GeForce GTX 780 Ti Round Up: EVGA, Gigabyte, MSI
Test System and 3DMark Benchmarks
How We Configured Our Test Systems: We tested the graphics cards in this article on a Gigabyte GA-Z68X-UD3H-B3 motherboard powered by an Intel Core i5 3570K quad-core processor and 16GB of Corsair Vengeance DDR3-1866 RAM. We installed Windows 8.1 Enterprise 64-bit onto a pair of Intel 730 Series 480GB solid state drives (SSDs) configured in a RAID 0 array. Once that was complete, we fully updated the OS and installed the latest DirectX redistributable along with all of the drivers, games, and benchmark tools necessary to complete our tests. Since we're also interested in 4K gaming performance, we ran the tests on a Dell UltraSharp UP3214Q 4K Ultra HD monitor.
|
|
|
 |
|
 3DMark 11
|
Although Futuremark’s 3DMark 11 has been around for several years, it still provides a good look at gaming capabilities. We ran this benchmark on the Performance preset at 1280 x 720 resolution at on the Extreme Preset at 1920x1080. |
|
|
|
|
|
Fire Strike has two benchmark modes: Normal mode runs in 1920x1080, while Extreme mode targets 2560x1440. GPU target frame buffer utilization for normal mode is 1GB and the benchmark uses tessellation, ambient occlusion, volume illumination, and a medium-quality depth of field filter. The more taxing Extreme mode targets 1.5GB of frame buffer memory and increases detail levels across the board. Extreme mode is explicitly designed for CrossFire / SLI systems. GT 1 focuses on geometry and illumination, with over 100 shadow casting spot lights, 140 non-shadow casting point lights, and 3.9 million vertices calculated for tessellation per frame. And 80 million pixels are processed per frame. GT2 emphasizes particles and GPU simulations. Tessellation volume is reduced to 2.6 million vertices and the number of pixels processed per frame rises to 170 million. |
Futuremark's newer 3DMark Fire Strike test reiterated what we saw in 3DMark 11. The overall scores tend to exaggerate the situation, which is why we like to post the frames per second results as well. In doing so, we can see that just a few FPS separate the cards.
That said, it's again odd to see MSI's card lag behind. As you'll see on the following pages, this is a theme throughout the roundup. We tripled checked the sensors to make sure the card was configured to run in OC mode, and then further verified this by benching the card in Gaming Mode and Silent Mode, which produced gradually lower scores (since both presets are slower).
Temps were not an issue with MSI's card, so it's likely the additional power afforded by the dual 8-pin feeds on the EVGA card allow it to maintain higher boost clocks for longer periods of time.
