Gaming PC Benchmarking Guide February 2011

If you are like me then you probably do not have the latest gaming PC out there. Even if you built a new machine it will probably have at least one obsolete part easily within a month or two. Because of this dilemma I have created the following gaming PC benchmarking criteria, which has some modern games and game engines as well as some older ones which still scale well.

A problem a benchmarker will face, especially when comparing an older machine with a newer one is that sometimes the older machine will not be able to run whatever game or benchmark as opposed to the new machine. Not necessarily saying the whole program won’t run but saying that it won’t run at the exact settings that the program runs on the faster, newer machine. Some settings will simply never run because the GPU will never ever have the ability to render those settings as it’s limited at a hardware level.

The specifications for my main gaming PC which is now old are:

OS: Windows XP Professional (Corporate)
Motherboard: Gigabyte GA-MA770-DS3
CPU: AMD Athlon 64 X2 5400+ 2.8 Ghz
Video Card: Sapphire ATI 3870 512 MB
Memory: G. Skill 4 GB DDR2 800 Mhz (limited to 3.4 GB by the 32-bit OS)
Sound Card: Creative Sound Blaster Live Platinum
Storage: Western Digital 750 GB 7200 RPM HD
Case: generic black case from newegg.com
Power Supply: generic 480 Watt
Peripherals: NEC DVD burner

My concern was to create a benchmarking guide that a normal person that is not running the latest hardware AND does not have an unlimited budget can use to test their system against ours and all the upcoming gaming PCs and parts we plan to review. How to do that? It’s rather simple. I sat there and hunted down many free games and demos that are currently available on the internet for download.

These benchmarks are run at a reasonable resolution that will yield great graphics while getting high frame-rates which a real gamer will use in a real world without risking to suffer lag in a (ranked) game.

As far as my picks go, I could have picked to run other games and just have FRAPS show my the framerates but FRAPS uses the hard drive a lot, especially to record and that would quickly become a bottleneck.

I will discuss why I picked those programs to benchmark now rather than some others which might be more popular. The list is the following:

The RealStorm Global Illumination Bench 2006 test has the following settings:

Demo: 1/5 Global Illumination Compare
Resolution: 1280×1024
Shadows: On
Reflections: On
Anti Alias: On
Depth of Field: On
Volume Lights: On
Radiosity: On

Download it from: http://www.realstorm.de/

This is the default benchmark option for this old benchmarking tool that is now discontinued but it will destroy the living shit out of any system out there. This benchmark uses straight-up RAW CPU processing power. It does NOT have multi-core support and it will simulate the max speed a single core will deliver in a system. Not every program has multi-core support and it’s still extremely important to have each core be as fast as possible. To a limited degree this benchmark does test the RAM as well but it’s mainly for the CPU. It stressed the living shit out of the system by making the CPU render everything, ignoring the GPU.

On my system, under the settings I listed above my computer yielded 2569 raymarks (the raw score used only by this benchmark), as well as 2.61 fps (frames per second) average, a minimum score of 1.7 fps, and a maximum score of 9.27 fps. Yes, that’s really really low but go ahead and run the benchmark on your own system. I hope you have good cooling, because you will need it! ;]

The Battleforge 1024×768 test has the following settings:

Shadow Quality: Very High
Resolution: 1024×768
Texture Quality: High
Fullscreen: On
Shader Quality: High
Anti-Aliasing: 8x
SSAO: Off
MultiThread Rendering: Auto-Detect
FX Quality: Very High
VSynch: Off
Cloud Shadows: Off
Glow: Off

Download from: http://www.battleforge.com/

Battleforge is a FANTASTIC free-mmo-rts that has kind of been abandoned by EA but it’s still free and many, many people still play it. I have mastered everything in the game and my friends are now all bored of it but I will play it once in a while. You can check my original review of the game here as well as check out my first strategy guide here for doing Battlegrounds as well as my second strategy guide for Battleforge here that shows you how to farm the mission Raven’s End by yourself.

