Products

Versal PL NVMe SSD Speed Test

Using Versal AI Edge Series VEK280

Posted on December 2, 2024 | Jeff Johnson

Hands down, the best solution for non-volatile storage on Versal designs, is NVMe solid-state drives. One of the main reasons why they are such a good fit is because they interface with the host device over PCI Express. Versal Prime, AI Core and AI Edge devices all have integrated PCIe blocks, and most of them are PCIe Gen5 compliant. This means that there are no extra IP costs to connect an NVMe SSD to the Versal devices because everything you need is already built into the device. [Read More]

Quick look at Ethernet FMC Max

Opsero's new gigabit Ethernet FMC with SGMII

Posted on October 9, 2024 | Jeff Johnson

Ten years ago, almost to the day, I launched the Opsero Ethernet FMC. To my delight and surprise, many of our customers are still buying them today. Gigabit Ethernet still finds many uses in FPGA designs today, because it is simple to setup, the cabling is easy to buy and install, the throughput is ample for many applications and it is extremely versatile. This year, I felt that it was time to upgrade our Ethernet FMC offering, so we’ve launched two new products: Quad SFP28 FMC which I have introduced to you in earlier posts (here and here), and Ethernet FMC Max, which I would like to introduce to you in this post. [Read More]

Introducing the Quad SFP28 FMC

High-speed Ethernet connectivity for FPGAs

Posted on July 16, 2024 | Jeff Johnson

In the next few days I will kick off the launch of another new Opsero FMC product: the Quad SFP28 FMC. This FMC card has 4x SFP28 slots that are compatible with SFP, SFP+ and SFP28 modules. The FMC and the reference designs that we are currently developing will enable 4x 10G/25G Ethernet links on a multitude of FPGA/MPSoC/RFSoC development boards including the newer Versal ACAP boards. We’ve already pushed working 10G/25G designs to the Github repo for the ZCU104, ZCU102, ZCU106, ZCU111 and ZCU208 with more coming soon. [Read More]

The M.2 M-key Stack FMC Unveiled

A fresh approach to getting more from FMC

Posted on April 3, 2024 | Jeff Johnson

The FPGA Mezzanine Card standard (VITA 57.1) has significantly enhanced the FPGA ecosystem by decoupling the FPGA board from the input/output components. Years ago, if you wanted to process samples from an ADC, you would buy an FPGA board with an on-board ADC. If you wanted to add a DAC, or upgrade the ADC, you would replace the entire board! The FMC innovation provided developers with choice, so that they could select the right FPGA for their specific I/O requirements. [Read More]

Using NVMe SSDs with Versal VCK190 and VMK180

Posted on March 18, 2024 | Jeff Johnson

High-capacity non-volatile storage is pretty handy in the intensive computing applications that the Versal ACAP adaptive SoCs get employed in. NVMe SSDs are a perfect way to provide that storage because they can directly interface with the Versal’s integrated blocks for PCIe. Those integrated blocks are Gen4 compliant which makes for an extremely high bandwidth connection between the FPGA fabric and the storage medium. Over the past couple of weeks, my team and I have been bringing up an NVMe SSD on the Versal AI Core VCK190 Evaluation kit using the FPGA Drive FMC Gen4 adapter. [Read More]

A Smart Camera implemented in PetaLinux 2022.1 on ZCU104

Using a Raspberry Pi camera

Posted on July 12, 2023 | Jeff Johnson

In this post we’re going to build a smart camera using a Raspberry Pi camera, the ZCU104 board and PetaLinux. We’re going to do this by leveraging the Smartcam app which was originally designed for the AMD Xilinx Kria SoM. Our version of the Smartcam app can take video inputs from a Raspberry Pi camera, a USB camera or a file, and it can output video to a DisplayPort monitor, a file or via Ethernet over Real-time Transport Protocol. [Read More]

Benchmarking an FPGA based AI Vision application

Docker, Ubuntu and PetaLinux put to the test

Posted on July 10, 2023 | Jeff Johnson

Many smart vision applications need to make fast decisions: autonomous vehicles, drones, surveillance and industrial robotics are only a few examples. When developing these kinds of AI vision systems, understanding performance-affecting factors is critical. In this post, we’ll explore two such factors: the operating system and camera type. We’ll measure and compare the performance of the NLP-SmartVision app on the ZCU104 board. The setups we’ll use are: Operating system: Docker container on Certified Ubuntu 22. [Read More]

NLP-SmartVision in PetaLinux on ZCU104

Using Raspberry Pi cameras

Posted on July 5, 2023 | Jeff Johnson

In the last post we looked at how to run the Smartcam and NLP-SmartVision apps on the ZCU106 and Certified Ubuntu 22.04 LTS. One reader mentioned that running these apps in a Docker container on Ubuntu probably comes with a performance penalty when compared to running it on a lean PetaLinux build. This piqued my curiosity so in this post, we’re going to get the NLP-Smartvision app running in PetaLinux on the ZCU104 and then in the next post we’ll measure whatever penalty there may be to the throughput (in frames per second) and/or the glass-to-glass latency (in milliseconds). [Read More]

Develop smart vision apps for ZCU106 and RPi Camera FMC

Using Certified Ubuntu 22.04 LTS for Xilinx devices

Posted on June 15, 2023 | Jeff Johnson

If you want to develop smart vision applications using Raspberry Pi cameras, the Zynq UltraScale+ MPSoC and Ubuntu 22.04 LTS, we’ve just released some code and the hardware to help you get started. For the launch of our new RPi Camera FMC, we’ve ported the Kria Vitis Platforms and Overlays over to the ZCU106 armed with an RPi Camera FMC. The result is that you can now build and develop the Smartcam and NLP-Smartvision applications in Certified Ubuntu 22. [Read More]

Camera FMC: Connecting MIPI cameras to FPGAs

Posted on March 1, 2023 | Jeff Johnson

FPGAs and MPSoCs are ideally suited for machine vision applications due to their ability to process large amounts of data in parallel and at high speeds. FPGAs can run highly power efficient neural network implementations and benefit from ultra low latency connections to multiple image sensors. Given the inherent strengths of FPGAs for machine vision, it surprises me that GPUs have become the dominant hardware platform for deep learning applications1 in recent years. [Read More]