iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
AMD announced the Versal RF Series on December 10, 2024. The adaptive-SoC family combines direct RF-sampling converters, dedicated DSP hardware, AI Engines, programmable logic and Arm processors on a monolithic device. AMD claims up to 80 TOPS of DSP performance, RF ADC rates up to 32 GSPS, 14-bit resolution and direct sampling up to 18 GHz. Those are maximum or vendor-defined figures—not guarantees for every workload or signal.
The announcement targeted aerospace and defense systems, radar and spectrum operations, communications and advanced test equipment rather than consumer electronics. AMD originally described silicon samples and evaluation kits as expected in Q4 2025 and production shipments in the first half of 2027; those dates should not be treated as proof of current commercial availability.
Versal RF at a glance
| Capability | AMD-announced figure | How to interpret it |
|---|---|---|
| DSP performance | Up to 80 TOPS | Maximum theoretical figure that varies with device, configuration and workload |
| RF ADC sampling | Up to 32 GSPS | Device- and channel-dependent converter rate |
| Resolution | 14-bit, with calibration | Does not establish effective number of bits, SNR or SFDR |
| Direct RF sampling | Up to 18 GHz | Not a promise of uniform clean performance across that range |
| Dedicated DSP | FFT/iFFT, channelizer, polyphase arbitrary resampler and LDPC decoder | Fixed-function acceleration can reduce programmable-logic use |
| Availability stated in 2024 | Samples and kits Q4 2025; production H1 2027 | Original forecast; verify present status with AMD |
AMD’s announcement and product overview are available at AMD’s December 10, 2024 release and the Versal RF Series product page.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
What AMD introduced
Versal RF is described as AMD’s fifth generation of direct-RF devices, extending the Zynq RFSoC idea into the Versal adaptive-SoC architecture. A single die brings together:
#1 Best Overall
- Ready for Meshtastic: Start a LoRa mesh build faster with pre-flashed Meshtastic firmware. Use it to join or create a mesh network, test node behavior, or begin a DIY off-grid messaging project
- ESP32-S3 + SX1262 Wireless Core: Built around a dual-core ESP32-S3 MCU and SX1262 LoRa radio, supporting 862–930MHz LoRa plus 2.4GHz Wi-Fi and BLE 5.0 for mesh, router and sensor projects
- Low-Friction Starter Kit: The press-fit board design reduces basic assembly work, while the included antenna setup helps new makers avoid starting from a bare board with missing RF accessories
- Arduino, MicroPython and Grove Expansion: Use I2C, UART, SPI, GPIO/PWM and ADC access with compatible XIAO expansion boards or Grove modules to add sensors, displays or custom functions
- Compact Platform, Flexible Builds: The 21 × 18 mm XIAO form factor fits compact prototypes, wearables and embedded devices, while modular add-ons let you choose the GPS, display, power and enclosure your project needs
- RF analog-to-digital and digital-to-analog converters.
- Dedicated DSP hard IP.
- AI Engines for highly parallel dataflow computation.
- Adaptive programmable logic for custom datapaths and control.
- Arm-based processing for software, management and coordination.
- Memory and high-speed interfaces determined by the particular device.
This integration can reduce the number of separate converter, FPGA and processor devices and can remove some board-level JESD204 data links. It does not remove antennas, filters, amplifiers, protection, clock generation, power regulation, calibration or thermal engineering from an RF system.
Why direct RF sampling matters
In a conventional receiver, mixers translate signals to an intermediate frequency before conversion. A wideband design may then move samples over high-speed serial links to an FPGA, DSP or processor. Direct RF sampling places conversion closer to the antenna-facing path, allowing digital downconversion, channelization and analysis with fewer intermediate stages.
Putting conversion and compute on one die can reduce interconnect latency, board area, signal-integrity exposure and data-movement power. It can also let the device discard unwanted data early: channelize or decimate on-chip instead of sending every raw sample to external memory. The trade-off is concentration of thermal, mixed-signal and supply-design risk in one package, plus less freedom to select each converter and processor independently.
Inside the compute architecture
Dedicated DSP hard IP
AMD identifies hard blocks for FFT and inverse FFT, channelization, polyphase arbitrary resampling and LDPC decoding. These functions are common in radar, communications and instrumentation. Implementing them as fixed silicon can improve throughput, area and power compared with building every operation in soft logic, although supported modes and interfaces are less flexible.
