Training Example: BittWare – Review the Data, Give Your Score & Compare to the Real AI Evaluation

[H5] We are a world-class supplier of electronics and systems, ranging from COTS to custom and commercial to defense.

[H5] Browse our products and capabilities below including systems, modules, microelectronics, software, and IP for a wide range of applications.

[IMG: BittWare logo]

[H5] Rugged + Enterprise Boards

PCIe FPGA Cards →

Rugged VPX/VNX+ →

Custom Cards/Modules →

[H5] Systems + Software

TeraBox FPGA Servers →

WaveBox RF Appliances →

Partner IP & Solutions →

[H5] Advanced Packaging

Miniaturized Modules →

Ruggedization →

Advanced SMT & Bare Die →

3D Packaging →

SiP-System in Package →

Smart Sensors →

[H5] Component Services

IC Lifecycle Management →

IC Interposers →

BGA Reballing →

Pogo Pins →

MICTOR Cables →

[H3] Markets We Serve

Aerospace & Defense

RF + Satcom

AI/ML

Fintech + Low Latency

Broadcast + Video

[H3] Aerospace & Defense Solutions

[H2] Ready for Your Toughest Challenges

[H4] Learn About New VPX Products →

[H3] Satcom RF Solutions

[H2] RFX + WaveBox

[H4] Mix, Amplify, and Digitize All in One →

[H3] Positron Atlas

[H2] Advanced AI Inference

[H4] Built on BittWare →

[H3] ÜberNIC

[H4] Fully hardware-accelerated network stack, with hardware built on BittWare. →

[H3] manifold CLOUD

[H2] Service-Based Video Architecture

[H4] Built on BittWare →

[H2] In-House Design & Manufacturing

Our customers count on us for a collaborative approach that ensures their solutions are done right. From enterprise‑scale PCIe deployments to specialized rugged embedded designs, our engineers bring deep experience and hands‑on problem‑solving.What sets us apart is our fully in‑house capability — not just our equipment, but the people who know how to work alongside customers to shape the right solution. That partnership makes us a reliable, risk‑reducing force in their development process.

[IMG: Images depicting card design, manufacturing, and development]

[IMG: AV-870p product thumbnail]

[H2] Featuring AMD, Altera, and Achronix Devices

To build the world’s most advanced electronics, we work with top silicon vendors to ensure the best value for customers both in terms of value and availability.

[H2] Solutions Built on BittWare

Pre-programmed solutions our customers offer without needing FPGA engineering teams and ready for enterprise and edge deployment.

[IMG: LMS UberNIC card thumbnail]

[H2]
LMS ÜberNIC

Low-Latency NIC for Fintech

[H2]
Positron Atlas

FPGA-Powered AI

[IMG: TeraBox 140xB server thumbnail]

[H2]
manifold CLOUD

Broadcast Multiviewer

[IMG: IA-440i PCIe card thumbnail]

[H2]
Atomic Rules TK242

200G Capture

[IMG: 250-U2 module thumbnail]

[H2]
Eideticom Analytics Processor

P2P Storage/Data Acceleration

[H2]
Your Solution

Build Your Solution on BittWare!

[H1] RFX 880

[H2] Direct RF Card Featuring the AMD RFSoC

[H4] Analog: 4x 5 GSPS 14-bit ADCs + 4x 9.7 GSPS 14-bit DACs

[H4] Digital: Up to 400 Gb/s I/O

[H2] Our Direct RF Products

[H2] RFX 880

[H4] 4 ADCs + 4 DACs w/400G I/O
Optional Integrated Nyquist Filter/Amp

[H2] RFX 770

[H4] 4 ADCs + 4 DACs w/ PCIe Gen4 x8x8
Optional Integrated Nyquist Filter/Amp

[H2] WaveBox

[H4] 1U Enclosure w/ up to 3x RFX 880
2U Dual CPU w/up to 2x RFX 770

[H2] Integrated Amplification and Filtering

[H2] With RFX, You've got Options

At the heart of RFX products is the revolutionary AMD RFSoC chip, giving you a range of advantages over discreet systems. However, BittWare has leveraged our analog and digital design heritage to bring you more than expected from a typical RF card:Direct sampling between 50 MHz and 6 GHzBuilt-in amplifier/filter options to target first, second, or third NyquistWideband options to bypass the filtering/amplifiersUp to 400 Gb/s digital I/O with options for PCIe- or QSFP-focused architecturesModular build options for our WaveBox RF serversStandalone-capable using QSFPs for digital I/O and no PCIe

[H2] Powered by 3rd Gen AMD Zynq UltraScale+ RFSoC

RFX cards and modules are built around the powerful AMD RFSoC devices. We bring analog signals in, but it’s the direct RF chip approach that packs so much analog and digital functionality that many racks of previous equipment are no longer necessary!Lower power by eliminating JESD204 interfacesSmaller physical spaceProgrammable (FPGA) logic for diverse requirements and emerging standardsBuilt-in ARM processing lets you do more in a standalone deployment

[IMG: RFSoC 3D diagram]

[H2] Designed for WaveBox

The RFX family is designed as standalone cards or as modules in servers. We recommend the WaveBox family of enclosures and servers for the optimum mix of features and the advantage of purchasing an integrated system. Learn more on the WaveBox product page.

[H2] Where RFX Fits

At the heart of RFX cards and modules is the AMD RFSoC chip. It combines many of the traditionally discreet RF analog and digital components into a compact, configurable device. Applications including Satcom and test/measurement can benefit from these revolutionary devices.But with RFX we’ve gone further—bringing to the PCIe form factor optional amplifiers, and filters. The RFX is configurable too, with configurable inputs/outputs and Nyquist filters.

[H2] Want More Details?

[H3] Request the RFX 880 Hardware Reference Guide (HRG)

The HRG gives you much more detail about the card such as block diagrams, tables and descriptions.

Request the HRG

[H2] Analog Mixing, Amplification, and Filtering Options

[H2] On-Card ADCs and DACs

Choose from on-card filtering and amplification targeting Nyquist zones, or choose “wideband” and bring your own external filtering. Here’s the input ADC:

These ADC channels are each factory configurable (talk to us for pricing and availability). DACs provide 30 to 6 GHz with low-pass filtering. This transmit side diagram is below:

[H2] Utility Software Included with RFX and WaveBox products

In development, configuring advanced RF devices using scripts can be problematic. That’s why we developed RFX Surfer, and best of all it’s free with many of our direct RF cards and enclosures (including source code)!Watch our demo video to get the basics of what this GUI can do!

Transcript (click to show/hide)

Welcome to this demo of RFX Surfer, a graphical user interface designed to simplify the configuration, monitoring, and analysis of RFX and WaveBox products over Ethernet. RFX Surfer acts as a versatile client that can connect to many BittWare RF products including the RFX 880, 770, and WaveBox, providing users with detailed control, tone generation, and real-time visualization.Connecting to Your RFX CardGetting started with RFX Surfer is straightforward. Simply provide the IP address of your RFX product and hit connect. Once connected, the software retrieves key information stored on the card, including serial number, software version, firmware version, build time, acquisition type, and VCXO frequency. This data is displayed clearly on the host PC, offering immediate insight into the card’s identity and status.Hardware Monitoring and DiagnosticsWithin the hardware monitor tab, users can view vital system information such as temperatures from multiple sensors distributed across the card, and see alerts triggered by the card’s monitoring circuitry. This tab also provides access to the boot log — a valuable resource for diagnosing issues during development or characterization phases.Configurable Card SettingsThe settings tab presents a rich set of configurable parameters that adapt to the specific options of the connected card. These include clock source selection, PLL configurations, and specialized RF ADC and DAC settings associated with the AMD RF Data Converter IP.RFX Surfer’s GUI intelligently adapts according to the connected card model, enabling or disabling options as appropriate to prevent configuration errors. This smart behavior ensures users can only select valid settings, helping prevent mistakes during setup.Waveform Generation and VisualizationThe waveform tab focuses on the DAC outputs of the RFX card. Here, users can specify tones for up to four DAC channels independently. The interface provides graphical visualization of these tones in both the frequency and time domains, making it easy to understand the output signal characteristics.Users can select single or dual-tone modes, adjust frequencies and amplitudes, and observe the expected output of waveform. For example, setting a tone to 1.2 GHz reflects in the visualization. When the DAC output is connected to external test equipment such as a spectrum analyzer, you’d see a similar output. Of course, for demonstrations we can use loopback cables from one or more DAC channels to the analog inputs—the ADCs—and use the RFX Surfer Plot tab to see a real-time analysis of the signals from the loopback. In fact, let’s now switch to that now.Real-Time ADC Capture and FFT AnalysisSo now on the Plot tab, RFX Surfer enables real-time capture of snapshot data packets from the card’s ADCs. As data streams in, the tool performs Fast Fourier Transform (FFT) analysis on each packet, presenting live views in both time and frequency domains.Users control which ADC channels to enable, start and stop streaming, and monitor detailed logs showing card configuration updates. These visualizations can be saved as images for reporting or troubleshooting purposes.We also support saving a snapshot to a binary file, which you can then open in BinView. This tool allows you to look in more detail, see the raw data, or do further post-processing.Integrated Signal Generator ControlRFX Surfer also includes the ability to control external signal generators via IP, sending commands to adjust frequency and amplitude remotely. Users can manually set frequencies, sweep across ranges with defined start and stop points, step sizes, power levels, and automatically log FFT measurements at each step.This integrated functionality streamlines characterization workflows by synchronizing signal generation with data capture and analysis within a single interface.Dual-Tone and Intermodulation MeasurementsFor advanced testing, dual-tone configurations are supported on respective DAC and ADC channels. Enabling the intermodulation measurement option allows the software to lock onto multiple tones and display detailed detected tone information from the ADC side.Cross-Platform SupportRFX Surfer runs natively on Windows and Linux, providing flexibility for a range of development and production environments. This cross-platform compatibility ensures users can deploy the tool wherever needed without compromise.Documentation and Source Code AvailabilityComprehensive user documentation accompanies RFX Surfer, covering all GUI widgets, underlying architecture, communication protocols, and troubleshooting guidance. This helps new users quickly understand the tool’s capabilities and allows experienced users to explore advanced features confidently.Additionally, the source code for RFX Surfer is included with RFX purchases for you to customize.Summary and BenefitsIn summary, RFX Surfer is a smart, user-friendly interface that automates many of the complex tasks associated with configuring, monitoring, and analyzing RFX cards. It eliminates common configuration errors through its adaptive GUI, accelerates characterization with integrated signal generation and data analysis, and supports Linux and Windows platforms.Thank you for watching this demo of RFX Surfer. For more information, visit us at BittWare.com.

[H1] RFX 880 Block Diagram, Data Sheet, and Specifications

[H2]
RFX 880 Data Sheet

Download

[H2] Board Specifications

[H3] RFSoC
AMD Zynq UltraScale+ RFSoCXCZU43 in a G1517 packageCore speed grade -2
[H3] Processing Subsystem (PS)
Application Processing Unit Quad-core Arm Cortex-A53 MPCore at 1.2 GHzReal-Time Processing Unit Dual-core Arm Cortex-R5F MPCore at 525 MHz
[H3] Analog
4x 5 GSPS 14-bit ADCsWideband build option provides +0 to -27 dB gain from 50 MHz to 6 GHz (or 7.5 GHz) with no filtration1st Nyquist (L-Band) build option provides +41 to -29 dB on all 4 channels, constrained between 400 MHz and 2.5 GHz by filters FL1 and FL22nd and 3rd Nyquist filters are also available on amplified channels as a special orderMax input power -22 dBm on the amplified version and +15 dBm on the wideband

[H3] Analog (Continued)
4x 9.7 GSPS 14-bit DACsRange of +3 to -24 dBm (high power end of this range is a function of frequency)Constrained by a 4.8 GHz LPFBalun used in both ADC and DAC channels begins its frequency roll-off at 6 GHzProgrammable clocksExternal reference and triggersPush-on SMPM connectors with optional SMA pigtails
[H3] External digital interface
Up to 400 Gb/s available via front panel 2x QSFP-DDAMD Hard IP support for dual 100GbE

[H2] Interested in Pricing or More Information?

