Print
For a different view of computer performance, we also rely on Fritz Chess Benchmark and ScienceMark 2.0. Fritz Chess Benchmark is based on the popular Chessbase engine, so it’s considered real-world (although slightly out of date for the free benchmark.) Running it is straightforward: simply fire up the app and click start. You should note Fritz will indicate how many “processors” it’s going to use. The number will include all of the physical cores in your chip as well as any virtual cores. The result gives you the performance of your machine versus a 1GHz Pentium III as well as how many kilo nodes per second it can compute. A kilo node per second is how many moves per second are being computed. The benchmark ships free with copies of Fritz 9 or can be found online.
ScienceMark 2.0 is another synthetic that’s rooted in a real-world engine. It uses mathematical algorithms common in scientific and engineering applications and also stresses memory performance and latency. The caveat to this benchmark is that it doesn’t seem to be particularly multithreaded. And back in the days when Intel Pentium 4 would get soundly splattered by the Athlon 64 in ScienceMark 2.0, the company would grouse that the authors of ScienceMark 2.0 weren’t interested in working with Intel in addressing optimizations for its CPUs. But with Intel taking the lead with Core 2 and Core i7, Intel doesn’t seem to object to this test as much anymore. Running it is easy. Download the installation file and decompress it. Execute the file and click on File, Run All Benchmarks. The results will give you an overall ScienceMark score as well as subscores for molecular dynamics, cryptography, memory, among other benchmark scores.
Although these are single-threaded, two popular benchmarks are great for calculating the math prowess of a CPU: Prime95 and SuperPi. Both have actually long been favored by overclockers as stress tests but both also will give you overall scores as a performance indicator. The weakness with both tests, obviously, is the lack of multi-threading. The preferred version of SuperPi is 1.5 and has been modified by XtremeSystems.org to make it more amenable to stress testing. It’s available here. To run it, execute the app and select Calculate and select 1M and the program will calculate pi to one million digits.
Like SuperPi, Prime95 is considered more of a stress test than a benchmark. In fact, we use a custom blend of Prime95 developed by an OEM PC builder to stress test many of the overclocked PCs we review. Prime95 is a distributed project used to search for Mersenne prime numbers. To run the benchmark, first download it from www.mersenne.org. Start the application, dismiss the stress testing screen, and go to Options Run Benchmark. When it’s complete, the results are dumped into a file named results.txt that should be in the same folder where the executable resides. Open the file and you should find results for each separate run you conducted reported in milliseconds. You can compare the results to others at: http://v5www.mersenne.org/report_benchmarks/.
Comments are closed on this article
Philips
October 25, 2011 at 6:12am
Thank you for the casualty of providing all of these information. This is a lot actually. These are very helpful.
Azrael808
February 16, 2011 at 8:13am
This is a great article; I'm currently using this as a basis to benchmark my home machine so I can see how it stacks up against other rigs, but mainly so I can see how much an upgrade improves the overall system. Some useful comments too! :)
I did have one question: when taking an average of the benchmark runs, why do you use the median, and not the mean score? Is it simply because a mean score will most likely produce a result that wasn't actually obtained by a run of a utility?
jlfrank83
January 27, 2010 at 8:12am
The article states that real-world performance is what you are shooting for when you run your benchmarks, Yet they also want you to disable antivirus, disconnect from your network, manually launch idle process, etc, before benchmarking. Isn't creating a "virtual clean room" tainting your results as much as a synthetic benchmark application or biased video driver would? I don't go through all these preflight steps before playing a game or running Photoshop, so why would I do them before benchmarking and expect realistic, real-world results?
mrvander
January 27, 2010 at 8:40am
... one has to eliminate as many variables as possible in order to produce repeatable results. This is a basic of any scientific testing method - whether it be life sciences or computer sciences. Once you can produce a repeatable result - you have established a control (or a baseline.) THEN add in the other variables to determine the effect it has, preferably one at a time. Turn on your virus scanner. Repeat the testing and compare it to your control. Analyze the data. Add in another variable, say your network. Repeat and rinse.
This is also why they don't recommend using FRAPS and just running through a level of your favorite game. It's not accurately reproducable. It can give you a general idea over many runs, but it can vary wildly.
Of course you can customize this to your heart's content and maybe just a general idea is all you need. The article does mention the idea of "play more games" as opposed to just endlessly benchmarking one's PC and analyzing every last little detail to get that extra tenth of a framerate.
nekollx
January 26, 2010 at 6:06pm
Daz 3d is good real world test since the preview/maniptlation window is GPU bound but renders are CPU bound
------------------------------
Coming soon to Lulu.com --Tokusatsu Heroes--
Five teenagers, one alien ghost, a robot, and the fate of the world.
Modred189
January 26, 2010 at 5:55pm
No furmark?
It's one of the most stressing GPU-only benches I know of.
And I must say, real-world benchmarking your graphics card is the best way to go. Pick a game or two you know well (HL2 works well), and play through a level a few times, measuring your fps with fraps. Throw the data into excel and average the columns. You can even make nice graphs from multiple data sets and compare them to show what scenes your PC is best at and which it has issues with.
jlfrank83
January 27, 2010 at 8:19am
Your columns are going to be radically different if you are controlling the action yourself. It's better to use a game that has in-game, real time rendered cutscenes, like Mass Effect or Assassin's Creed. If you use a playable section of a game, your graphics card will be rendering differently every time, since each playthrough is going to vary in length, FOV, etc.
Modred189
January 27, 2010 at 5:54pm
I have done this a few times before, and if you play through the level a few times first, you can get it relatively reproducible. The problem with in-game benchmarks is that they disable a lot of things like AI, which is a big CPU drain.
bikerbub
January 27, 2010 at 10:39am
GTA IV also has a benchmark option. it's not very long, but it is an actively rendered test with AI and lots of gunfire. also somewhat humorous, as you are a clown (if i remember correctly) riding on a moped, through the streets, weaving in between the Liberty City skyscrapers.
