Portland, Oregon: September 22, 2010 — At processing rates of up to 306 mega complex pixels per second, the new OptNgn 2D-FFT library elements are the perfect choice for inexpensive, low power sensor data stream processing of convolution and filtering-based applications from video, medical, radar/sonar, and scientific-based real-time sources.
2D-FFT Usage and Examples
The library elements, can process thousands of frames per second, which makes them particularly useful in convolution coprocessor based object finding, such as a streaming FFT-Filter-IFFT component that finds matches between the input stream and a target object library.
About the Library
These FFT library are delivered either individually as instantly downloadable and are high throughput complex 2-dimensional FFT instances, with 2-power dimension sizes, tuned for use inside FPGAs. Each element is instantly available now, via web download.
See our screen cast: . http://youtu.be/1VGaHWh9H8Q?hd=1 for a short video introduction.
Benefits
These elements can supply image processing horsepower previously only reserved for the fastest CPUs. Be the first to embed these powerful operators into your inexpensive, low power embedded systems, with an FPGA component.
Individual FFT Library elements are priced from $75-$800, as InstaIP Xilinx and Altera netlists, while the Mentor Graphics Precision elements are delivered as Precise-Encrypt RTL, which allows for mapping to any FPGA families supported by the Mentor Precision Synthesis toolkit.
About OptNgn Software, LLC
OptNgn produces and sells FPGA coprocessor acceleration libraries and services to customers that need to accelerate their systems using FPGA technology. Using proprietary technology, the company’s click-to-buy products help to reduce time to market and power, while radically improving system throughput over comparable solutions. Visit the company online at www.optngn.com to learn more, download reference designs, and view system demonstrations or to make an InstaIP purchase.

Leave a Reply
You must be logged in to post a comment.