8x CoaXPress-12 Frame grabber & Image Processing System
- Up to 100 Gb/s bandwidth via 8x CoaXPress-12 links
- Ideal for AI, Recording & Streaming solutions
- Inline Image Processing option including:
- On-FPGA real-time Compression (.JPEG, Lossless, Quality+)
- Real-Time High Dynamic Range (HDR) correction
- Supports 100+ synchronized cameras using the InfiniVision
- Memory: 8GB @ up to 300GB/s sustained throughput
- Full-height, half-length PCB & bracket PCIe card (Gen. 3 x16)
- Dual-slot
Proc1C10N-CXP12: Octo CXP 12 Frame Grabber with Real-Time Image Processing
The Proc1C10N-CXP12 is a high-performance Octo CXP 12 Frame Grabber designed for real-time image acquisition, preprocessing, and compression in demanding multi-camera vision environments. Built on Gidel’s Proc10N FPGA module with Altera Stratix 10 NX FPGA technology, it integrates embedded Tensor blocks and HBM2 memory, delivering 143 INT8 TOPS / FP16 TFLOPS of processing power. As a result, it can process and analyze data from up to 8× CXP-12 links, making it ideal for advanced vision-based AI systems.
Performance and Real-Time Image Processing
The Proc1C10N-CXP12 is available as a plug-and-play CXP-12 Frame Grabber, or, alternatively, as part of a complete solution that integrates image acquisition, modular real-time preprocessing, enhancement, and advanced compression IPs (.JPEG, Lossless, Quality+).
In practice, this means lower latency at the edge; moreover, teams can offload significant work from the host.
Inline enhancement features include:
-
High Dynamic Range (HDR) – Captures superior details in high-contrast lighting conditions.
-
White Balance – Maintains color accuracy across variable lighting conditions.
-
Dynamic Luminance Balance – Preserves consistent brightness under changing illumination.
-
Gamma Correction – Optimizes brightness and contrast for improved clarity.
Real-time compression with Dynamic ROI delivers:
-
Extended recording times without compromising quality.
-
Lower transmission bandwidth for efficient data handling.
-
Accelerated offline compression, reducing storage needs and post-processing time.
InfiniVision Architecture – Multi-Camera Octo CXP 12 Frame Grabber
Gidel’s InfiniVision architecture enables deterministic, synchronized acquisition from 100+ CXP-12 cameras. As a result, it manages bandwidth, connectivity, and scalability across large systems.
Meanwhile, a PCIe Gen3 x16 host interface ensures ultra-fast data transfer, and 8 × CXP-12 links provide up to 96 Gb/s aggregated bandwidth. In addition, on-board HBM2 and DDR4 memory sustain throughput even under extreme loads.
Flexible Operating Modes
The Proc1C10N-CXP12 supports two operating modes selectable via firmware:
-
InfiniVision: Designed for synchronized multi-camera setups, combining all incoming data—even across multiple cards—into a single unified buffer with dynamic resolutions and formats.
-
ProcFG: Tailored for precision and line-scan applications, offering fixed frame sizes, pixel formats, and ROI grabbing (uncompressed).
As a result, the Proc1C10N-CXP12 adapts to diverse workflows—from complex multi-camera systems to single-camera applications requiring consistent, high-performance acquisition.
SDK and Development Tools for
The Proc1C10N-CXP12 is supported by Gidel’s SDK, featuring intuitive GUIs and APIs for easy integration. Moreover, the ProcVision Suite adds advanced FPGA programming, debugging, and validation tools, enabling rapid customization of data pipelines, real-time processing, and compression workflows—so teams can deploy optimized, application-specific solutions faster and with reduced risk.
Why Choose the Proc1C10N-CXP12? (Octo CXP 12 Frame Grabber)
-
Supports up to 8 × CXP-12 links for high-bandwidth acquisition.
-
FPGA-based real-time processing with HDR, white balance, and gamma correction.
-
Dynamic ROI compression for storage and bandwidth efficiency.
-
Flexible operating modes to suit synchronized multi-camera and deterministic capture.
-
Advanced SDK + ProcVision Suite for rapid development and integration.
Looking for a compact Quad CXP-12 Frame Grabber? See HawkEye-CXP12.
Need an Octo CoaXPress-6 Frame Grabber, refer to Proc10A-CXP6.
For a Jetson-based Quad CoaXPress-12 solution, visit FantoVision40-CXP12 Edge Computer Vision System.
For a Jetson with Quad CXP12 & 10GigE Vision Interfaces, see FantoVision40 Edge Computer Vision System.
General
- 8x CoaXPress-12 links
- 4x CoaXPress-12 links
- PoCXP
- 1.25 Gb/s | CXP-1
- 2.5 Gb/s | CXP-2
- 3.125 Gb/s | CXP-3
- 5 Gb/s | CXP-5
- 6.25 Gb/s | CXP-6
- 10 Gb/s | CXP-10
- 12.5 Gb/s | CXP-12
- 20.83 Mbps (CXP-1 to CXP-6)
- 41.66 Mbps (CXP-10, CXP-12)
- Gidel PHS connector for daughterboards
- JTAG
- On-the -fly selective ROI acquisition
- Link Aggregation mode
- Chunk Data
- Mono, Bayer, RGBA (8, 10, 12, 14 and 16 bits/color)
- RGB (8, 10 and 12 bits/color)
- Horizontal: 16 K pixels (64-bit)
- Vertical: 65 K lines
- Area
- Line Scan
Environmental conditions
- Continuous Operation: 10 - 80% (non-condensing)
- Peak Operation: 10 - 90% (non-condensing)
The Gidel Proc1C carrier board uses only one of the three Proc10N connectors. The two additional connectors available on the Proc10N board and can be used, per user design, for additional 400 Gb/s bandwidth, RDIMM interface and many additional I/Os.
