Storm-1 Stream Processor Reduces Systems Cost and Supports Analytics-Enabled Systems

SAN JOSE, Calif. – October 14, 2008 – Stream Processors Inc. (SPI), a fabless semiconductor company focused on making parallel processing simple, today announced the availability of the industry's first single-chip 8-channel DVR processor solution. Combining full software programmabilty in C with the compute performance of a dozen traditional DSPs, SPI's Storm-1 defines a new generation of processor that will drive a wave of smarter, more cost-efficient video surveillance products. With all functions implemented in software, innovation and time-to-market restrictions imposed by hardware-oriented ASICs can be eliminated.

" – As vendors look to reduce costs and meet new feature demands such as analytics, the traditional choices are becoming increasingly restrictive", comments Bengt Christensson, VP Business Development for SPI, "The Storm-1 platform breaks this negative trend, allowing customers to adopt a flexible software-driven model for faster development and lower costs".

Unlike traditional DSPs that are dependent on hardwired blocks to try to keep up with feature demands, SPI solves this problem by providing a highly efficient parallel processor platform. The processor headroom can be used for any combination of resolutions, framerates, codecs and other tasks, providing an effective way for vendors to reduce development costs, differentiate, and get new features to market without having to wait for hardware spins.

Standards-based Linux platform and high-level framework with a rich API allows for simple integration and plug-in of software libraries. SPI's video surveillance library includes H.264, MPEG-4 and MJPEG soft codecs, preprocessing such as deinterlacing and noise filtering, and display processing functions such as alpha-blending overlays and scaling. The encoders are tuned for video surveillance with attention to clarity of motion and rate control based on dynamic regions of interest, such as faces. The libraries are available as object or C source to allow for differentiation. Furthermore, the soft decoder enables interoperability with any brand of IP Camera, critical for Hybrid DVR / NVR functionality.

SPI and partners are adding to Storm-1's growing analytics library, including basic functions such as tamper detection, face detection, and object-based motion detection, as well as more traditional analytics functions, such as perimeter monitoring, and indexing for fast video search.

About Storm-1
Storm-1 is the industry's most powerful and cost-efficient processor for video and imaging applications. ANSI C programming enables software reuse, a faster learning curve, and reduced development effort.

Storm-1 features an on-chip Host CPU (330 MHz, 32-bit MIPS 4KEc) and a DSP subsystem (700 MHz, 80 x 32-bit 80 ALUs). A rich set of I/Os includes 8 BT.656 video inputs, HD display ports, Gigabit Ethernet and PCI. Four models range from the low-power G30 with 32,000 MMACs, to the high-performance G220 with 224,000 MMACs to allow for software-compatible scaling from single-port encoders and IP Cameras, to multi-channel Hybrid DVRs and analytics appliances.

Availability
All four Storm-1 chips are available for volume orders. Please contact SPI at sales@streamprocessors.com for the DVR platform options, including Reference Design Kits (RDKs) available from SPI's ecosystem partners.

About Stream Processors, Inc.
Stream Processors, Inc. (SPI) is a privately held fabless semiconductor company developing an innovative stream processing technology that that will help consumer and industrial companies accelerate time to market for their products, while greatly reducing system development costs. SPI was founded in 2004 in the recognition that the new era of compute-intensive applications requires radically increased levels of processor performance and efficiency. The company’s stream processing technology and award-winning development tools delivers dramatic improvements in productivity, performance, power, and cost by making parallel processing easily accessible. Additional information can be found at http://www.streamprocessors.com/.

# # #