Rank #2
- 2.4GHz Dual Mode WiFi + Bluetooth Development Board
- Support LWIP protocol, Freertos;ESP32 is a safe, reliable, and scalable to a variety of applications
- SupportThree Modes: AP, STA, and AP+STA
- Ultra-Low power consumption, Compatible with Arduino IDE
- 1PCS 30Pin ESP32 Development Board 2.4GHz WiFi Dual Cores Microcontroller Integrated with Antenna RF Low Noise Amplifiers Filters
AI Engines
AI Engines contribute parallel, streaming computation for kernels such as filtering, transforms, beamforming, channelization and matrix-style operations. “AI Engine” here should not be read as proof that the device is an 80-TOPS neural-network accelerator.
Programmable logic
Programmable logic remains valuable for custom signal paths, protocols, deterministic control, data movement and mission-specific algorithms. It supports designs whose waveforms or processing chains may change after deployment. That flexibility requires FPGA timing analysis, hardware/software partitioning, RF validation and substantial verification effort.
Arm processing
The Arm subsystem handles configuration, system control, monitoring, communications stacks, security functions and coordination between software and datapaths. It is complementary to—not a replacement for—the high-throughput RF and DSP fabric.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →What “80 TOPS” does and does not mean
AMD’s 80-TOPS statement is a maximum theoretical DSP-compute claim. It depends on the selected device, operation format, block utilization, configuration and an optimal workload. The announcement does not make it an apples-to-apples comparison with GPU or NPU neural-network TOPS.
Rank #3
- RP2040-LoRa-HF development board integrates the new generation SX1262 RF chip, 850 ~ 930MHz frequency band, supports FSK, GFSK, LoRa modulation, featuring better anti-blocking and ultra-long distance communication
- LoRa development board with -148dBm high reception sensitivity and programmable emit power up to 22dBm, supports preamble detection, with CRC, up to 256 bytes data packet engine
- Based on Raspberry Pi RP2040 microcontroller chip, Dual-core Arm Cortex M0+ processor, flexible clock running up to 133 MHz, 264KB of SRAM, and 2MB of onboard Flash memory
- Supports C, comprehensive SDK, online Dev resources and tutorials to help you easily get started
- Package Content: RP2040-LoRa-HF x1, 2DB antenna 850~930MHz ~11cm x1, IPEX 1 to SMA adapter cable ~17cm x1, USB Type-C adapter board x1, FPC cable ~15cm x1
- The figure may combine contributions from hard IP, AI Engines and other DSP resources.
- Operation-count conventions, such as multiply and accumulate treatment, matter.
- Real applications lose capacity to buffering, control, memory movement, precision requirements and underutilized blocks.
- A channelizer, beamformer, modem and spectrum monitor will expose different bottlenecks.
AMD also reports up to 19 times the DSP compute of a Zynq UltraScale+ RFSoC Gen 3 device in channelizer mode. That is an AMD theoretical comparison under stated assumptions, not a universal 19-times application-speed improvement.
Putting 32 GSPS, 14-bit and 18 GHz in context
32 GSPS
A 32-GSPS converter produces an enormous raw data stream. Usable system throughput depends on converter count, decimation, channelization, memory bandwidth, external links, latency and whether raw samples leave the device. High-rate conversion is most useful when on-chip processing reduces data before transport.
14-bit resolution
Nominal resolution is not the same as effective number of bits. Engineers still need effective-number-of-bits, SNR, spurious-free dynamic range, linearity, clock-jitter and calibration data at the frequencies and temperatures that matter to the design.
Up to 18 GHz
The 18-GHz wording describes a maximum direct-sampling or observable-frequency capability under device conditions. It does not mean every 18-GHz signal has the same dynamic range, instantaneous bandwidth or spectral purity. Input power, filtering, alias zones, clock quality, active-channel count, temperature and board implementation all affect results. Nor does 32 GSPS automatically mean 32 GHz of usable signal bandwidth.
Rank #4
- Advanced Control Interface: Three-wire SPI to control pin and state locking pin, allowing all functions including point frequency sweep and frequency hopping, stepping to 1K, low frequency step can be 0.1K, according to crystal frequency
- Complete Development Package: Default + -50ppm 25M active crystal oscillator with circuit diagram in PDF format and STM32 test program provided for easy integration
- Professional Circuit Design: This ADF4351 source development board features well designed circuit board layout for optimal performance and reliability
- Software Compatibility: Can be controlled by the upper computer official software for convenient programming and configuration
- Accessible Pin Configuration: All control pins are leaded out for convenient access and flexible integration with your projects
Where Versal RF fits
Aerospace, defense and spectrum operations
Phased-array radar, electronic-warfare systems, signals intelligence and wideband spectrum monitoring benefit from simultaneous channels, deterministic latency and reconfigurable processing. Final qualification, ruggedization and radiation requirements remain program-specific.
