benchmark News - Page 49 of 51 - CNX Software

July 13, 2012July 13, 2012 by Jean-Luc Aufranc (CNXSoft) - No Comments

ACE CoSy Compiler Framework Outperforms LLVM by up to 25% for ARM9 processors

ACE (Associated Compiler Experts) announced that their 2012 CoSy compiler development system delivers better performance than the latest LLVM 3.0 compiler on ARM9 processors. Using an ARM9 processor as reference, CoSy compiler framework (2012) shows more than 15% performance improvement using Livermore benchmark loops and 25% in MiBench against LLVM 3.0. The CoSy compiler framework is also more than 7% ahead of LLVM on the EEMBC CoreMark benchmark. The company explains this feat by the use of “CoSy’s unique flexible phase ordering of cutting-edge code optimizations and the addition of new CoSy features” such as compile-time code generator feedback injected into optimization algorithms in order to augment realistic and accurate decision making in architecture-independent optimizations. ACE did not publish the benchmark results publicly, and I could not find any reference to “cosy” or “ace” in EEMBC Coremark database, so I could not check those results myself. CoSy Compiler Framework is […]

June 25, 2012June 25, 2012 by Jean-Luc Aufranc (CNXSoft) - 11 Comments

Monkey, an Open Source High Performance Embedded Web Server

Some time ago, I mentioned 5 web servers (mathopd, thttpd, busybox httpd, boa and lighttpd) suitable for embedded systems (including those featuring no MMU processors) and low end machines. I’ve recently come across Monkey web server, a lightweight open source Web Server for Linux (2.6.29 or greater), which has been designed with focus in embedded devices. Monkey is currently supported on ARM, x86 and x64, although a quick analysis of the source code shows it forks, so it won’t be supported on processors that do not feature a memory management unit (MMU) without modifications. Monkey supports the following features: HTTP/1.1 compliant Virtual Hosts Asynchronous networking model (event-driven) Indented configuration Plugins Support C API Interface Other features through base plugins: SSL Security Log writter Directory Listing Shell: Command line The developers have benchmarked Monkey against busybox httpd and nginx on a now well-know ARMv6 platform: the Raspberry Pi. They used Siege […]

June 21, 2012 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

Phoronix Designed a Solar Powered ARM Cluster with 48 Pandaboards

Michael Larabel of Phoronix did a “little” week-end project at the Massachusetts Institute of Technology (MIT), building an ARM cluster with 48 Pandaboards mostly with standard Pandaboards (OMAP 4430), although some Pandaboard ES (OMAP 4460) were also thrown in the mix. The 96-core cluster consumes just over 200 Watts under load and can be powered by a solar panel. The boards are stacked up vertically in an industrial trashcan, and are powered via USB instead of AC adapters. Michael’s PandaBoard ES testing indicates about a 3 Watt idle per board, 5 Watt under load, or 6 Watts under extreme load. The cluster idle power consumption is just under 170 Watts while idle, and goes slightly over 200 Watts under load. Each Pandaboard runs Ubuntu 12.04 ARM OMAP4 installed on SD cards. MIT did some testing and provided Phoronix with benchmark results such as tests following Green500 benchmarking procedure, but they are […]

June 19, 2012July 17, 2012 by Jean-Luc Aufranc (CNXSoft) - 7 Comments

ARM vs x86 Servers Benchmark – Calxeda EnergyCore ECX-1000 vs Intel Xeon E3-1240

Calxeda has released the results of ApacheBench benchmark comparing their ARM-based EnergyCore solution to an Intel Xeon server in order to showcase the performance and the much lower power consumption of their servers. Here’s the setup: Hardware: Single Calxeda EnergyCore ECX-1000 @ 1.1 GHz, 4 GB of DDR3L-1066 memory, 1Gb Ethernet network port and 250 GB SATA 7200rpm HDD Intel Xeon E3-1240 @ 3.3 GHz, 16 GB memory and 1Gb Ethernet network port. No info on hard drive provided Software: Ubuntu Server v12.04 Apache Server v2.4.2 ApacheBench v2.3 (16k request size) They performed power measurements every 2 seconds and averaged the results. Power supply overhead and hard drive power consumption were not excluded in the measurement, but the entire SoC and DDR memory power consumption were included together. For the Intel server however, they could not measure directly, so they used published TDP values for the CPU (80 W) and I/O […]

