When to use: Size the network uplink and switch capacity for an IP camera deployment. Per-camera bitrate scales with resolution, frame rate, and codec. H.265 roughly halves the H.264 bitrate and H.265+ can save 60–70% on largely static scenes. Add overhead for protocol packaging and bitrate spikes during high-motion events.
This calculator determines the total aggregate network bandwidth consumed by an IP camera system, accounting for resolution, frame rate, codec selection, and number of cameras. Network engineers and system designers use it to size switch uplinks and NVR network interfaces.
Per-camera bitrate is a function of resolution, frame rate, and codec. A 1080p camera at 30 fps in H.264 produces approximately 4 Mbps; the same camera in H.265 produces approximately 2 Mbps (50% reduction). At 15 fps the bitrate halves again. Total aggregate bandwidth is the per-camera bitrate multiplied by the number of cameras, with an additional overhead factor for network protocol packaging and bitrate spikes during high-motion events.
H.265 (HEVC) achieves equivalent quality to H.264 at approximately half the bitrate through improved intra/inter-frame prediction and entropy coding. H.265+ (proprietary smart codecs from Hikvision, Dahua, etc.) can achieve 60–70% reduction on largely static scenes by using background modeling to reduce bitrate during periods of inactivity.
The aggregate bandwidth must be carried on the NVR's uplink port and the switch backplane. A 1 GbE link can sustain approximately 700–800 Mbps of practical throughput; systems exceeding this require 10 GbE uplinks.
IEEE 802.1Q VLAN standards are commonly applied to isolate camera traffic from general IT traffic on the same switch infrastructure. ONVIF Profile S and Profile T define the streaming protocols (RTSP/H.264/H.265) used by IP cameras to ensure interoperability between cameras and VMS platforms. NFPA 731 addresses network infrastructure reliability for security systems. Switch selection should meet the switching capacity requirements of the full camera payload without packet loss.
Always specify a network overhead factor of at least 20% above the raw camera aggregate bitrate to account for protocol headers, TCP/UDP overhead, and motion-event bitrate spikes. High-motion scenes (busy intersections, entrances) can temporarily produce 1.5–2× the nominal bitrate in Variable Bit Rate (VBR) mode.
Do not confuse aggregate recording bandwidth with storage bandwidth — the NVR's disk write throughput must also support the total bitrate. Dedicate separate VLANs for camera traffic to prevent camera video from competing with corporate data traffic. Plan switch uplinks and NVR NICs for the full calculated load, not the average.
Select the camera resolution and codec, choose the frame rate, enter the number of cameras, and set the network overhead percentage (20% is typical). The tool shows per-camera bitrate, raw aggregate bandwidth, and the total with overhead. The recommended uplink (1 GbE, 10 GbE, or 25 GbE+) is automatically selected based on the total bandwidth. Use the result to specify the switch uplink ports and NVR network interface card.
A 4K (8MP) camera at 30 fps in H.264 uses approximately 16 Mbps; in H.265 approximately 8 Mbps; in H.265+ as low as 5–6 Mbps on static scenes. At 15 fps, halve those figures. Scene activity and VBR settings significantly affect real-world bitrates.
Yes, most modern NVRs and VMS platforms support both codecs simultaneously. Each camera stream is decoded independently. Calculate the bandwidth contribution of each group separately and sum the totals — this tool supports that workflow through its per-camera bitrate output.
20% is the standard planning overhead for RTP/RTSP streaming over UDP on a dedicated camera VLAN. If cameras share a network with general IT traffic, or if the system uses multicast streaming, increase the overhead to 25–30%.
At 15 fps, 32 cameras × 2 Mbps (1080p H.265) = 64 Mbps raw. With 20% overhead: 77 Mbps total — well within a 1 GbE uplink. At 30 fps and H.264, 32 × 4 Mbps = 128 Mbps × 1.2 = 154 Mbps, still within 1 GbE. 4K systems at 30 fps H.264 with 32 cameras would exceed 1 GbE and require 10 GbE.
Motion detection reduces storage, not live streaming bandwidth. Even when cameras record only on motion, they still stream video continuously to the NVR for motion analysis. Only substream or scheduled recording reduces the amount written to disk.
Try our Physical Security Studio
More calculators, simulators, and guides for this discipline.