The Proc1C10N-CXP12 Frame Grabber is a highly modular solution, designed to be tailored to meet unique application requirements.
Looking to adapt the Proc1C10N-CXP12 to match your vision?
The Proc1C10N GigE frame grabber offers two powerful customization paths to meet specific application requirements:
-
Modular Pre-Configured Features
Gidel provides a range of pre-embedded modules tailored to your specifications—such as real-time compression, HDR, and other advanced features. (See the Options tab for available configurations.) -
User-Level FPGA Customization
Leverage Gidel’s development tools and IP libraries to integrate your proprietary FPGA logic and extend the system’s capabilities for acquisition, image processing, and control.
The Gidel ProcVision Suite delivers a complete toolchain for advanced user-level FPGA customization of the data flow, image pipeline, image processing, and more—ensuring optimal performance for your vision or imaging application.
Proc1C10N FPGA Resources
| Model | Proc1C10N-CXP12 |
|---|---|
| FPGA | Stratix 10 NX (NX2100) |
| ALM | 702,720 |
| FPGA-MLAB | 11Mb |
| FPGA-M20K | 6,847 |
| AI Tensor Blocks | 3,960 AI Tensor Blocks; peak 143 INT8 TOPS / 286 INT4 TOPS |
| Peripheral Memory | |
|
|
94.5 Mb @ Max throughput of 90 GB/s |
|
|
Up to 16GB @ up to 512GB/s Sustain Throughput |
|
|
Up to 128 GB with ECC |
|
|
15 GB/s |
Grabbers SDK
Application Interfaces
- InfiniVision
- ProcFG
- CameraConfig – Camera discovery and configuration
- ggvcon – GigE Vision network configuration
- InfiniVision with supporting examples
- ProcFG with supporting examples
- Gen<i>Cam GenTL producer libraries compatible with C/C++ compilers
- InitCam for developing user Gen<i>Cam camera configuration application
- GigE for developing camera network communication applications
Software Compatibility
- MVTec Halcon machine vision software
- Camera control Gen<i>Cam based application
- Windows 11
- Windows 10
- Windows Server 2022
- Windows Server 2019
- Windows Server 2016
- Linux (kernel 2.6.x- 6.12)
| Name | Description | Type | Size | |
|---|---|---|---|---|
|
Proc1C10N-CXP12
|
Datasheet | 290.4 KB | ||
|
Proc1C10N-CXP12
|
Block Diagram | PNG | 68.2 KB | |
|
Proc10N: Datasheet
|
FPGA module | 194.4 KB | ||
|
Proc10N: Block Diagram
|
FPGA module | JPG | 73.9 KB | |
|
Proc1C: Datasheet
|
Carrier Board | 133.5 KB | ||
|
Proc1C: Block Diagram
|
Carrier Board | JPG | 57.1 KB |
FAQ
The Proc1C10N-CXP12 is an ultra-high-performance Octo CXP card designed for extreme acquisition and AI tasks. It supports:
-
Up to 8 × CXP-12 links, allowing for simultaneous acquisition from 1 to 8 cameras depending on each camera’s link requirements.
-
Flexible configurations for 4 × CXP-12 links.
-
Full support for PoCXP (Power over CoaXPress) for simplified, single-cable connectivity.
The board is powered by the Altera FPGA, featuring embedded Tensor blocks and high-speed HBM2 memory. By offloading both the massive data management and AI processing to the hardware, the system ensures low latency and consistent throughput while keeping the host CPU utilization negligible, even under extreme data loads.
Yes. One of the core strengths of the Proc1C10N-CXP12 is the option for modular inline FPGA pre-processing using Gidel’s off-the-shelf IPs. You can immediately implement real-time enhancements—such as High Dynamic Range (HDR), Real-Time Compression (Lossless, JPEG, or Quality+), Dynamic Luminance Balance, and beyond. By performing these tasks on the FPGA, you process the data before it even reaches the host memory.
Virtually zero. Because the acquisition path, synchronization, and modular ISP/compression tasks are performed entirely on the FPGA, the host CPU is freed from the heavy lifting of raw data handling. This enables you to run high-level AI inference and complex application logic on the host without any performance bottlenecks or dropped frames.
The Proc1C10N-CXP12 is supported by the ProcVision™ Suite, a modular Vision SDK for FPGA development. It provides a complete ecosystem for building custom imaging systems, allowing you to insert proprietary algorithms and AI models directly into the FPGA pipeline. This environment includes dedicated tools for real-time validation and debugging, significantly reducing development risk.
By utilizing Gidel’s InfiniVision™ architecture, the board provides a flexible infrastructure for deterministic, synchronized acquisition from 100+ CXP-12 cameras. Combined with the massive bandwidth of HBM2 memory and a PCIe Gen3 x16 interface, it guarantees zero-frame-loss performance in the most demanding AI-driven vision environments.
Yes. The Proc1C10N-CXP12 is ideal for raw data and lossless recording as well as full image processing and encoding. It can capture wide dynamic range 12-bit data and perform all necessary on-the-fly preprocessing—including high-quality deBayer, HDR, and JPEG compression. To further reduce host load and support real-time control, the board can simultaneously provide thumbnails and ROI streams in parallel.
All Gidel products come with a 12-month warranty, dedicated technical support, and regular firmware and driver updates to ensure high-performance reliability in demanding, long-term deployments.