June 15, 2012May 25, 2014 by Jean-Luc Aufranc (CNXSoft) - 129 Comments

WM8850-MID Android 4.0 Tablet Unboxing and Review

I recently bought an Android 4.0 tablet with a 7″ capacitive touchscreen based on the new WonderMedia WM8850 processor (Cortex A9 + Mali-400). In my first post, I called this tablet UGoos UG-LMW70, but I just received a no brand package with nothing mentioning Ugoos or the model number UG-LMW70 anywhere. Android settings just show WW8850-mid for the model number. The tablet comes with a 5V power supply, a miniUSB to USB cable, a short miniUSB to USB female adapter, headphones and a user manual in English explaining how to use Android, but nothing about the tablet itself. The tablet has vol-/+ buttons and a power button (left side below), and a power jack, microSD slot, headphones jack, a USB host port, a mini HDMI port, a miniUSB port and a microphone (right to left). They did not include a mini HDMI to HDMI cable, so I did not try HDMI […]

June 7, 2012 by Jean-Luc Aufranc (CNXSoft) - No Comments

IAR Systems Releases Embedded Workbench for ARM Version 6.40

Earlier today, IAR Systems has announced the release of version 6.40 of Embedded Workbench for ARM. This new version introduces several new features, enhancements and optimizations. The enhancements include a new source browser and text editor, with functionality such as auto-completion, code folding, block selection, block indentation, bracket matching, and zooming. Word/paragraph navigation have also been improved. IAR Systems also enhanced the compiler with improvements to the stack usage analysis functionality that provides calculations of the maximum stack depth for each call graph root and new features such as support for C++ source code and recursion. A new linker directive check can be used to calculate the stack usage at link time to verify that the used stack space does not exceed the allocated memory. The inline assembler has also been improved and expanded with a large number of new operand constraints and modifiers. In a previous post “Green Hills […]

June 4, 2012June 4, 2012 by Jean-Luc Aufranc (CNXSoft) - No Comments

Texas Instruments Showcases OMAP 5 GPU Capabilities with GLBenchmark 2.5

In February, Texas Instruments compared the browser rendering performance of its OMAP 5 platform to the Nvidia Tegra 3, and the benchmark results showed OMAP 5 was twice as fast @ 800 MHz than the Tegra 3 @ 1.3 GHz. Today, they’ve uploaded a video showing the 3D performance a tablet based on OMAP 5 (and PowerVR SGX 544-MP GPU) against a market-leading tablet (I guess it’s the Apple iPad 3 with PowerVR SGX 543 GPU) using GLBenchmark 2.5 with the following setup: 1080p resolution. On-screen and off-screen tests were conducted to compensate for the “Vysnc Limited” scenarios. (24bpp) 16bit depth tests to better represent high-end game content Power saving mode disabled and device connected to a power source. No background tasks and airplane mode “on.” The on-screen results show 38 fps for the OMAP 5 vs 34 fps for the other platform, or a 12% performance boost. Off-screen results […]

June 3, 2012June 18, 2018 by Jean-Luc Aufranc (CNXSoft) - 30 Comments

Linaro Android Puts Stock Android To Shame on TI Pandaboard (OMAP4430)

The team at Linaro has done an amazing job at optimizing Android 4.0 for ARM and Bernhard Rosenkränzer, Android Engineer at Linaro among other things, has put all those optimizations together to showcase a demo at Linaro Connect Q2.2012 in Hong Kong with 2 pandaboards: One running Stock Android 4.0.4, the one released by Google (AOSP) One running Linaro Android 4.0.4 Both hardware, android version and benchmark software (oxBench) are the same, and the results are quite amazing with Android Linaro achieving about 60 fps in all 0xBenchmark tests (OpenGL Cube, OpenGL Blending, OpenGL Fog and Flying Teapot) whereas Android stock achieving 30 fps. They selected a benchmark tool that is mainly CPU bound, as they cannot optimize the GPU code since they can only access binary blobs. Apparently, most of the improvements were possible thanks to toolchain and code optimization (to be able to build) such as using gcc […]