Communications and satellites
Digital upconversion and downconversion, resampling, channelization and LDPC decoding map naturally to satellite, military and software-defined-radio systems. The architecture is useful where waveforms and channel plans evolve.
Test and measurement
Oscilloscopes, spectrum analyzers, RF generators and other wideband instruments can use direct conversion plus on-chip DSP to capture, transform and analyze signals with less external data movement.
Recommended Free Tools
Pre-6G research
AMD has identified advanced communications and pre-6G experimentation as target areas. That is a forward-looking application position, not evidence that Versal RF is a finished commercial 6G platform.
Best Value
- 2.4GHz Dual Mode WiFi + Bluetooth Development Board
- Support LWIP protocol, Freertos
- SupportThree Modes: AP, STA, and AP+STA
- Ultra-Low power consumption, Compatible with Arduino IDE
- ESP32 is a safe, reliable, and scalable to a variety of applications
Versal RF versus other design approaches
| Approach | Potential strengths | Trade-offs |
|---|---|---|
| Versal RF integrated SoC | Direct conversion, adaptive logic, hard DSP, AI Engines and Arm control in one device | Complex development, concentrated thermal risk and dependence on device availability |
| Discrete RF converter plus FPGA | Component-level choice and easier substitution in some programs | More board area, JESD204 links, interconnect power and integration work |
| Earlier Zynq UltraScale+ RFSoC | Potentially more mature for existing designs | Less compute headroom according to AMD’s comparison; suitability is workload-specific |
| GPU or CPU processing | Software-friendly prototyping and broad ecosystems | Often less attractive for deterministic latency and SWaP-constrained RF datapaths |
No architecture wins universally. A discrete design may be preferable when sourcing flexibility or independent converter selection matters more than integration. A GPU may be sensible for exploratory software workloads, while Versal RF is aimed at tightly coupled, high-rate and reconfigurable datapaths.
Power, thermal and integration limits
AMD projects up to 80% lower dynamic power for selected hard-IP functions versus comparable soft-logic implementations. This is an engineering projection for those functions, not an 80% reduction in total board or system power. A monolithic RF device can still create high thermal density, demanding clock and power-distribution requirements, and difficult analog isolation. External RF components and cooling remain part of the design.
Availability and development path
AMD’s 2024 announcement said tools were available at announcement, samples and evaluation kits were expected in Q4 2025, and production shipments were expected in the first half of 2027. As of August 18, 2026, the cited material does not independently confirm that production shipments have begun.
Free tools Windows power users keep installed
One-click scans. No signup required.
AMD’s current Vivado page lists Vivado 2026.1 with Versal RF Series support. That establishes software listing, not universal access to production parts or evaluation kits. Hardware teams should also plan for the broader adaptive-SoC flow, including Vitis for software and acceleration development; see AMD Vitis.
The official AMD evaluation-kit storefront is the appropriate buying route, but its existence alone does not prove that a Versal RF-specific kit is in stock. Pricing was not publicly stated in the cited materials and is likely to be quote-based and device-specific.
Checklist for evaluating a design
- Define frequency range, instantaneous bandwidth, channel count and synchronization requirements.
- Obtain ENOB, SNR, SFDR, phase-noise and jitter data for the intended operating conditions.
- Map the workload to hard DSP, AI Engines, programmable logic and Arm software.
- Estimate data reduction before external memory or links; do not size the system from GSPS alone.
- Budget power, cooling, clocking, calibration and RF isolation early.
- Confirm package, temperature, security, lifecycle and qualification requirements.
- Verify evaluation-kit, sample and production status directly with AMD.
- Assess internal Vivado/Vitis, RF, verification and embedded-software expertise.
Bottom line
Versal RF is most compelling for wideband, multichannel RF systems that need deterministic low-latency processing, reconfigurability and tight size, weight and power integration. AMD’s 80-TOPS, 32-GSPS, 14-bit and 18-GHz figures explain the family’s ambition, but they are starting points for an engineering trade study—not substitutes for converter measurements, workload benchmarks, thermal analysis or verified supply dates.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitches
