FPSLabs : Introductory Gotfrag Hardware Benchmark

Posted by in October 30th, 2005

Posted in: Systems

In order to better understand how we’re different than the other benchmarking sites, we’ve come up with this introductory benchmark to show you how our tests are going to be run.

Welcome to the first ever quantitative benchmark in GotFrag’s history. The purpose of this initial benchmark isn’t to show you any direct comparison, since I’m only benchmarking my own personal system. The point of this article is to introduce you to how we’re going to be running these benchmarks, and how we’ve changed them from the average every-day hardware website benchmark to better suit your needs as a gamer. First off, let’s get to the test bed:

AMD Athlon 3500+ Processor, 2.2GHz, 512KB cache, Newcastle Core
Visontek X800 Pro Graphics Card
Abit AV8 Motherboard, K8T800 Pro chipset
Sound Blaster Audigy2
2×512MB Kingston Value Ram 3-3-3-8 (dual channel)
2×160GB Maxtor Diamondmax 9 (raid-0)
NEC 2510A DL DVD-RW

Due to limitations of game availability, we were only able to test four games in this introductory article, but it will more than illustrate the point we’re getting at with these benchmarks. The games tested are Call of Duty 2 (demo), Quake 4, and Counter-Strike: Source, and Counter-Strike.

We have decided to go against the grain of normal hardware sites and test all these games at lower resolutions, because we know that that is what the majority of you use. We’ve decided on 800×600, 1024×768, and 1280×1024. The reason for the odd-sounding 1280×1024 resolution is that it is the native resolution of many LCD screens out there, and thus the resolution that games would look best on for many people. Another way we’re changing up the rules is with background programs. Normally, a website would disable all programs before testing to get the maximum result. However, gamers don’t play with only one program running. So, all tests are done with mIRC, Xfire, and Ventrilo running in the background.

All tests were both done in “pure” mode, which means no eye candy like AA or AF, and there was also a batch done in 6xAA/16xAF mode. For a recap on what this means, check our earlier article on AA and AF.
{mospagebreak heading=Introduction&title=Counter-Strike: Source, Call of Duty 2}

Counter-Strike:Source
Our first game is Counter-Strike: Source. This game, as the title would imply, runs off of valve’s HL2 engine called “source.” The engine was designed from the ground up to be built around ATi cards, so the performance should be optimal in this game.

Counter-Strike Source

As the benches show, at 1280 resolution and maximum eye candy, cs:s still provides you with playable frame rates on the X800 Pro. With eye candy turned off, the video card performs so well, that the bottleneck in the frame rate is actually the CPU, which being a 3500+, is no slouch. So if you’re looking for 100 FPS constant in cs:s, look no further than a $200 X800 Pro, but you may want to invest in a more powerful processor.

Call of Duty 2
The second game tested was Call of Duty 2. As opposed to the first CoD, this game uses a completely new engine, built from the ground up and unique to this game. It runs on Direct3D, which theoretically should be good for ATi cards, but it also has its roots in the Quake 3 engine, which has been ATi’s nemesis for years. Let’s look at the results.

Call of Duty 2

As one can see, CoD2 is one heck of a stressful game for graphics cards. The same setup that could chew up and spit out CS:S struggles severely on CoD2. In fact, the highest possible quality setting in CS:S gets you about the same FPS as the lowest possible setting in CoD2 that we tested! Time will tell if this is simply an issue of optimizing a driver for a new engine, or if it really is that hellacious on your GPU.
{mospagebreak title=Quake 4, Counter-Strike 1.6}

Quake IV
Our third game is Quake 4, which, with CPL backing, is most likely going to be the “next big thing” when it comes to 1v1 games. It runs on a very slightly modified Doom 3 engine, and looks very similar to it in many ways. Being the same Doom 3 engine, it also doesn’t take too kindly to ATi cards. Let’s see how the X800 Pro did.

Quake 4

As you can see, our X800 Pro didn’t fair all to well against the demands of the advanced Doom 3 engine. With an average FPS even lower than CoD2, this is the lowest scoring game of our roundup. One interesting note is that the 800 and 1024 resolution tests are nearly identical, leading one to believe that the 800 resolution is CPU rather than GPU limited. However, if you’re planning on playing Quake 4 competitively at high quality settings, you’re looking at an investment of a 7-series GeForce at least.

Counter-Strike 1.6
Our final game is the venerable old counter-strike. Seeing as the engine is based off of the Quake 2 engine and is over 8 years old, we should be able to assume that nearly every test will be CPU limited.

Counter-Strike 1.6

Evidentially, our suspicions were right. Any setting we tested didn’t matter; we would always get the same 200ish average FPS. So no worries to any of you out there who are considering an Athlon64 system; you WILL get 100FPS for most of the game.

This has been our very first GotFrag Hardware benchmark. We’re still working out the kinks of this operation, so any feedback you can give us on what you would like to see would be greatly appreciated. Thanks ahead of time, and welcome to the only hardware site dedicated to people like us.

Popularity: 7% [?]

Support FPSLabs! Share and Enjoy: