Page 6
Sandra - Synthetic Tests
SiSoftware's Sandra (the System ANalyser, Diagnostic and Reporting Assistant) is an information & diagnostic utility. It should provide most of the information (including undocumented) you need to know about your hardware, software and other devices whether hardware or software. Sandra provides similar level of information to Norton SI, Quarterdeck WinProbe/Manifest, etc. The Win32 version is 32-bit and comes in both ANSI (legacy for Windows 98/Me systems) and native Unicode (Windows NT4/200X/.Net) formats. The Win64 version is 64-bit and comes in native Unicode format.
Do note that all the SANDRA benchmarks are synthetic and thus may not tally with real-life performance. The latter stands for whatever your environment is, i.e. which applications you run with what amount of data and so on. It is up to you to decide whether what Sandra measures is what you want to measure.
- Download SANDRA (special Guru3D.com edition)
Below you can find the scores of Sandra starting with memory performance:
It will be quite difficult to understand what we present to you today. Interpreting data in the way we tested and what we can show you simply is hard to comprehend, especially with all the mathematic BIOS stuff timings and dividers. Memory tweaking and overclocking is close to science. Let me try to explain a little how to interpret the chart.
When you look at for example the first three XMS2-5400 results then we can tell you that our CPU at all times was running at 3600 MHz. The last XMS2-5400 results where it says 285 FSB x14 is a full system overclock. That's throughout our benchmark session you'll notice the maximum overall performance gain as not only the memory runs faster, the CPU is also clock higher (~4 GHz).
The same goes for the XMS2-6400 Pro scores, we tested in a similar way.
First of at that FSB our mainboard was at the very peak of it' capability. The clock generators started having a hard time and there were tiny variances with every score we took. What we did was we each ran the test five times and presented you the average. We are hitting a number over 8000 MB/sec. How on this earth is that possible? DDR2 can run asynchronously from the system bus, which means that it does not need to run at the frontside bus speed. In fact thanks to the ABIT mainboard we strapped the Nortbridge, it can run to 1066 MHz, which allows us to give the memory a 3:4 ratio of the FSB, which means that on a system running a standard 200MHz FSB the memory in fact would run at 267Mhz, or the standard PC2-4200 533 MHz DDR2 speed.
Now with the XMS2-5400 at supported timings we got: 285/3*4(x2)= the DDR2 running at 760 MHz, that's way beyond what is advertised, nice.
The the XMS2-6400 Pro was able to achieve a 288/3*4(x2)= 768 MHz frequency, but then my mainboard gave me the proverbial middle finger and pretty much I had reached the maximum potential overclock. The memory could have gone higher though, I'm positive of that.
PC Mark 2004PCMark 04 is the latest version of the popular PCMark series. PCMark04 is an application-based benchmark and a premium tool for measuring overall PC performance. It uses portions of real applications instead of including very large applications or using specifically created code. This allows PCMark04 to be a smaller installation as well as to report very accurate results. As far as possible, PCMark04 uses public domain applications whose source code can be freely examined by any user.
Info and download - Download!
These are synthetic benchmarks and synthetic tests for some reason can be weird. Look at the highest FSB and then compare the 1 GB and 2 GB kit with each other. Now the the XMS2-5400 should be faster compared to the the XMS2-6400 Pro when you look at the timings yet it isn't. My best bet is that PCMark seems to have a soft spot for 2 GB and prefers it over 1 GB and somehow includes that in the score. I ran this test a couple of times and the score was consistent.
