3D game running on FPGA shown to be 50x more efficient than on x86 hardware

3D game FPGA

Sphery vs. shapes is an open-source 3D raytraced game written in C and translated into FPGA bitstream that runs 50 times more efficiently on FPGA hardware than on an AMD Ryzen processor. Verilog and VHDL languages typically used on FPGA are not well-suited to game development or other complex applications, so instead, Victor Suarez Rovere and Julian Kemmerer relied on Julian’s “PipelineC” C-like hardware description language (HDL) and Victor’s CflexHDL tool that include parser/generator and math types library in order to run the same code on PC with a standard compile, and on FPGA through a custom C to VHDL translator. More details about the game development and results are provided in a white paper. Some math functions were needed, including: floating point addition, subtraction, multiplication, division, reciprocals, square root, inverse square roots, vector dot products, vector normalization, etc. Fixed point counterparts were also used for performance reasons and to […]

ChipWhisperer-Husky is a palm-sized power analysis and fault injection tool (Crowdfunding)

power analysis fault injection tool STM32F Target Board

NewAE Technology’s ChipWhisperer-Husky is a compact tool designed for side-channel power analysis and fault injection with features such as a high-speed logic analyzer used to visualize glitches, real-time data streaming for attacking asymmetric algorithms, and support for JTAG/SWD programming. The security research company explains its device delivers a more stable and reliable experience compared to other off-the-shelf test gear such as oscilloscopes and function generators thanks to features such as synchronous sampling, which means the sample clock of your target device and the sample clock of ChipWhisperer-Husky can be perfectly aligned, or the ability to generate glitches, including clock glitches that can be less than a  nanosecond wide.   ChipWhisperer-Husky key features and hardware specifications: Synchronous clock for capture board and target board for significantly improved performance over a typical asynchronous oscilloscope setup 12-bit 200MS/s ADC for capturing power traces – It can be clocked at both the same clock […]

Ztachip open-source RISC-V AI accelerator performs up to 50 times faster

Ztachip RISC-V AI accelerator

Ztachip is an open-source RISC-V accelerator for vision and AI edge applications running on low-end FPGA devices or custom ASIC that is said to perform 20 to 50 times faster than on non-accelerated RISC-V implementations, and is also better than RISC-V cores with vector extensions (no numbers were provided here). Ztachip, pronounced zeta-chip, is not tied to a particular architecture, but the example code features a RISC-V core based on the VexRiscv implementation and can accelerate common computer vision tasks such as edge detection, optical flow, motion detection, color conversion, as well as TensorFlow AI models without retraining. The open-source AI accelerator has been tested on Digilent ArtyA7-100T FPGA board in combination with a PMOD VGA module to connect to a display and an OV7670 VGA camera module. You can then build the sample found on Github with the Xilinx Vivado Webpack free edition and flash it to the board […]

$5 CH347 board is a USB 2.0 bridge to I2C, SPI, UART, JTAG, and GPIO

small CH347 development board

MuseLab USB-HS-Bridge is an inexpensive ($5) board based on WCH CH347 chip with a USB 2.0 Type-C interface that acts as a bridge for I2C, SPI, UART, and JTAG interfaces, as well as GPIOs. It’s notably useful to debug and download bitstreams to FPGA development boards, but it can also be used to connect various peripherals such as I2C sensors, SPI flash devices, UART devices to basically any host with a spare USB 2.0 host port. USB-HS-Bridge specifications: Chip – WCH CH347 high-speed USB to UART, I2C, SPI and JTAG chip (See link to the datasheet for details) USB – 1x USB 2.0 Type-C port with up to 480 Mbps data rate I/Os – 2x 16-pin header with 2x UART interfaces up to 9 Mbps baudrate 1x I2C for EEPROM or sensors 1x SPI master interface with 2 chip select signals to control up to 2x SPI slave devices. The […]

