El Dorado Hills, Calif. — August 6, 2013 — The Embedded Microprocessor Benchmark Consortium (EEMBC) today announced FPMark™, a new benchmark suite that tracks the performance of embedded processors with floating-point hardware units (FPU), an increasingly popular and necessary feature to support graphics, audio, motor control, and many other high-end processing tasks. Uniquely, FPMark contains single (32 bit) and double (64 bit) precision workloads, as well as a mixture of small to large data sets to support microcontrollers to high-end processors, respectively. The EEMBC FPMark allows users to evaluate FPU performance on the basis of consistent and controlled data, delivering honest, reliable, and unbiased metrics to serve the needs of processor vendors, compiler vendors, and system developers.
Using floating-point (FP) representation enables more accurate calculations of fractional values than fixed-point numbers (integers) because exponents allow the decimal point to shift. Moreover, floating- point math makes numerical computation much easier and many algorithms implemented with floating point take fewer cycles to execute than fixed-point code (assuming similar precision). To take advantage of this efficiency, many embedded processors include hardware floating-point units (FPUs) to support these higher levels of precision.
The EEMBC FPMark Suite uses 10 diverse kernels to generate 53 workloads, each of which self-verify to ensure correct execution of the benchmark. These workloads are built on the same infrastructure as EEMBC MultiBench™, allowing the user to launch multiple contexts and demonstrate multicore scalability, as well as greatly simplifying the effort required to port the benchmarks to bare metal or implementations running Linux. The kernels in FPMark include a mixture of general-purpose algorithms (such as Fast Fourier Transform, linear algebra, ArcTan, Fourier coefficients, Horner’s method, and Black Scholes) and complex algorithms (such as a neural network routine, a ray tracer, and an enhanced version of Livermore Loops)1.
“Until now, the industry has lacked a reliable, useful, and consistent floating-point benchmark. In the same way that EEMBC CoreMark® was intended to be a “better Dhrystone,” FPMark provides an extreme improvement over the easily manipulated Whetstone and Linpack,” said EEMBC president, Markus Levy. “The FPMark will expose and highlight the performance gains from innovations in FPU development in terms of real application performance.”
"Developing a reliable floating-point benchmark is a complex challenge – one that EEMBC overcame using many years of benchmark development experience. While many people have attempted to create a floating-point benchmark, most do not comprehend the extra effort required to ensure that the workload executes comparably regardless of compiler or hardware used,” said Linley Gwennap, president and principal analyst of The Linley Group. “For example, it’s important that the FPMark was constructed in such a way to support advanced compiler optimizations, but not at the expense of optimizing away work that must be done during the execution of the benchmark.”
Similar to EEMBC CoreMark, certified scores are not required for FPMark, but EEMBC will promote the use of certified scores for members to ensure high-quality results. Non-members may obtain the entire FPMark suite, including source code and documentation, for only $495 for a single user corporate license or $195 for academic licensing. Contact EEMBC for more details.
# # #
EEMBC, the Embedded Microprocessor Benchmark Consortium, founded in 1997, develops and certifies real-world benchmarks and benchmark scores. The processor benchmarks help designers select the right embedded processors and compilers for their systems. The system benchmarks help consumers and IT professionals select the best smartphones, tablets, and networking firewall appliances. Every processor submitted for EEMBC benchmarking is tested for parameters representing different workloads and capabilities in communications, networking, office automation, automotive/industrial, and embedded Java. System tests focus on two areas – smartphones/tablets and network appliances. With members including leading semiconductor, intellectual property, and compiler vendors, and leading OEMS, EEMBC establishes benchmark standards and provides certified benchmarking results through the EEMBC Technology Center. For more information, visit www.eembc.org.
EEMBC, CoreMark, and BrowsingBench are registered trademarks of the Embedded Microprocessor Benchmark Consortium. All other trademarks appearing herein are the property of their respective owners.
1 Additional Technical Information
· A Fast Fourier Transform takes any function and converts it to an equivalent set of sine waves. The analysis is important in applications such as audio, spectral analysis, and image compression.
· Horner’s Method is used to approximate the roots of a polynomial.
· The Linear Algebra benchmark, derived from Linpack, is useful for understanding balancing forces in structural engineering, converting between reference frames in relativity, solving differential equations, and understanding rotation and fluid flow, among other problems.
· ArcTan, also known as inverse trigonometric functions, calculates the angles of a right triangle by using the ratio of two sides of the triangle to calculate the angle between them.
· Fourier Coefficients is a numerical analysis routine for calculating series or representing a periodic function by a discrete sum of complex exponentials.
· Neural Net is a small but functional back-propagation neural net simulator; neural net algorithms are computer programs that can identify complex relationships among data.
· Black Scholes is a mathematical model developed to calculate the value of financial derivatives, such as stock options.
· Enhanced Livermore Loops, one of the more popular FP benchmarks, are loops of computer code, extracted from operational programs used at Lawrence Livermore Labs that test the computational capabilities of parallel hardware and compiled software.
· LU Decomposition – apps like solving linear equations or matrix inversion
· Ray tracer - technique for image generation by tracing light path through pixels in an image plane and simulating the effects of its encounters with virtual objects.