CCTV storage calculator for Kuwait's 120-day rule

For a typical 8-camera 4 MP system recording at 15 fps in H.265 with the official 50% motion assumption, 120 days needs about 15.6 TB of usable capacity — around 17 TB of raw disk once file-system and database overhead is allowed for. Four 2 MP cameras need roughly 4.3 TB; 16 cameras at 4 MP need about 34 TB. Camera count, resolution and codec change the answer more than anything else.

Law No. 61 of 2015 requires commercial premises to install and maintain surveillance systems, and Ministerial Decision No. 5003 of 2015 sets the technical specification: all cameras recorded for a minimum of 120 days at maximum resolution and 15 fps, calculated at 50% motion, with UPS backup and authenticated export so footage is admissible as evidence. This calculator uses those exact parameters in MOI mode.

A drive sold as 8 TB formats to less, and the NVR also uses space for its database, indexes and bitrate spikes above the nominal figure. We add 10% headroom on top of the recorded footage so the system reaches day 120 rather than starting to overwrite at day 112.

Yes, and it is the step most often missed. The calculator gives usable capacity. RAID 5 needs one additional disk of the same size for parity, RAID 6 needs two, and RAID 1 or 10 doubles the total. A 32 TB usable requirement on RAID 6 with 8 TB disks means six drives, not four.

If the cameras and recorder both support it, yes — it typically cuts storage by around 45% for the same image quality, which often decides whether an existing NVR can reach 120 days at all. Mixing H.265 cameras into an H.264-only recorder does not help, because the recorder has to handle the stream. Check both ends before assuming the saving.

All three are a manufacturer's own compression layer built on top of standard H.265, not a separate codec — a camera and NVR from the same brand both have to support it for the saving to be real. Uniview's Ultra Code is the most aggressively marketed: Basic claims roughly a further 30% smaller than plain H.265, Advanced (deep-learning-based) roughly 50% smaller. Hikvision markets H.265+ as up to ~75% smaller in an ideal static scene, and Dahua's Smart H.265+ (Smart Codec) makes a similar claim — the calculator uses a more conservative ~35% for both, since a real scene with people and vehicles moving through it never behaves like the vendor's demo footage. Every one of these figures is the vendor's own, not an independent measurement, which is why the calculator flags a note whenever you select one: confirm the exact camera model and NVR firmware actually support it before sizing disks around the saving.

For a corridor, a stockroom, or a camera that only needs to confirm someone passed through — yes. It records at roughly a third the bitrate of a 4 MP camera, which is meaningful across a large count. It is a poor choice anywhere identification matters, such as a till, an entrance, or a gate, where the extra storage cost buys real evidentiary value.

Recording bandwidth is the sum of the camera bitrates: 16 cameras at 4 MP and 15 fps is roughly 48 Mbps continuously into the NVR. That is comfortable on a gigabit switch, but it should sit on its own CCTV VLAN, and remote viewing adds a second stream per viewer on top of it. Large sites approaching several hundred Mbps need the recorder on a 10 G uplink.

Residential villas and private premises are not bound by the commercial specification, so shorter retention is a genuine choice there — switch the calculator to Custom mode for 30 or 60 days. Shops, malls, hotels, banks, restaurants and similar establishments are covered by the requirement, and falling short can bring fines and licensing consequences.

Yes. We size the disks against this formula, supply surveillance-rated drives rather than desktop ones, configure the RAID level and retention policy, and verify at handover that the system actually holds the full period. Sizing on paper and confirming on the running system are two different things, and the second is the one that matters at an inspection.