Winstone and Winbench Results
First I think it’s important here to look at the subsystem performance, because it should give us a good base line and make looking at the application based benchmarks that follow easier to understand.
Subsystem performance gain with iP4 upgrade
Well I’d say we have a lot to look forward to in the rest of the benchmarks if they follow this trend. A gain in graphics scores of 56 ~ 71% is quite impressive, disk scores are a bit better and both the FPU and CPU scores are significantly higher. So I’d suspect to see a significant difference using the two different systems.
Remember when looking at the CPU score that it includes testing the memory speed, and even though the actual iP4 CPU’s speed tested here is over twice the speed of the iP3 used in MHz, I’d not expect it to give a test score of 100%+ faster due to the memory not being twice as fast. But since the DDR memory, along with the larger L2 cache, should give better CPU Mark scores the result does indicate one of two things: That either the performance the CPU offers the user is not measured well by this test or the CPU may be over twice as fast in actual MHz but doesn’t really offer twice as much CPU performance due to its design. Just remember that raw CPU power is not the only thing that affects overall system performance. How well it performs certain functions can be just as important, but not show in a test such as the CPUmark 99.
Winstone Performance Gain with iP4 System
Wow! Over a 100% higher Content Creation score with the iP4 upgrades vs. the iP3 800EB and SDRAM. That is very impressive. What is interesting to look at here is that the two tests give significantly different scores, that clearly shows that not only is the iP4 system faster than the iP3 but that the actual CPU architecture is quite different. Intel has stated that the iP4 is a CPU geared for the latest multimedia applications that include not only 3D but also video and sound, and the Content Creation benchmark results would seem to confirm that.
Be the first to discuss this article!