Benchmark Results: Cinebench R11
We found this particular test to be very dependent on the CPU used. For instance, running both the Radeon HD 5670 and HD 5770 with an AMD Athlon II X2 250 drops the results from around 52-57 FPS to around 43-45 FPS.
Of the two mainstream cards, the Radeon HD 5770 takes the top spot, but the HD 5670 is not that far behind. A mere five frames per second separate the two, five frames away from a base of 60 FPS. If you look at the numbers, that additional frame rate costs about 25 W. Both Radeon HD 5870 cards are limited by the Phenom II X4 955 BE. V-sync was disabled, confirmed by observing more than 60 FPS in the benchmark at certain points.
The Radeon HD 5670 adds about 40 W, while the Radeon HD 5770 adds 70 watts to the base system’s power use. Both of these cards show just how much progress AMD has made since the 2900 XT. They offer significantly higher performance for significantly less power.
Since we were able to generate performance measurements with this test, we can calculate the performance per watt ratio.
Header Cell - Column 0 | Radeon 2900 XT | Radeon HD 5670 | Radeon HD 5770 | Radeon HD 5870 1 GB | Radeon HD 5870 2 GB |
---|---|---|---|---|---|
FPS | 32.27 | 52.99 | 57.62 | 59.73 | 59.27 |
Power (minus base system power consumption) | 100 | 43 | 70 | 100 | 121 |
FPS/watt | 0.32 | 1.23 | 0.96 | 0.59 | 0.48 |
As with Crysis, the Radeon HD 5670 offers better efficiency than the HD 5770. Both of the Radeon HD 5870s have to run full-speed in Photoshop CS4 and Cinebench R11, so they are handicapped in the power department compared to the Radeon HD 5670 and HD 5770.