Our technical sales team is ready to provide availability and configuration information, or answer your technical questions.

"*" indicates required fields

[H1] LMS ÜberNIC | NIC Cards Leveraging PCIe 5.0 + CXL

[H2] Fully Hardware-Accelerated Network Stack

[H2] First Use Case Targets Financial Services Industry (FSI)

The First FPGA-Powered CXL NIC

[H3] PCIe 5.0 + CXL

Liquid-Markets-Solutions (LMS) have done something unique: moved the full network stack into an FPGA. What’s even better is that FPGA is Agilex I-Series from Intel, with the groundbreaking performance of PCIe 5.0 + CXL.
Called ÜberNIC, this family of NIC cards is built on BittWare hardware for enterprise-class deployment in datacenters.

[IMG: LMS UberNIC card thumbnail]

[H3] Game-Changing Performance for Fintech

[H4] ÜberNIC for Low-Latency Trading

[H2] Today's NICs Don't Stack Up for Market Trading

Fact: today’s market feeds regularly pass 850 Mb/s—the point at which even the best competitive NIC’s network stack begin to dramatically lose sufficient latency performance.Enter ÜberNIC, the first commercially-available FPGA NIC with the entire network stack in accelerated hardware, allowing for consistent low latency—ready for today’s markets and the future.

[IMG: NYSE and 825 Mb Stream]

[H3] Key Features at a Glance for Financial Services Industry (FSI)

[H3]
Performance

Raw performance improvement that’s consistent through the critical peak market feed moments when you need it most!

Jump to Performance Graph

[H3]
Utility

More fiber pairs per card means getting more out of each server. Powerful enough for ancillary services like PCAP and timestamping on the same card.

[H3]
Future Compatible

Support for latest PCIe 5.0, CXL, and link speeds up to 400G.

[H3] Foundation

[H3]
100% Offload

By moving the complete network stack into FPGA hardware, the ÜberNIC achieves unmatched performance.

[H3]
RFC & IEEE Compliant

[H3]
Fiber Port Density Leader

Single-slot card supports up to 16× 10/25G fiber pairs

[H3]
Multi-GB On-board Memory

[H3]
Single-Slot, Compact Form Factor

[IMG: Intel Agilex and CXL logos]

[H4] Tap into the Power of Agilex® 7

Central to the performance gains of ÜberNIC over competitors is the Intel Agilex 7 FPGA. This chip brings the core features like PCIe 5.0 and CXL—plus I/O giving double (or more) the port density of even top competitors.

[H3] Performance & Usability

[H3]
Full RTT (95th percentile): Sub-Microsecond Latency

At its core, ÜberNIC is a low-latency focused NIC. With the combination of full hardware acceleration and low-latency PCIe transfers, you get unmatched performance—even during peak market times.

[H3]
Significantly Higher Throughput Without Performance Loss

[H3]
Highly Consistent

[H3]
Multi-GB DDR Buffer Ensures Lossless Reception

[H3]
Plug & Play x86 I/O (BSD/POSIX; Raw Frame; Custom API)

[IMG: Manufacturing equipment]

[H4] Built on BittWare

BittWare brings enterprise-class products to market using the latest technology such as Agilex FPGAs. Today’s high-speed interfaces require significant care in the design and manufacturing process to make a reliable, high-quality product at scale.Pictured is our manufacturing facility at ISI, a Molex company, in California.

[H3] Extensibility

[H3]
FPGA-Exclusive Functions

Only FPGA-based solutions with bundled hardware network allows for all these extensible functions.