Acromag XMC-7A50-AP323 – An XMC module based on AMD Xilinx Artix-7 FPGA

XMC Module Artix-7 FPGA

Acromag XMC-7A50-AP323 is an XMC (Switched Mezzanine Card) module based on a Xilinx Artix-7 FPGA with 48 TTL I/O channels plus a 16-bit ADC for 20 differential or 40 single-ended analog inputs. Designed for commercial off-the-shelf (COSTS) applications, Acromag XMC modules are RoHS compliant, and suitable for automation applications, scientific development labs, as well as aerospace and military applications. Acromag XMC-7A50-AP323 module specifications: FPGA – AMD Xilinx Artix-7 (XC7A050) FPGA with 52,160 logic cells, 65,200 Flip flops, 2,700 kb block RAM, 120 DSP slices Storage – 32Mbit QSPI flash memory FPGA Digital I/O 48x I/O channels controlled in groups of eight channels, 5V tolerant TTL, RS485, and LVDS interface options: Build Option A: 24x EIA-485/422 channels Build Option B: 24x TTL and 12x EIA-485/422 channels Build Option C: 24x LVDS channels Analog Input 20 differential or 40 single-ended inputs Flexible scan control 16-bit A/D resolution 8μs conversion time FIFO buffer […]

XRF16 Gen3 SOM features Xilinx Zynq UltraScale+ ZU49DR RFSoC with up to 6GHz bandwidth

Avnet XRF16 Xilinx RFSoC Gen3 SOM

We’ve written about Xilinx Zynq UltraScale+ MPSoCs that combine Arm Cortex-A53/R5 cores and Mali-400 GPU with Ultrascale FPGA fabric several times over the course of a few years. But AMD-Xilinx also offers the Zynq UltraScale+ RFSoC single-chip adaptable radio platforms that support up to 7.125GHz analog bandwidth. The topic came to my attention because of an upcoming ZU49DR SoM from iWave Systems that seems to be under development but also noticed Avnet had launched a solution last year with the XRF16 Gen3 SoM featuring the same third-generation Zynq Ultrascale+ ZU49DR RFSoC with 16 RF-ADC, 16 RF-DAC channels, and 6GHz RF bandwidth. Avnet XRF16 specifications: Main chip – Xilinx Zynq UltraScale+ Gen3 ZU49DR RFSoC with Quad-core Arm Cortex-A53 processing subsystem Dual-core Arm Cortex-R5F MPCore up to 533MHz 16x ADCs, 14-bit up to 2.5 GSPS 16x DACs, 14-bit up to 9.85 GSPS (10 GSPS Available) 1 GbE, PCIe Gen1/2, SATA, USB2/3 UltraScale+ […]

STEPFPGA FPGA board is programmable with a Web IDE (Crowdfunding)

STEPFPGA education FPGA board

STEPFPGA MXO2Core miniature FPGA development board is based on Lattice MXO2-4000 FPGA, and designed for education with an easy-to-use Web IDE, instead of the more traditional tools that can be frustrating to use, and detailed tutorials. The board also comes with a 2-digit segment display, some LEDs, push buttons, and a 4-way DIP switch, as well as two rows of twenty pins for I/O expansion, and a USB Type-C port used for power, programming, or mass storage. STEPFPGA MXO2Core specifications: FPGA – Lattice Semi MachXO2 X02-4000 FPGA with 4320 LUTs Display – 2-digit segment display USB – 1x USB Type-C port for power, programming (UART), and mass storage Expansion – 2x 20-pin headers with up to 36x GPIOs, SPI, I2C, 3.3V, VBUS, GND; breadboard-compatible Misc – 2x RGB LEDs, 8x red LEDs, 4-way DIP switch, 4x push buttons Power Supply – 5V via a USB port Dimensions – Small four-layer […]

MNT Pocket Reform 7-inch modular mini laptop takes a range of Arm (and FPGA) modules

MNT Pocket Reform

MNT Pocket Reform is an open-source hardware mini laptop with a 7-inch Full HD display, an ortholinear mechanical keyboard, and trackball, that follows the path of its older and bigger sibling:  the MNT Reform 2 laptop initially launched with an NXP i.MX 8M quad-core Arm Cortex-A53 module. The new laptop will not only support a similar “NXP i.MX 8M Plus” module but also a range of other Arm modules namely an NXP Layerscape LS1028A module with up to 16GB RAM, the Raspberry Pi CM4 module via an adapter, Pine64 SOQuartz (RK3566, up to 8GB RAM), as well as based on AMD Xilinx Kintex-7 FPGA for industrial use. MNT Pocket Reform specifications: Available system-on-modules Standard: NXP i.MX 8M Plus quad-core Arm Cortex-A53 @ 1.8GHz with 4 or 8 GB DDR4, Vivante GC7000UL GPU, 2.3 TOPS NPU NXP Layerscape LS1028A dual-core Arm Cortex-A72 with 8 or 16GB DDR4, Vivante GC7000UL GPU Raspberry […]

Exit mobile version
UP 7000 x86 SBC