03 February 2005
Arithmatica Develops Integrated Flow with Cadence Encounter RTL Compiler to Accelerate Design and Verification of Math-Critical Chips
Arithmatica, Inc ., the silicon math company, announced an integrated, front-end flow for mutual customers of Arithmatica and Cadence Design Systems, Inc . intended to improve quality of silicon for math-critical chips. Developed as part of the Cadence® OpenChoice Program, the flow includes best-in-class Arithmatica CellMath silicon IP as well as leading-edge Cadence synthesis technology - EncounterT RTL Compiler synthesis -- and Encounter Conformal® formal verification tools. Users can quickly verify equivalency to design specifications, making it ideal for customers that face very aggressive design and verification cycles, such as those involved in consumer electronics. Cadence is the largest supplier of electronic design technologies, methodology services and design services, while Arithmatica is the only company focused on using advances in silicon math to increase speed and/or lower costs for math-intensive ICs.
"We are pleased to welcome Arithmatica as the newest member of the OpenChoice program," said Jan Willis, senior vice president of industry marketing at Cadence. "Cadence's growing strengths in the RTL design and verification space are leveraged by integrating quality IP with Encounter RTL Compiler. Our collaboration with Arithmatica will enable customers to bring their designs to market faster and with higher quality of silicon."
The Cadence OpenChoice program enables interoperability and facilitates open collaboration with IP providers to build, validate, and deliver accurate models optimized for Cadence design and verification solutions. The program aims to ensure IP quality and provide the semiconductor industry with access to optimized IP through a complete IP catalog. This optimizes the electronics design chain and accelerates customer time to market.
New Flow Streamlines CellMath IP Delivery
The new integrated flow improves silicon end-results for math-critical chip design in two key ways. First, Arithmatica's work with Cadence ensures that CellMath IP, delivered typically as a gate-level netlist, is proven against the bit-accurate Verilog simulation model with the aid of Cadence's Conformal formal equivalency checking technology, which includes unique data path capabilities to verify equivalence of the behavioral model and netlist. Second, use of Encounter RTL Compiler ensures that the CellMath IP architectures are efficiently integrated and optimized within an overall chip design. Using CellMath IP in such a flow, NVIDIA Corporation achieved significant chip area reductions in major floating-point blocks.
Chris Malachowsky, co-founder and vice president of hardware engineering at NVIDIA Corporation, said, "NVIDIA relies on this type of collaboration to help us reach new thresholds of graphics performance within very aggressive design cycles. Gaining the performance of CellMath IP while using multiple synthesis vendor flows has helped us set the bar for GPU performance with the NVIDIA® GeForceT 6 series GPU, which some have called 'a gamer's dream.'"
Dave Burow, Chairman of the Board at Arithmatica, added, "The best way to create design flows that really work is through the collaboration of all parties working on challenging production designs. NVIDIA is great at providing the access and feedback needed by its EDA and IP suppliers to integrate new technologies into their flow. Many of our target customers, like NVIDIA, are pushing the performance envelope, and within these customers we often see RTL Compiler being used. We expect that many mutual customers of Cadence and Arithmatica can benefit from this flow."
Become a member of the CIMdata PLM Community to receive your daily PLM news and much more.
Tell us what you think of the CIMdata Newsletter. Send your feedback.
CIMdata is committed to your privacy. Your personal information will never be sold or shared outside of CIMdata without your express permission.