Conclusion
Nothing can be said in this conclusion that is more valuable than the actual data itself. If, for some reason you skipped the individual charts, jumping right to the conclusion, you’ll likely not grasp what this story is all about: balance.
For each chart, we recommended a minimum level of CPU and GPU needed to play the game at that particular resolution. By tallying results of all 28 tests (seven games times four resolutions), we’ll summarize how often each solution was able to reach our targeted level of performance.
Again, a word of warning here: while valuable in many ways, this chart has the potential to be misused. Our minimum recommendations are just that--minimums--and only a guideline for obtaining maximum image quality and playable performance, as affordably as possible.
Apart from the GeForce GTX 285, which failed to reach the target framerate in Crysis when paired with anything less than Intel's Core i7-920, all of our graphics solutions met the same exact same targets today as in Part 1 of the series. Rather than summarizing graphics performance again, we will concentrate on the CPUs and drawing some conclusions from our first two “stock clocked” data sets.
One look at the totals chart for Part 2 and it’s easy to see that the graphics card will determine what resolution is playable for each game. In fact, these lines are so flat that looking at this chart alone would almost mislead us to believe the CPU has little impact on frame rates at all. This is far from the truth when it comes to outright performance or even the sweet spot of CPU/GPU balance. The fact remains that a $100 CPU does a far better job than a $100 GPU when it comes to maxing out a low resolution like 1280x1024.
While the Phenom II X2 550 Black Edition was sufficient to play all seven games, the Radeon HD 4850 was only powerful enough to push our ambitious levels of detail and AA in two of these titles. This doesn’t make the Radeon HD 4850 a bad gaming card, it just means settings will need to be adjusted down to find the desired compromise between performance and image quality.
Often, this least-expensive (acceptable) solution is still unbalanced, and would benefit from adding more CPU muscle to the mix. There is certainly good evidence to support the use of a quad-core processor for gaming. Of course, spending too much on the CPU and reducing the graphics budget can quickly result in a GPU limitation, especially as you scale resolution.
At the end of the day, you can't generalize or summarize the amount of hardware it takes to cut through your favorite game smoothly. Instead, it's necessary to look at the individual charts for each game and resolution. Also consider that these test systems were clean and only running the essentials. Additional background applications and multi-tasking would give us more reason to step up to a higher-end CPU.
Last round, we saw the stock Pentium E6300 severely hold back our graphics cards and even prevent playable performance in two of the tested games. As with the Pentium E5x00s (some of our favorite budget gaming CPUs), the real value in these chips is exposed through their massive overclocking headroom. The AMD Phenom II X2 550 BE, Phenom II X3 720 BE, and Intel Core 2 Duo E8400 all reached the same number of targets overall, only falling to the quad-core models in Crysis when a 1920x1200 resolution required a match-up with the GeForce GTX 295. While the E8400 did manage to deliver higher overall performance than the Phenom II X2 550 BE in most games, it didn't lead by much and often struggled to keep up with the more affordable Phenom II X3 720 BE. It will be interesting to see how all of these processors stack up to one another and compete with the quad-core CPUs when we overclock in Parts 3 and 4 of the series.
The Phenom II X4 955 Black Edition (the top processor represented in today’s data) performed admirably, even slightly besting the more expensive Intel Core 2 Quad Q9550. When comparing this similar level of performance, keep in mind the LGA 775 platform does utilize DDR2-1066, rather than the DDR3-1600 memory used on the AM3 platform. While both these processors manage to offer playable performance in the same number of situations as the Core i7-920, neither was powerful enough to allow the GeForce GTX 295 to take as many victories as it did dropped into an X58-based platform in Part 1.
Until it was limited by graphics performance, Intel's Core i7-920 held a significant lead between the first two parts of our series. Clearly, it's important to use a high-end CPU (and even overclock it) if you want to see the best possible performance in a graphics card review. However as we have seen thus far, that's not going to relate to the performance a gamer can expect in a less-muscular platform. Most enthusiasts are on a fixed budget, and many can’t afford the money for a tuned LGA 1366-based platform, such as those often used for our testing. Those folks can study the charts in this series to decide on the best balance for their games, resolutions, and budget. For those who want the best performance with the smallest price tag, we’ll shift focus in our next two parts towards overclocking, a procedure many enthusiasts use to maximize the performance that can be squeezed from their budgets.
Follow us on Twitter for more tech news, reviews, and exclusive updates!
