Blackmagic Design Decklink IP 100G

The Blackmagic Design DeckLink IP 100G is a PCIe Gen 4 capture and playback card built for SMPTE 2110 IP network systems, supporting up to eight simultaneous UHD video channels via dual 100G Ethernet ports. It enables ultralow latency with GPU Direct RDMA and supports macOS, Windows, and Linux, working seamlessly with software like DaVinci Resolve and Adobe Premiere Pro.

Advanced Design

• Dual 100 Gb/s QSFP28 Ethernet ports enable full redundancy for critical broadcast workflows

• PCIe 8-lane Gen 4 interface fits into both 8- and 16-lane PCIe slots

• 6-pin auxiliary power connector supports stable, high-speed performance

• Integrated active cooling maintains thermal stability under load

Video Standards and Formats

• Handles up to eight simultaneous channels in HD, UHD 4K, and DCI 4K at up to 120 fps

• Compatible with SD (PAL and NTSC), 720p, 1080p, DCI 2K, UHD 4K, and DCI 4K formats

• Supports static HDR metadata with HLG and PQ

• Colour formats: 8-, 10-, and 12-bit RGB 4:4:4; 8- and 10-bit YUV 4:2:2

• Colour spaces include Rec. 601, Rec. 709, and Rec. 2020

Software and OS Compatibility

• Works with DaVinci Resolve, Final Cut Pro, Adobe Premiere Pro, AVID Media Composer, and others

• Fully compatible with macOS, Windows, and Linux

• Bundled with Media Express, DaVinci Resolve, Disk Speed Test, Livekey, and Blackmagic Desktop Video Utility

Broadcast and Network Protocols

• Conforms to SMPTE 2110 standards: 2110-20 (video), 2110-21 (timing), 2110-30 (audio), and 2110-40 (ancillary data)

• SMPTE 2022-7 for stream redundancy across networks

• NMOS IS-04 and IS-05 compliance enables automated discovery and connection management

• Uses IEEE 1588-2008 PTP for frame-accurate time sync

Security and Performance

• Acts as a hardware-level firewall by isolating video to frame buffers

• Prevents unauthorised access and rogue software interaction on managed IP networks

• Ideal for broadcast servers, GPU-accelerated pipelines, and virtual production stages requiring high throughput and minimal latency