Anyways, this full game is FREE and it includes a built in benchmarking tool. The way to use it is to login first to the game and then not login to your character, instead hit back, and select options, and go to the graphics screen and select to run the benchmark.

I picked this part of the test to run at 1024×768 because when I play the game competitively and most of the time, I run it only at this resolution to get the max amount of framerates and no lag.

At this resolution my system put out an average framerate of 9.2 fps, a minimum framerate of 3.4 fps, and a maximum framerate of 54.9 fps.

Again, if you never checked out the game, I encourage you to do so, especially if you are a massive RTS player. I like this game and benchmark because it taxes your CPU, RAM, and GPU. Every unit in the game moves and attacks in a complex way and it’s a great example of showing how well your system performs playing a real time war game with a ton of units.

The Battleforge 1280×1024 test has the following settings:

Shadow Quality: Very High
Resolution: 1280×1024
Texture Quality: High
Fullscreen: On
Shader Quality: High
Anti-Aliasing: 8x
SSAO: Off
MultiThread Rendering: Auto-Detect
FX Quality: Very High
VSynch: Off
Cloud Shadows: Off
Glow: Off

Download from: http://www.battleforge.com/

The same as above except with better eyecandy because of the higher resolution.

My average framerate was 6.8 fps, my minimum framerate was 0.5 fps, and my maximum framerate was 53.5 fps. Do you see now why I play it at a lower resolution? The game looks almost the same to me anyways, so might as well avoid lag!

The Dirt 2 1280×1024 max settings test has the following settings:

Resolution: 1280×1024
Refresh Rate: 60
Multisampling: 8x MSAA
Vsync: Off
Aspect Ratio: Normal
Gamma: 1.0
Night Lighting: High
Shadows: High
Particles: High
Mirrors: Ultra
Crowd: Ultra
Ground Cover: High
Drivers: Ultra
Distant Vehicles: Ultra
Objects: Ultra
Trees: Ultra
Vehicle Reflections: Ultra
Water: Ultra
Post Process: Ultra
Skidmarks: On
Ambient Occlusion: High
Cloth: High

Download from: http://www.codemasters.com/downloads/details.php?id=39424

In my opinion, you cannot get more intense for beating up your gaming system than playing a racing game OTHER than playing a real flight simulator game with all the options on. Think about how fast a system needs to render what’s going in a racing game, especially 200 MPH or higher being scaled realistically and you will see how these games are system killers.

Codemasters games are pure unadulturated eye candy, especially Dirt 2, GRID, and F1 2010. The games keep getting prettier and still run very efficiently despite the graphic quality increase. These games are also system killers because of how great they simulate the physics needed to create a realistic racing and driving experience.

Although I play it usually at a lower resolution, I tested it at 1280×1024 just to stay consistent with my future gaming PC reviews as well as my upcoming new gaming PC that I plan to buy this year so we can see the before and after results. My average framerate was 15.7 fps and my minimum framerate was 13.4 fps.

This IS the game that made me realize I needed to upgrade my machine to a newer system.

The demo (although I have the full game) includes a built in benchmarking tool so it’s a great test.

Tom Clancy’s HAWX 1280×1024 max settings test:

Screen Resolution: 1280×1024
Refresh Rate: 60 Hz
Antialiasing: 8x
VSync: Off
Full Screen: On
View Distance: High
Forest: High
Environment: High
Texture Quality: High
HDR: On
Engine Heat: On
DOF: On

Download from: http://www.bigdownload.com/games/tom-clancys-hawx/pc/tom-clancys-hawx-demo/

My friend Chris Mosso, which was my top Lieutenant in my massive guild in Auto Assault, always kept recommending for me to try out Tom Clancy’s HAWX saying it was an amazingly fun game and of course, he was completely right. I hadn’t had that much fun playing a game like that where you fly around fighting for your life since Descent: Freespace. The game is a permanent adrenaline rush and is simply total eye candy. I’ve shown this game to some friends of mine that are not gamers and it got them dizzy from just staring at the screen when I play.

