Testbed and Methods

Our selection of gaming tests was intended to match this low-end graphics card. New games are too slow on a card of the RADEON 9200 class or don’t launch at all. Some games produce visual artifacts.

We could only pick up a few competitors to the Stealth: the lowest-performance cards we had on our hands were GeForce FX 5600 Ultra (350/700MHz), ATI RADEON 9600 XT (500/600MHz) and S3 DeltaChrome S8 (300/600MHz). We also had a Gigabyte Maya II R9200 (250/400MHz, 128MB of graphics memory). We decided to transform the RADEON 9600 XT and the GeForce FX 5600 Ultra into a RADEON 9600 and a GeForce FX 5600, respectively, by reducing the clock rates of the cards (to 325/400MHz for the RADEON and to 325/550MHz for the GeForce). The frequencies of the S3 DeltaChrome S8 remained the same.

Of course, it would be logical to use RADEON 9600 SE or GeForce FX 5200 cards, but we didn’t had them at out disposal at the moment.

The testbed was configured as follows:

• CPU: AMD Athlon 64 3400+  (2.20GHz, 1MB L2);
• Mainboard: ASUS K8V Deluxe;
• RAM: OCZ DDR PC-3200 Platinum EB (2x512MB, ÑL2.5-3-2-8);
• HDD: Seagate 7200.7, Serial ATA-150, 8MB buffer;
• Audio: Creative Sound Blaster Audigy 2;
• Software: Microsoft Windows XP Pro SP1, DirectX 9.0b;
• Drivers:  ATI CATALYST 4.4, NVIDIA ForceWare 60.72 (52.16 for 3DMark03), S3 Graphics 6.14.10.1632.

We used the following games for benchmarking the cards:

First Person 3D shooters:

• RTCW: Enemy Territory;
• Call of Duty;
• Unreal Tournament 2003;
• Unreal Tournament 2004 Demo;
• Tron 2.0;
• Highly Anticipated DX9 Game 2.

Third person 3D shooters:

• Tom Clancy’s Splinter Cell;
• Prince of Persia: Sands of Time;
• Max Payne 2: The Fall of Max Payne.

Simulators:

• IL-2 Sturmovik: Forgotten Battles;
• X2: The Threat;

Real-time strategies:

• C&C Generals: Zero Hour.

Semi-synthetic benchmarks:

• Final Fantasy XI Official Benchmark 2;
• Aquamark3.

Synthetic benchmarks:

• Futuremark 3DMark03 build 340.

We ran the tests in 1024x768 and 1280x1024 resolutions because 800x600 is low by today’s standards, while 1600x1200 is too heavy on a card like the Stealth S110. That’s also the reason for us to avoid enabling full-screen antialiasing or anisotropic filtering. Graphics quality settings were at the maximum in every game, so when the Diamond Stealth S110 shows an indecently low speed, its owners will have to reduce the graphics quality to achieve a higher performance.