the processor and speed of the device. Check with the manufacturer or contact your local Tri- Ed/Northern Video representative to make sure that your system is capable of meeting the requirements of your installation.

Video Codecs - H.264 vs. M-JPEG

The H.264 codec for compressing and decompressing video comes in several varieties. Blu-ray and broadcast HDTV use the High Profile standard (Level 4.1 of H.264) and can deliver full 1080p resolution. This standard of H.264 is CPU and storage intensive and is rarely used in the IP video surveillance space for this reason.

The Baseline Profile of H.264 is less CPU and storage intensive, but cannot deliver the high-quality video that the High Profile standard delivers. The Baseline Profile is the most commonly used standard in IP video surveillance systems. M-JPEG offers high-quality video but less processing power, and it requires more storage space. With the cost of storage decreasing steadily, M-JPEG is a viable standard for IP video when quality is the most important feature required by the customer.

It is critical to determine the proper codec required of the system design so that the end user ultimately receives the video quality and system performance desired.

Running a Parallel Nework for Better Perform ance

One way to maximize the success of an IP video network is to run a parallel network dedicated to security. There are many reasons to do this, but some of the most important are:

• a greater level of security when using your own network as opposed to a customer’s network and all of its vulnerabilities;
• easier network troubleshooting—should issues arise if you do not have to worry about devices on your customer’s existing network;
• ]parallel network allows for the use of all of the available bandwidth on your switches without interfering with existing network operation; and
• more IT departments accepting a parallel network because it will not decrease the performance of an existing network and can be easily detached in the event of network issues.

Parallel networks are a “best practice” when implementing an IP-based video system.

How to Capture a License Plate

How many of us have been on a job where the customer is convinced that he or she absolutely must see license plates? Well, here is a good guideline to follow to ensure success on your job.

To capture license plates, users will need 5 pixels per inch or 60 pixels per foot. If users have a driveway that measures 20 feet horizontally, they need a camera that has 1,200 horizontal pixels to capture license plates. A single 1.3-megapixel camera is more than capable of providing that type of resolution.

This article originally appeared in the June 2012 issue of Security Today.