Hardware Test: Recommended System Configurations
After our initial set of disappointing test results for the bare-minimum setup, we simulated a graphics card update. To do this, we used an older GeForce 8800 GTS 512 MB so we could test how much of a limitation the CPU imposed on the game, and how play might improve by speeding up the processor clock. That’s why we also elected not to turn on PhysX, because we wanted to load the CPU as much as possible.
Test: Average Frame Rates vs. Clock Rate: Single Core vs. Dual Core
For this test, we used the CPU from the previous test (Low-End PC 1) to compare a similar dual-core PC (Low-End PC 2) that was also supposed to represent a configuration that met the minimum requirements. We varied the clock speeds from 2.0 GHz to 2.6 GHz, which involved underclocking as well as overclocking our test rigs, emulating a full range of processors. We tested the same three scenes as before, but this time we used the frame rates from the individual game scenes and those from the animated sequences to calculate a cumulative value for each specific clock rate. We ran three test runs for both single- and dual-core PCs, and got results that offered no surprises.
Without GPU-assisted PhysX, the recommended configuration delivered fluid play only for minimal or mid-range settings.
Results: Average Frame Rates vs. CPU Clock Rates
Although the maximum frame rates measured didn’t change much, and were more or less limited by the platform configurations we were testing, we did see minimal values increase measurably with processor clock rates. At the top-end of the range, in fact, the game became playable. Subjective differences between individual clock rates are more noticeable than the reported differences in frame rates may suggest. As the clock rate increased, we saw far fewer hiccups or stutters in game play and faster level load times. At higher AA and AF levels, the CPU performed up to expectations.
Results: Single- vs. Dual-Core
Not surprisingly, the dual-core processor delivered better results across the board at the same clock rates, even though the game itself in these configurations hardly benefited from the second core. This difference is perhaps best understood in terms of minimum frame rates observed. On a single-core system, the operating system and the game have to compete for the same core, whereas on a dual-core system, this bottleneck is alleviated somewhat. Despite all of this, we recommend using an AMD dual-core CPU running 2.6 GHz or faster. Anything less will result in a less satisfactory gaming experience.
Game play was tolerable to acceptable. The best approach is to use the lowest graphics settings and live with the resulting modest but playable frame rates.
Visuals
Things don’t change much here from the previous test results for the rock-bottom configuration, as low graphics settings produce unattractive results. It’s a good thing most of this game takes place in the dark or at low light levels since this ambience hides more than zombies, in this particular case.
