Embedded vision systems are the keystones to many advanced capabilities in modern applications, including augmented vision, automotive, machine vision, medical, and smart city cameras. Modern vision systems require high resolution, high frame rates, and increasing pixel depth, presenting significant data rate and memory bandwidth challenges. AMD offers both cost-optimized and high-performance MIPI-based solutions for camera sensor capture and display, supporting- D-PHY, C-PHY, CSI-2, and DSI protocols. Additionally, AMD offers a breadth of image signal processing IP for color conversion, correction, balancing, and other operations required by many image sensor applications. The flexibility of AMD devices helps to meet the unique requirements of every design and enables developers to meet those needs as well as differentiate their solutions.
AMD MIPI IP Cores
AMD devices support multiple MIPI configurations, from single to multiple links, different data rates, and data types, and can mix receive and transmit datapaths on a single device. AMD devices are also expandable, limited only by the number of I/Os. C-PHY will require dedicated pinouts to accommodate for three distinct voltages levels and is expandable up to three lanes. To further accelerate or streamline solutions, programmable logic enables integration of the entire image signal processing (ISP) chain while remaining scalable and field upgradable as ISP algorithms improve.
MIPI Support across AMD Products
| Cost-Optimized Portfolio | High-Performance Devices |
|---|---|
|
|
AMD offers MIPI-based solutions for camera sensor capture and display through cost-optimized and high-performance devices that enable edge-type applications across a broad set of markets.
Automotive
The AMD automotive portfolio solves industry challenges within ADAS, automated driving, in-vehicle Infotainment & driver information, and electrification & networking. Our adaptable architecture allows users to achieve low-latency compute and the ability to connect a wide array of interfaces – including MIPI. As enhancements to the MIPI protocol are released in the toolchain, flexible silicon from AMD can adapt – letting users design their systems with the latest interfaces for the fastest data transmission possible. AMD is introducing C-PHY support for reduced I/O count and higher bandwidth MIPI applications.
AMD offers high-performance, low-power consumption MIPI-based solutions for camera sensor capture and display, supporting D-PHY, CSI-2, and DSI protocols.
A broad range of industrial and healthcare IoT applications can leverage MIPI support on our cost-optimized portfolio platforms. These include vision-guided robotics, security cameras, smart cities, smart retail, endoscopy, and machine vision. If you are looking to design with our Spartan™ 7 FPGA or Zynq™ 7000 SoC families, start with these kits.
Spartan 7 SP701 Evaluation Kit
The SP701 evaluation kit, equipped with the best-in-class performance-per-watt Spartan 7 FPGA, is built for designs requiring sensor fusion such as industrial networking, embedded vision, and automotive applications. The SP701 features video interfaces including MIPI CSI, DSI, and HDMI. It also features high I/O availability and I/O expansion capability via Pmods and FMC connectors making it the largest IP development canvas for Spartan 7 FPGA users.
Featuring the Spartan 7 XC7S100 FPGA
| System Logic Cells (K) | 102 |
|---|---|
| Memory | 1,100Kb DRAM, 4,320Kb Block RAM |
| DSP Slices | 160 |
| Maximum I/O Pins | 400 |
| Additional Features | 8 clock management tiles (1 MMCM + 1 PLL) |
| 1 analog to digital converter (XADC) | |
| 1 configuration AES / HMC block |
Zybo Z7: Zynq 7000 ARM/FPGA SoC Development Kit
The Zybo Z7 from Digilent is built around the Zynq 7000 device, integrating dual Arm® Cortex®-A9 processors with 7 series programmable logic. The Zybo Z7 kit offers a video-capable feature set that includes a MIPI CSI-2 compatible Pcam connector, HDMI input and output, high DDR3L bandwidth, and other multimedia and connectivity peripherals. This feature set provides an affordable solution for complex embedded vision applications.
Featuring the Zynq 7000 XCZ7010 SoC or XC7Z020 SoC
| Device | XCZ7010 | XC7Z7020 |
|---|---|---|
| System Logic Cells (K) | 28 | 85 |
| Memory | 1,024 MB DRAM 2.1 Mb block RAM |
1,024 MB DRAM 4.9 Mb block RAM |
| DSP Slices | 80 | 220 |
| Maximum I/O Pins | 100 | 125 |
| Additional Features | 2 clock management tiles (1 MMCM + 1 PLL) | 4 clock management tiles (1 MMCM+1PLL) |
| 1 analog to digital converter (XADC) | 1 analog to digital converter (XADC) | |
| 1 AES and SHA block | 1 AES and SHA block |
For AI or video-focused industrial, medical imaging, endoscopy, healthcare edge AI and more, our high-performance device platforms open up added performance and features. If your application includes a need for video encoding or machine learning, check out our high-performance line based on the Zynq UltraScale+™ MPSoCs and Versal™ adaptive SoCs.
