Situation
NVIDIA is the worldwide leader in programmable graphics processor technologies. NVIDIA has long been known for innovative design methodologies driven by the extreme complexity, pricing pressure and short product life of 3D PC accelerators. NVIDIA's goals were straightforward but demanding: meet aggressive timing in several critical path math functions while keeping the die area small, using a cell-based design methodology.
Solution
Creating a customized and affordable solution by combining CellMath IP and expert datapath design services, Arithmatica was able to assist NVIDIA in meeting their high-performance goals while significantly reducing area and power consumption within their major silicon math blocks.
Results
NVIDIA was able to achieve a new threshold of performance in their next-generation of graphics processors while reducing the chip area dedicated to calculations in many of their major silicon math blocks, typically by 20 to 30 percent. Capitalizing on the time-to-market and differentiated product advantage of CellMath IP, NVIDIA was able to tape-out on time and within their specifications.
Client Perspective
"We worked closely with Arithmatica's engineering team and we were able to embed their CellMath IP in a variety of designs without any schedule impact. The result is that our new product line will be the highest performance GPU in the market today, and we are positioned to be very aggressive in terms of expanding our market leadership. It is refreshing to work with a young company that delivers on its commitments and we look forward to a long working relationship with Arithmatica."
Gopal Solanki,
Senior Vice President,
Platform Business Unit,
NVIDIA
“This is the biggest generation-to-generation performance leap that we have ever seen with a new graphics processing unit.”
Jen-Hsun Huang,
President and CEO,
NVIDIA
