Performance in Windows XP
As usual, I will first run the synthetic benchmarks.
The SiSoftware 2007 suite features an updated enhanced-functionality interface, runs on three platforms (Win32 x86, Win64 x64, WinCE ARM), contains 13 tests and 34 informational modules, and supports a large range of devices thanks to the developer’s collaboration with Intel, AMD, ATI, SiS and VIA. The program is supported in six languages and has a free Lite version for personal and educational purposes. SiSoftware Sandra measures the overall performance of the system as well as that of each of its subsystems.
PCMark benchmarks the computer performance in office and office-related applications and also produces performance scores for the main subsystems (CPU, memory, graphical, and disk subsystem). PCMark 2005 carries on the tradition of complex benchmarks of the series and uses fragments of real-life applications as tests. This makes it somewhat more relevant for end-users as opposed to fully synthetic benchmarks. After running a series of 11 tests on the different components of the system, the program calculates an overall performance score in units called PCMarks. PCMark 2005 can check a computer out at processing HD video and encoding audio, and offers enhanced tests of the CPU and hard disk under multi-threaded load. The overall score is calculated by the formula: PCMark Score = 87 x (the geometric mean of the basic tests), where the geometric mean is calculated as (Result 1 x Result 2 x…) divided by the number of results.
The results are close to those of the older Napa Refresh platform. This could be expected because the CPU is the same, the higher FSB frequency being the only difference. The CPU performance is lowered in the battery mode due to the activation of the power-saving features implemented in the Merom core, from old SpeedStep to new FSB Frequency Switching. The CPU frequency drops below 1GHz, and the test results get much lower. As for the graphics subsystem, Graphics Media Accelerator X3100 looks preferable to GMA950. The memory subsystem behaves typically for the Santa Rosa platform: the memory performance is half lower in the battery mode. The Napa Refresh platform didn’t have such a serious performance hit because it didn’t have such aggressive power-saving technologies.
I didn’t benchmark the notebook in Business Winstone 2004 and Multimedia Content Creation Winstone 2004 because they wouldn’t start on the new platform.
SYSMark 2004 SE is intended to reveal a system’s performance under different types of load. It simulates a user who is solving practical tasks in a few popular applications. Multithreading is taken into account. The benchmark issues a few ratings that are indicative of the system performance under different loads. SYSMark 2004 SE is mainly positioned as a tool for testing desktop systems and includes applications that are not often run on mobile computers. That’s why the results for each test load are shown separately.
The 3D Creation script simulates a user who is rendering an image into a BMP-file in 3ds max 5.1 and is also working on web-pages in Dreamweaver MX. After these operations are done, a 3D animation is created in a vector graphics format.
SYSmark 2004 SE is sensitive to the CPU performance. The notebook delivers a typical result of a 1.8GHz Merom when powered from the mains. When powered by the battery, the Santa Rosa platform proves to be slower than the Napa Refresh with the same CPU due to its more advanced power-saving technologies. You’ll see a similar picture in almost all of the SYSMark tests.
The 2D Creation script simulates a user creating a video out of a few RAW-format fragments and audio tracks in Premier 6.5. Waiting for the operation to complete, the user is also modifying an image in Photoshop 7.01 and then saves it to the hard disk. When the video clip is ready, the user edits it and adds special effects in After Effects 5.5.
The mains result is for a CPU frequency of 1.8GHz; the battery result is for 0.8GHz.
The next test simulates the work routines of a professional web-master. The user unzips the content of a website while using Flash MX to open an exported 3D vector graphics clip. Then the user modifies it by including more pictures and optimizes it for faster animation. The resulting clip with special effects is compressed with Windows Media Encoder 9 to be broadcast via the Internet. Next, the website is compiled in Dreamweaver MX while the system is being scanned for viruses with VirusScan 7.0 in the background.
The results depend on the CPU frequency again, and this frequency is lowered in the batter mode due to the reduction of the FSB clock rate.
The next script simulates an ordinary user who’s receiving a letter with a .zip attachment in Outlook 2002. While the received files are being scanned for viruses with VirusScan 7.0, the user looks through his e-mail, enters some comments into the Outlook calendar, and then opens a corporate website and some documents with Internet Explorer 6.0.
This is the only test from this suite that does not depend directly on the CPU frequency and, accordingly, on the power mode. However, there is a huge difference in the notebook’s performance in the two modes because the Santa Rosa platform can lower the memory frequency, too.
In the Document Creation script the user is editing text in Word 2002 and is also using Dragon NaturallySpeaking to convert an audio file into a text document. This text document is then converted into PDF format with Acrobat 5.0.5. And finally, the document is employed in a PowerPoint 2002 presentation.
Again, the results depend on the frequencies.
The final script from SYSMark 2004 SE includes the following: the user opens a database in Access 2002 and creates a few queries. Documents are archived with WinZip 8.1. The results of the queries are exported into Excel 2002 and are used to construct a diagram.
The final SYSMark 2004 test agrees with the previous ones.