The FPBench™ working group is developing a comprehensive suite of Floating-Point benchmarks.
The FPBench™ benchmark will provide a standardized, industry-accepted method to measure floating-point performance
- Targets graphics, audio, motor control, and many other high-end processing tasks
- Supports embedded processors with floating-point hardware
- Utilizes pseudo-random input data that can be generated at run time to guarantee that the compiler cannot calculate the operations at compile time
- Will perform output data verification
- IEEE 754-2008 compliant verificaation with libm functions accurate to .55ulp

FPBench™ will benefit processor vendors, compiler developers, and system developers
- Evaluate FPU performance on the basis of consistent and controlled data
- Analyze the performance delta between single (32-bit) and double (64-bit) precision
Join the EEMBC FPBench™ working group to help ensure a meaningful and fair representation for your company’s products
- FPBench™ working group is actively looking for additional participants to further define and elaborate the FPBench™ benchmark suite
- Participation open to all EEMBC subcommittee and Board of Directors members

|