|ASUS M4A88TD-V EVO/USB3 Motherboard|
|Reviews - Featured Reviews: Motherboards|
|Written by Bruce Normann|
|Thursday, 15 July 2010|
Page 9 of 16
EVEREST Benchmark Results
EVEREST Ultimate Edition is an industry leading system diagnostics and benchmarking solution for enthusiasts PC users, based on the award-winning EVEREST Technology by Lavalys. During system optimizations and tweaking it provides essential system and overclock information, advanced hardware monitoring and diagnostics capabilities to check the effects of the applied settings. CPU, FPU and memory benchmarks are available to measure the actual system performance and compare it to previous states or other systems. Furthermore, complete software, operating system and security information makes EVEREST Ultimate Edition a comprehensive system diagnostics tool that offers a total of 100 pages of information about your PC. All of the benchmarks used in our test bed: Queen, Photoworxx, and AES, rely on basic x86 instructions, and consume very low system memory while also being aware of Hyper Threading, multi-processors, and multi-core processors.
I recently tested the AMD AM3 platform with Everest Ultimate Edition and the triple-core 720 Black Edition Processor against the Intel I5-750 and the $200 CPU beat up on the $100 processor, with one fewer core to fight with. No surprise there, and things aren't going to get any better with the AMD 555BE dual-core driving the ASUS M4A88TD-V EVO/USB3 CPU socket.
The EVEREST Queen benchmark shows some good scaling with the AMD 555BE overclock, just like a good synthetic benchmark should. The i5 with 4 cores and a substantial overclock almost doubles the score of the triple-core 720BE, and the 555BE comes within 20% of the triple core Phenom II, with its 4.0 GHz overclock. The Queen benchmark focuses on the branch prediction capabilities and misprediction penalties of the CPU. It does this by finding possible solutions to the classic queen problem on a chessboard. At the same clock speed theoretically the processor with the shorter pipeline and smaller misprediction penalties will attain higher benchmark scores.
The Photoworxx benchmark tests for penalties against pipeline architecture. This synthetic benchmark stresses the integer arithmetic and multiplication execution units of the CPU and also the memory subsystem. Due to the fact that this test performs high memory read/write traffic, it cannot effectively scale in situations where more than two processing threads are used. The EVEREST Photoworxx benchmark performs the following tasks on a very large RGB image:
The Everest documentation hints that this benchmark is highly dependent on memory write speed, and we've see ample evidence of that in prior tests. Here we see another huge gain by the ASUS P55 motherboard, in part due to the strength of the integrated memory controller on the Lynnfield CPU. The Phenom II X3 and M4A79T Deluxe actually lag a little behind the 555BE and the M4A88TD-V EVO/USB3. The results also scaled nicely with the overclocks on the AMD CPU, eventually besting the M4A79T/720BE combo by 22%.
The AES integer benchmark measures CPU performance using AES data encryption. It utilizes Vincent Rijmen, Antoon Bosselaers and Paulo Barreto's public domain C code in ECB mode and consumes only 48 MB of memory. The AES test shows a very strong dependence on the number of cores doing the processing. The AMD 555BE CPU scaled nicely in this test; for a 25% increase in CPU clock, the score increased by 26%. Some of that gain is due to raising the main bus speed; otherwise we never would have topped the increase in clock rate, even by 1%.
The Zip Library provides another case where the number of cores exerts the most influence. Neither the dual-core or the triple-core is a match for the i5 platform, with four cores, and a brand new memory controller. The ease at which the M4A88TD-V EVO/USB3 motherboard and processor combination overclocked is evident here, as is the improved performance; a 26% improvement again when going from the stock clock of 3.2GHz to 4.0GHz.
So far, memory performance seems to have had an impact on some of the synthetic benchmarks in Everest, so it's a good idea to look at the pure memory benchmarks that are available in the test suite. EVEREST Ultimate Edition offers three simple memory bandwidth tests that focus on the basics; Read, Write, and Copy. In order to avoid concurrent threads competing over system memory bandwidth, the Memory benchmarks utilize only one processor core and one thread. This will also aid in equalizing the comparison here between Quad and Tri-Core processors.
Once again, we see that Intel has a winner with their new Core microarchitecture, and particularly the i5, with its improved memory controller. There's no contest here between the AMD and Intel platforms, even though they both now have integrated memory controllers on the CPU. We see similar results here from overclocking on the ASUS 880G board; 21%, 17%, and 25% increases in memory performance with the 25% increase in CPU clock.
Now, let's look at some additional benchmarks with PCMark Vantage, which emphasize real-world activities.