Anyways, although this game is super eyecandy, it does have an AMAZINGLY efficient game engine. I like to include this game in the benchmarking guide because it represents for me what a well written game’s performance would be like with a typical game system. My system got an average framerate of 23 fps and a maximum framerate of 127 fps.

The fun demo includes the benchmarking tool, so it’s totally worth getting.

X3 Terran Conflict 1280×1024 max settings test

Resolution: 1280×1024 Fullscreen
Antialiasing: 8x
Anisotropic Texture Filtering: 16x
Texture Quality: High
Shader Quality: High
More Dynamic Light Sources: On
Ship Colour Variations: On

Download from: http://www.egosoft.com/download/x3tc/demos_en.php

My friend Ramiro became a huge fan of the X series after I lent and gave him a copy of X Beyond The Frontier many years ago. As kids, we grew up playing Elite 2: Frontier on my Amiga 600. We thought it was the most epic game every made and and I played it religiously exploring star systems and reading up on their composition and learning a lot about astronomy as well as doing missions nuking planets from space as well as other crazy things like assassinations. I even dabbled with Privateer 1 and 2 later on, but those games were much simpler than both Elite 1 and Elite 2: Frontier.

Anyways, although I played X: Beyond The Frontier a lot and a little of X2, I kind of gave up on that series, especially when I later got into playing Eve Online and found it kind of pointless to play a game like that single player. The people who make the X series have ALWAYS impressed the living hell out of me with how efficient their game engines are as well as how scalable, detailed, and completely beautiful they are.

This game engine will rock the socks of your CPU, RAM, and GPU. I couldn’t believe my eyes as to how detailed the final part of the benchmark was when I saw the massive, super-detailed space station being rendered as it was, on my old gaming PC.

Let’s see the numbers… My system only got a framerate of 17.012 fps in this benchmark but believe me, considering the settings and how beautiful everything looks, that was still higher than I expected it to be. Still, I would maybe play X3 when I get my next gaming machine.

Trackmania Nations

Resolution: 1280×1024
Antialiasing: 16 samples
Shadows: Complex
Shader Quality: PC3 High
Texture Quality: High
Max Filtering: Anisotropic 16x
Geometry Details: Normal
PostProcess FXs: On
Force Dynamic Colors: On
Force Motion Blur: On
Force Bloom: On
Water Geometry: On
Stadium Water Geometry: On
Trees Always High Quality: On

Download from: http://trackmania.com/

Trackmania is still one of my favorite racing game series of all time. This benchmark is done using the game Trackmania Nations that has always been the free version of Trackmania and the one that most people in the world play. I highly recommend getting Trackmania United if you are serious about having FUN in playing a racing game MMO.

This game is a massive physics simulator and it has looked good right from the start. The game engine is probably even as efficient as probably the Unreal engine except that Epic Games doesn’t put out free games! At 1280×1024 my system puts out an average framerate of 31.8 fps.

Okay, so there’s the list. Yeah, you might say that who cares because my system is running Windows XP and therefore DirectX 9 but the way I see it, gaming is a lot like racing cars in the real world. You might run DX11 on your Windows 7 box but if you barely run stuff and my system gets higher framerates than yours, that’s pretty sad. Only real results matter in the real world! I say fuck it, compare apples to oranges. I just care if the system will be able to run a game 100% stable as well as with decent performance.

So that about wraps it up. I hope you use my February 2011 benchmarking guide to test out your system and post some results below as comments. I know my system is a 2007 average price gaming system but let’s see what my next PC yields! And let’s see what the gaming PCs I will review soon will show us. Will these brands defend their speed claims when being tested in the real world? Stay tuned!

Visits: 2350