[H3]
MAC-based Timestamping

[H3]
Market Data Handling & Line Arbitrage

[H3]
Data Preprocessing, Filtering, & Enhancement

[H3]
MAC-based PCAP & Port Spanning

[H3]
Simultaneous Network I/O Support for SW & HW EUC

[H4] Ultra-Low Latency Performance

Showing RTT latency ÜberNIC up to 10 Gb/s with an FSI focused payload of 64B UDP

[H2] ÜberNIC Network I/O RTT Latency

[H3] FSI Focus (64B UDP Payload)

[H3] ÜberNIC Ultra+ | ÜberSock v2.0 | PCIe | W7-2495X

[H3] Configuration Overview

[H4] Each ÜberNIC configuration is single width and passively cooled.

[H5] ÜberNIC Ultra | ÜberNIC Ultra+
[H6] Single Width, Low Profile | Up to 8× 1/10/25G Links

[H5] ÜberNIC Ultra Wide | ÜberNIC Ultra+ Wide
[H6] Single Width, FH ¾L | Up to 16× 1/10/25G Links

[H3] Product Features

[H4] ÜberNIC is Available in a Variety of Configurations

[H3]
ÜberNIC

Ultra

PCIe 3/4/5 x16 | CXL x16

ÜberNIC Bitstream

Full Network Stack in Hardware

Low Profile (HHHL)

Intel Agilex I-SeriesFPGA: AGI 019

Link Speeds1/10/25/40/50/100/200/400 GbE

QSFP-DD up to 8× 1/10/25G,or 1× 400G, or 2× 200G, or 4× 100G

16GB DDR4 Total (Two Banks)

Get Pricing

[H3]
ÜberNIC

Ultra+

PCIe 3/4/5 x16 | CXL x16

User-Programmable ÜberNIC

Full Network Stack in Hardware

Low Profile (HHHL)

Intel Agilex I-SeriesFPGA: AGI 023

Link Speeds1/10/25/40/50/100/200/400 GbE

QSFP-DD up to 8× 1/10/25G,or 1× 400G, or 2× 200G, or 4× 100G

16GB DDR4 Total (Two Banks)

Get Pricing

[H3]
ÜberNIC

Ultra Wide

PCIe 3/4/5 x16 | CXL x16

ÜberNIC Bitstream

Full Network Stack in Hardware

Single Slot, FH ¾L

Intel Agilex I-SeriesFPGA: AGI 019

Link Speeds1/10/25/40/50/100/200/400 GbE

QSFP-DD up to 16× 1/10/25G,or 1× 400G, or 2× 200G, or 4× 100G

Up to 64GB DDR4 Total (Two Banks)

Get Pricing

[H3]
ÜberNIC

Ultra+ Wide

PCIe 3/4/5 x16 | CXL x16

User-Programmable ÜberNIC

Full Network Stack in Hardware

Single Slot, FH ¾L

Intel Agilex I-SeriesFPGA: AGI 023

Link Speeds1/10/25/40/50/100/200/400 GbE

QSFP-DD up to 16× 1/10/25G,or 1× 400G, or 2× 200G, or 4× 100G

Up to 64GB DDR4 Total (Two Banks)

Get Pricing

[H3] Get in Touch

Ready to learn more about ÜberNIC? Reach out to BittWare to learn more!

"*" indicates required fields

[IMG: 250-SoC PCIe card]

[H1] Learning Resources

[H2] Articles, White Papers, Videos and Sample Applications

[H2] Learn from BittWare

We regularly publish content that can help you in choosing the best technology to fit your needs. Browse our library of articles, white papers,  videos and some sample FPGA applications.

[H2] Upcoming Events

[H3] Where to find us in the next 3-4 months

[H2] STAC Summits – London & New York

[H3] April 29 (London) and May 20 (New York), 2026

If you’re in Fintech/FSI you know about STAC events! Visit us for the latest on low-latency and more acceleration technology. Event Website

[H2] 41st Space Symposium

[H3] Colorado Springs, Colorado | Exhibits: April 13-16, 2026 | Molex Booth

Get up to speed on our RF technology that applies to ground stations for satcom down/up transmissions. We have staff at the Molex booth 631 (Broadmoor Hall).Event Website

[H2] Past Events

2025

[H2] WEST

[H3] San Diego, California | February 10-12, 2025

[H2] 62nd Annual AOC International Symposium & Convention

[H3] National Harbor, Maryland | December 9-11, 2025 | Molex Booth

[H2] SC25

[H3] St Louis, Missouri | November 17-20, 2025

[H2] Military & Aviation Exhibition

[H3] Tel-Aviv | November 10, 2025 | Avnet Abacus Booth

[H2] STAC Fall Summits - London & New York

[H3] October 7 (London) and 28 (New York), 2025

[H2] MOSA Industry & Government Summit & Expo

[H3] National Harbor, Maryland | Exhibits: August 27-28, 2025

[H2] STAC Spring Summits - London & New York

[H3] April 29 (London) and May 13 (New York), 2025

[H2] Webinars

[H2]
Panel Discussion on Direct RF

We talk about the RFX and WaveBox products plus firmware an IP from Atomic rules. It's a discussion of all things direct RF!

[H2]
MACsec and IPsec on Altera™ FPGAs

Join us for a discussion of IP from Xiphera on MACsec and IPsec! Recording now available including Q&A.

[H2]
Data Capture/Record Using COTS

Our three experts discuss the use of COTS for data capture and recording. Co-sponsored with Mouser Electronics.

[H2]
HPC Next-gen Intel Agilex FPGAs

Special guest César González explains an example application using FPGAs at Barcelona Supercomputing Center + we cover next-gen Agilex devices.

[H2]
FPGA Panel Discussion

Three FPGA experts discuss the "data deluge" and various ways FPGAs are helping solve it.

[H2]
Arkville PCIe Gen4 Data Mover

This webinar introduces the Arkville IP core for moving data at up to 220 Gb/s over PCI Gen4 using Intel Agilex FPGAs.

[H2]
Intel Agilex Computational Storage

This webinar focuses on using Intel Agilex FPGAs plus Eideticom's NoLoad IP for storage acceleration.

[H2]
Intel oneAPI + 2D FFT Demo

We explore using the Intel oneAPI programming model for FPGAs, plus have a look at our 2D FFT on the 520N-MX demo. Learn about how these high-level tools get you to market faster!

[H2]
S7t-VG6 Achronix Card

Featuring 400 Gb/s and built on 7nm process, meet the S7t-VG6 VectorPath card from BittWare. Hear from Achronix on some of the innovative new chip features like Network-on-Chip and AI/ML-focused hard IP!

[H2]
TeraBox FPGA Servers

From the ultra-dense 1U 1400DN to the new 200DE edge server, our TeraBox line of FPGA servers are pre-integrated for ease of development and ready for enterprise-class deployment at scale.

[H2]
Computational Storage Acceleration

FPGAs are changing NVMe storage by moving compute closer to FLASH. Hear from Eideticom's work with Los Alamos National Lab and Myrtle.ai's new AI-based SEAL recommendation engine.

[H2] Articles and Papers

[H2]
FPGA Neural Networks

[H3] White Paper
We look at the inference of neural networks on FPGA devices, illustrating their strengths and weaknesses.

[H2]
Obsolescence Guide for AMD Alveo Accelerators

This guide explains the range of AMD Alveo cards that have been discontinued and suitable replacements from BittWAre. Cards covered are Alveo U25N, U200, U250, and U280.

[H2]
Homomorphic Encryption Acceleration

[H3] White Paper
FPGA acceleration enables this unique solution that allows compute on encrypted data without decrypting or sharing keys.

[H2]
Two Approaches to Rapidly Developing Customized FPGA Solutions

We've got a long history of custom solutions. Learn some of our best tips for getting to market faster with two approaches to custom requirements.

[H2]
Crossbar Switch: Sharing Resources in oneAPI

[H3] White Paper
Using oneAPI to efficiently share FPGA resources using a butterfly crossbar switch. Includes three use cases.

[H2]
oneAPI + 2D FFT using HBM2

[H3] White Paper
UPDATED for oneAPI
We use Intel's oneAPI on 520N-MX (with HBM2) to accelerate 2D FFT. Topics Covered: HBM2, oneAPI, FFT.

[H2]
NVMe SSD Write Performance

[H3] White Paper
Benchmarks and details on getting the best performance when writing for long periods to FLASH-based NVMe SSD storage.

