How IP Security Cameras Capture and Store Video
A business installs cameras, sees the live feed working, and assumes the job is done. Months later something happens in the parking lot, someone goes to pull the footage, and discovers the system only kept the last four days, or the image is too low-resolution to read a license plate, or the recording stopped when the internet went down. The cameras were never really the question. How they capture, where they store, and how long they keep what they record is the part that determines whether a surveillance system actually does its job when it finally matters.
Modern security cameras are computers that happen to see. Understanding the path a single frame of video takes, from the lens to the moment someone reviews it weeks later, is what lets a business size a system correctly instead of discovering its limits during an incident. What follows is that path, and the few decisions along it that matter most.
From Lens to Recording: How an IP Camera Works #
An IP camera, the standard for modern business surveillance, is a self-contained device on the network with its own address, much like a computer. It does four things in sequence, continuously.
First it captures: the lens and sensor turn light into a digital image, many times per second. Then it encodes: a built-in processor compresses that video using a codec, commonly H.264 or the more efficient H.265, because raw video would be far too large to store or send. Then it transmits: the compressed stream travels over the network, usually an Ethernet cable, to a recorder or storage. Finally it is stored, waiting to be viewed live or pulled up later.
This is the fundamental shift from older analog cameras. An analog camera just sends a raw video signal to a recorder that does all the work. An IP camera is intelligent on its own, digitizing and compressing before anything leaves the camera, which is what allows the much higher image quality and the network features businesses now expect.
Where the Video Goes: The Recorder and Storage #
The captured video has to live somewhere, and for most business systems that somewhere is a Network Video Recorder, or NVR. Each camera streams its compressed video to the NVR, which records it to hard drives and manages playback.
One feature of this design matters more than businesses expect: a properly built system records locally to the NVR, not over the internet. The internet connection is used for remote viewing from a phone or laptop, not for the recording itself. That means if the internet goes down, the cameras keep recording to the NVR the entire time. A business that assumes its footage lives “in the cloud” and depends on a connection is making an assumption worth checking, because the most reliable systems capture locally first.
Storage can also live in other places depending on the system: cloud platforms, network-attached storage, or even a card inside the camera for smaller setups. But for a multi-camera business system, the local NVR remains the dependable backbone.
The Decision That Trips Businesses Up: Resolution Versus Retention #
Here is the tradeoff that catches people. Higher resolution captures more detail, the difference between footage that shows “a person was here” and footage that shows a readable license plate or an identifiable face. Modern IP cameras range widely, from around 2 megapixels up to 4K and beyond.
But resolution and retention pull against each other through storage. Higher-resolution video takes more space, which means a fixed amount of storage holds fewer days of it. The length of time a system keeps footage before recording over it, its retention period, is governed by camera count, resolution, frame rate, and how much motion the scene contains. Many businesses target somewhere between one week and three months of retention depending on their needs, and regulated industries may have specific requirements.
Setting every camera to maximum resolution is the common mistake here. A better approach uses high resolution where detail is critical, entrances, points of sale, anywhere a face or plate must be identifiable, and lower resolution for general coverage where simply seeing activity is enough. Matching resolution to purpose, camera by camera, keeps both storage cost and retention under control.
Why Many Systems Use Power over Ethernet #
A practical note on installation: many business cameras are powered over the same cable that carries their data, using Power over Ethernet, or PoE. A single Ethernet cable runs from the recorder or a network switch to the camera, supplying both the network connection and the electricity. This removes the need for a separate power outlet at every camera location, which is why it has become the standard for wired business installations. It is an installation convenience rather than a feature of the footage itself, but it meaningfully simplifies where cameras can be placed.
Frequently Asked Questions #
What is the difference between an IP camera and an analog camera?
An analog camera sends a raw video signal to a recorder that processes it, while an IP camera is an intelligent networked device that digitizes and compresses video itself before sending it over the network. The IP approach delivers much higher resolution, supports remote access and analytics, and is the standard for modern business systems.
Will my cameras still record if the internet goes out?
With a properly designed system, yes. A Network Video Recorder records footage locally to its own drives, and the internet connection is used only for remote viewing, not for recording. This is why local recording is considered more reliable than systems that depend entirely on a cloud connection to capture footage.
How long will a system keep my footage?
That depends on storage capacity, camera count, resolution, frame rate, and scene activity. Higher resolution and more cameras consume storage faster, shortening how many days are retained. Many businesses aim for one week to three months, and the retention period can be planned by sizing storage to the desired number of days.