EEMBC Delivers 64-Bit Android Platform Benchmark
AndEBench™-Pro 2015 Includes Support for Android 5.0, Off-Screen Rendering, and Other Functions
EL DORADO HILLS, CA-- Apr 22, 2015 - The Embedded Microprocessor Benchmark Consortium (EEMBC) today announced the availability of EEMBC AndEBench-Pro 2015 on Google Play™, a 64-bit enhanced version of the highly anticipated Android device benchmark. The AndEBench-Pro 2015 upgrade provides professional benchmarking that allows vendors and consumers to reliably assess performance. The 2015 version of AndEBench-Pro adds native support for 64-bit application processors while maintaining compatibility with 32-bit processors. AndEBench-Pro 2015 also includes off-screen rendering with checkpoints for the 3D graphics component of the benchmark to avoid screen-refresh delays.
Similar to the original AndEBench-Pro, AndEBench-Pro 2015 focuses on the key metrics that reflect the subsystems and usage models of applications. Hardware tests in the benchmark thoroughly exercise the central- and graphics-processing units (CPU and GPU), memory, and storage subsystems using carefully designed, tough-to-trick algorithms. The complete Platform test includes common tasks such as GUI rendering, XML parsing, image manipulation, data compression, and cryptography, all combined into real-world scenarios that truly provide meaningful comparisons.
Upgrading to 64-bit benchmarks can potentially benefit all native portions of AndEBench-Pro 2015, including CoreMark-Pro (the CPU-centric test within the benchmark), memory and storage bandwidth, and the 3D test. Furthermore, the Platform portion of the benchmark uses Android platform APIs, so these tests could also show some benefits from 64-bit processing, although the value is highly device dependent.
The 3D score of AndEBench-Pro 2015, indicated as frames/second, represents the performance of a device's GPU working in conjunction with the CPU to render a 3D game-like scene. For this portion of the benchmark, AndEBench-Pro 2015 includes a custom-designed Zombie scene that a good GPU will render smooth and effectively animated. Within the benchmark, the 3D test runs twice. While both tests render off-screen, the first run renders off screen, and only displays periodic checkpoints in a grid pattern, removing a bottleneck for higher performance GPUs. In the second run, the test renders to a fixed resolution and then scales to device resolution for visual verification (this is the part that makes the benchmark look cool). Although the 3D test is aimed at testing the system as a whole, screen performance and refresh rates are less important in the context of benchmarking the GPU performance, and thus only the first run (off-screen rendering without screen refresh) is counted into the final score because in this mode, screen capabilities and refresh rates are ignored.
"Performance analysis on an Android device is a complex undertaking. The possible hardware and software components present a wide range of variables within any device and any meaningful benchmark must comprehend these variables. Hence, we found it imperative to upgrade AndEBench-Pro to support 64-bit processors and use off-screen rendering for the performance result, despite the lack of 'eye candy' that so many graphics benchmarks deliver," said Markus Levy, EEMBC's President. "As with all our benchmarks, the expertise that has gone into developing AndEBench-Pro 2015 and the transparent availability of source code ensures that the benchmark is structurally sound and serves the industry's demanding needs."
With support for both English and Chinese languages, the AndEBench-Pro 2015 can be downloaded for free from Google Play™ and the Amazon™ Appstore for Android. Professional reviewers should contact EEMBC directly to obtain their specific version of AndEBench-Pro 2015 that will enable them to change benchmarking parameters and gain access to even more detailed scoring information that is not disclosed as part of the standard benchmark to aid in their analysis.
EEMBC encourages all vendors and manufacturers to join the consortium's working group to contribute to the definition and development of our next generation AndEBench. To join the working group, contact Markus Levy for details.
About EEMBC
EEMBC, the Embedded Microprocessor Benchmark Consortium, develops industry-standard benchmarks to test embedded processors and systems such as smart phones and network firewall appliances. EEMBC's benchmark development work is supported by yearly member dues and license fees.
EEMBC members include AMD, Analog Devices, Andes Technology, ARM, Atmel, Avago Technologies, Broadcom, C-Sky Microsystems Co, Cavium, Cypress Semiconductor, Dell, Freescale Semiconductor, Green Hills Software, IAR Systems, Imagination Technologies, Infineon Technologies, Intel, Lockheed Martin, Marvell Semiconductor, MediaTek, Mentor Embedded, Microchip Technology, Nokia, NVIDIA, NXP Semiconductors, Qualcomm, Realtek Semiconductor, Red Hat, Renesas Electronics, Samsung Electronics, Silicon Labs, Somnium Technologies, Sony Computer Entertainment, STMicroelectronics, Synopsys, Texas Instruments, Tilera, TOPS Systems, and Wind River Systems.
|
Related News
- Tensilica's Xtensa LX Processor Beats All Other 32- and 64-bit Processor Cores On EEMBC Consumer "Out of the Box" Benchmark
- EEMBC Publishes Benchmark Scores for Toshiba’s TMPR4927 MIPS-Based 64-Bit RISC Processor
- Micro Magic, Inc. Delivers Ultra Low Power 64-Bit RISC-V Core
- Andes Technology Features 32-bit A25MP and 64-bit AX25MP RISC-V Multicore Processors With Andes Custom Extension at TSMC 2019 Open Innovation Platform Ecosystem Forum
- Wave Computing Unveils New Licensable 64-Bit AI IP Platform to Enable High-Speed Inferencing and Training in Edge Applications
Breaking News
- TSMC drives A16, 3D process technology
- Frontgrade Gaisler Unveils GR716B, a New Standard in Space-Grade Microcontrollers
- Blueshift Memory launches BlueFive processor, accelerating computation by up to 50 times and saving up to 65% energy
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
Most Popular
- Cadence Unveils Arm-Based System Chiplet
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
- Esperanto Technologies and NEC Cooperate on Initiative to Advance Next Generation RISC-V Chips and Software Solutions for HPC
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- Arteris Selected by GigaDevice for Development in Next-Generation Automotive SoC With Enhanced FuSa Standards
E-mail This Article | Printer-Friendly Page |