Zynq UltraScale+ MPSoC ZCU104 Evaluation Kit
The ZCU104 evaluation kit enables designers to jumpstart designs for embedded vision applications such as surveillance, advanced driver assisted systems (ADAS), machine vision, augmented reality (AR), drones, and medical imaging. This kit features a Zynq UltraScale+ MPSoC EV device with integrated video codec and supports many common peripherals and interfaces for embedded vision applications. The included ZU7EV device is equipped with a quad-core Arm Cortex-A53 processor, dual-core Cortex-R5F real-time processor, Mali™-400 MP2 graphic processing unit, 4 Kp60-capable H.264/H.265 video codec, and 16nm FinFET+ programmable logic.
Featuring the Zynq UltraScale+ XCZU7EV-2FFVC1156 MPSoC
| System Logic Cells (K) | 504 |
|---|---|
| Memory | 38 Mb |
| DSP Slices | 1,728 |
| Video Codec Unit |
1 |
| Maximum I/O Pins |
464 |
Ultra96-V2 Zynq UltraScale+ ZU3EG
The Ultra96-V2 is an Arm-based development board built around the Zynq UltraScale+ MPSoC. Available in commercial and industrial temperature grades, the Ultra96-V2 has been designed with a certified radio module from Microchip. Additional power control and monitoring is possible with the included Infineon Pmics. Engineers connect to the Ultra96-V2 through a web server using integrated wireless access point capabilities or use the provided Linux Matchbox windows environment, which can be viewed on the integrated Mini DisplayPort™ video output. Multiple application examples and onboard development options are also provided.
Featuring the Zynq UltraScale+ MPSoC ZU3EG A484
| System Logic Cells (K) | 154 |
|---|---|
| Memory | 7.6 Mb |
| DSP Slices | 360 |
| Video Codec Unit |
None |
| Maximum I/O Pins |
252 |
Versal AI Core Series VCK190 Evaluation Kit
The VCK190 kit is the first Versal AI Core series evaluation kit, enabling designers to develop solutions using AI and DSP engines capable of delivering over 100X greater compute performance than today's server-class CPUs.
With a breadth of connectivity options and standardized development flows, the VCK190 kit features the Versal AI Core series VC1902 device, providing the portfolio's highest AI inference and signal processing throughput.
Featuring the Versal VC1902 Adaptive SoC
| AI Engines | 400 |
|---|---|
| DSP Engines | 1,968 |
| System Logic Cells (K) | 1,968 |
| LUTs | 899,840 |
| Application Processing Unit | Dual-core Arm Cortex-A72 |
| Real-Time Processing Unit | Dual-core Arm Cortex-R5F |
| Maximum I/O Pins | 770 |
| Programmable NoC Ports | 28 |
| Integrated Memory Controllers | 4 |
AMD also offers a platform designed specifically to meet the extreme vision and sensor processing requirements of unmanned aerial vehicle (UAV) and robotics development.
Autonomous Control & Robotics
The UAV and Robotics Platform (URP) meets the typical processing requirements encountered when designing UAV and robotics applications. It is a Zynq UltraScale+ MPSoC-based platform, providing a broad spectrum of processing capabilities. The small form factor board integrates sensors-on-board sensors, software, and FPGA logic. With its motor control units, position sensors, obstacle detection interfaces, and communication resources, the URP can be used for many applications.
Key Features and Benefits
What’s Included
| System Logic Cells (K) | 504 |
|---|---|
| Memory | 38 Kb |
| DSP Slices | 1.728 |
| Video Codec Unit | 1 |
| Maximum I/O Pins | 464 |
| Description | Device Support |
|---|---|
| Spartan™ 7 FPGA SP701 Application Example This example showcases a CSI-2 Subsystem IP with a PCAM camera module (Digilent), which is a popular interface used by MIPI CSI camera modules, and a DSI Subsystem IP with a display. These are both controlled by a MicroBlaze™ soft processor and all the IP is integrated in the Vivado™ Design Suite. The D-PHY is implemented on the SP701 Evaluation Kit using passive components. |
AMD Spartan 7 FPGA |
| Zynq UltraScale+ MPSoC ZCU102 and Versal Adaptive SoC VCK190 application example This design demonstrates the use of the MIPI CSI-2 RX (decodes and processes video data) and MIPI DSI TX subsystems on the Zynq™ UltraScale+™ ZCU102 board or Versal™ adaptive SoC VCK190 board. The system receives images captured by the IMX274 image sensor. Processed images are then displayed on either an HDMI monitor or a MIPI DSI display. |
AMD Zynq UltraScale+ MPSoC AMD Versal adaptive SoC |
Join the adaptive SoC and FPGA notification list to receive the latest news and updates.