Cadalyst - July 2008 - (Page 19)

cadalystlabsreview What We Requested Earlier this year, Cadalyst sent invitations to several vendors requesting mobile systems that had the fastest available processors — be they from AMD or Intel. We requested systems that had at least 2 GB of onboard RAM, at least 60 GB of total hard drive storage space, and a CD-RW optical drive. Submitted systems had to have Windows XP Professional preinstalled. No additional accessories were requested, and although some mobile workstations have provisions for accommodating a second battery, only a single battery was installed in each. Cadalyst received three systems, all based on Intel processors — one each from @Xi Computer, Lenovo, and Polywell. Hewlett-Packard usually sends a system for this annual review, but the company is about to refresh its lineup, and the new systems weren’t ready for evaluation. Dell, also a regular contributor to this roundup review, was going thorough personnel changes and wasn’t able to send a system in time for evaluation (Cadalyst will, however, be looking at one of Dell’s high-end 64 bit notebook systems in an upcoming standalone First Look review). The evolution of battery technology has made it possible to work longer and more productively, even with applications as demanding as CAD, DCC, and engineering. native OpenGL driver, the second using the native Direct 3D driver, and the third using any available accelerated driver. The final benchmark series was a run of the full SPECviewperf 10 benchmark (www.spec.org). All the results for the various viewsets are listed individually in the online feature table (www.cadalyst.com/0708laptop-table). The various subsets of this particular benchmark give performance indications for a variety of commonly used design applications. When all tests were completed, I calculated the results and generated a total index score based on the following criteria: The Cadalyst C2008 benchmark received a 3x weight factor for the total weighted index, the MAXBench 4 tests results received a 1x weight factor, and the SPECviewperf test results received a 1x weight factor for the total weighted index. These numbers are all reflected in the single performance score, which carries a weight of 7x in the overall ratings. Pricing for the system carries a weight of 4x in the overall ratings, and battery life has a weight of 3x. The system warranty and the vendor return policy each carry a weight of 1x in the overall ratings. All mobile workstations included in this roundup were evaluated on meeting the minimum system requirements. Testing Procedures I test notebook computers the same way I test conventional desktop workstations, using the same software and benchmarks but adding battery tests to determine how long these systems can be used in the field. All benchmarks are run at a screen resolution of 1,280 x 1,024, although most are capable of higher resolutions these days. Typically, I do spot tests at the native resolution as well, but I haven’t included the testing figures in the online feature table (www.cadalyst.com/0708laptop-table). I ran the Cadalyst C2008 benchmark using AutoCAD 2008 on all systems. As we go to press, AutoCAD 2009 is becoming available, and I expect Cadalyst will be using it for the next series of system tests. All testing was done using with the native AutoCAD drivers, both OpenGL and Direct 3D. All tests were conducted under Windows XP Professional with Service Pack 2 installed. Where the base video driver for Windows allowed optimization settings for AutoCAD, I applied and used those settings for the benchmark tests. The 3ds Max benchmark, MAXBench4, was run under Autodesk 3ds Max 2008. 3ds Max 2009 is nearing release but was not available when I began the benchmark tests for this article. Where the option was available, I set the baselevel Windows driver to use the optimized settings for 3ds Max, and if an accelerated driver for the application was available, I also tested with that driver. This typically resulted in three benchmark series being run — the first using the Battery Life The practicality of using a mobile workstation is dependent upon both its ability to run major design applications at a reasonable speed and to run them for long enough to get a significant amount of work done, so battery life is an important concern for these mobile systems. We typically run two types of battery life tests to get some basic figures about how long batteries actually last. Each system started with a battery that indicated a 100% charge. The first battery test was a simple test, with the tested computer being idle and allowed to run down to hibernation. The second test was a rundown test with the Cadalyst C2008 benchmark running in a continuous loop until the system went into hibernation. Quad-Core Systems Report Card Product @Xi Computer Xi PowerGo Duo XT, p. 20 Lenovo ThinkPad T61p, p. 22 Polywell Computers PolyNote C719ND, p. 24 Performance (7x) A+ B– A Price (4x) A A+ A– Battery Life (3x) A A+ A Warranty (1x) A A A+ Return Policy (1x) A+ A– A– Overall Grade A+ B+ A July 2008 cadalyst www.cadalyst.com 19 http://www.spec.org http://www.cadalyst.com/0708laptop-table http://www.cadalyst.com/0708laptop-table http://www.cadalyst.com

Table of Contents Feed for the Digital Edition of Cadalyst - July 2008

Cadalyst - July 2008 Editor’s Window CAD Central FIRST Robotic Beasts Draw Students to Engineering Power! Speed! Action! — Mighty Mobile Workstations are Packed and Ready for Adventure! Vico Virtual Construction Suite 2008 AMD ATI FireGL V8600 and FireGL V8650 — 1-GB and 2-GB Graphics Cards Gaining Independence and Influence The Current State of MCAD Rewriting the Rules of PDM BIM Goes Residential CAD Cartoon Issue Indexes Pleasures of Customization

Cadalyst - July 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.