[H2]
High Speed Data Capture

[H3] White Paper
Learn details about how we achieved 100 Gb/s over PCIe. Topics covered: sensor/signal processing.

[H2]
OVHcloud Anti DDoS

[H3] Case Study
Learn how OVHcloud uses BittWare FPGA cards in its anti-DDoS solution. Topics Covered: security, networking, packet processing.

[H2]
Timestamping

[H3] White Paper
We look at the use of very accurate timestamps on BittWare cards. Topics Covered: networking, packet processing.

[H2]
SmartNIC Shell

[H3] White Paper
Learn how to quickly build FPGA-powered NICs. Topics Covered: networking, packet processing.

[H2]
Loopback Example

[H3] White Paper
How to use AMD CMAC and configure packet processing pipeline. Topics Covered: networking, packet processing.

[H2]
Synthetic Traffic Generator

[H3] White Paper
How we turn our XUP-P3R card into a synthetic traffic generator for regression testing.

[H2]
StreamSurge

[H3] White Paper
Free 100G network load testing app for XUP-P3R customers. Topics Covered: networking, packet processing.

[H2]
Parser

[H3] White Paper
How we implement a packet parser using HLS C++ as compared to P4. Topics Covered: high-level synthesis, networking.

[H2]
NVMe Over Fabrics

[H3] White Paper
Using 250 family cards for NVMe storage. Topics Covered: flash, NVMe, storage.

[H2]
Comparing RTL to HLS

[H3] White Paper
Comparing HLS to RTL by building a networking RSS module. Topics Covered: development, networking, HLS, RSS.

[H2]
Acceleration of BWNNs

[H3] White Paper
Programming Stratix 10 using OpenCL for machine learning. Topics Covered: OpenCL, machine learning, Stratix 10.

[H2]
FPGA Accelerated NVMe Storage Solutions

[H3] White Paper
Learn how FPGAs are boosting performance of flash storage arrays in a number of ways. Topics Covered: storage acceleration, NAND flash, NVMe.

[H2]
Acceleration of CNNs Using FPGAs

[H3] White Paper
Using variable precision in FPGAs to build better machine learning inference networks. Topics Covered: machine learning, application tailoring, Arria 10

[H2]
Accelerate Kafka Producers with FPGAs

[H3] White Paper
Learn how FPGAs can accelerate this popular streaming platform. Topics Covered: streaming data, compression.

[H2] Video Library

[H2]
StreamSurge Network Test

We show two XUP-P3R FPGA boards running StreamSurge, which is a network test application we are making available for our customers.

[H2]
Liquid-Cooled FPGA Boards

See four 300+ AMP liquid-cooled FPGA cards in a TeraBox server.

[H2]
Liquid Cooling at 320A

Live demo from the SC18 show; we brought an XUPVVP board with liquid cooling installed, running out power test load at 320A FPGA core power.

[H2]
OpenCL on Intel Stratix 10

Presentation from the Supercomputing event including a case study from Paderborn University.

[H2]
Close-up with Xilinx VU13P

Unboxing the extremely large VU13P Virtex UltraScale+ FPGA. It's got 50% more logic than even the XUP-P3R's VU9P.

[H2]
A10SA4 with Intel Arria 10

With a powerful Intel Arria 10 GX FPGA, BittWare's A10SA4 offers powerful acceleration in a low-profile form factor.

[H2]
TeraBox X5000TT Expansion Chassis

In-depth look at this unusual solution for connecting dozens of FPGAs to a single host.

[H2]
BWNN White Paper Preview

Preview of the BWNN white paper you can find in the Resources section of the BittWare website.

[H2]
Find/Replace Demo on A10PL4

Demo of find/replace acceleration using IP from Axonerve, running on the Intel Arria 10-based A10PL4 card.

[H2]
Liquid Cooling 8× Cards

Carlos shows how we can install eight XUP-VV4 cards including liquid cooling in a single TeraBox server—and monitor them remotely.

[H2]
e4 FPGA Chassis from LDA Technologies

Sergey from LDA Technologies takes us through the unique 1U e4 FPGA server.