Due to the lack of heavily multithreaded consumer software, it makes little sense to buy a workstation/server-class CPU such as the Opteron 2435 and use it solely for gaming or general office tasks.
However, we recognise that some bit-tech readers will want a single PC to perform every task, which is why we’ve designed a comprehensive suite of consumer and professional application benchmarks for this review. These range from 3D animation and rendering packages to real-world HPC (high-performance computing) simulations that are usually run on mammoth clusters.
Here is a list of the consumer applications we used for our testing:
Sisoft Sandra 2009 SP1
Media Player Classic
Here's a list of the games we used for our testing:
X3: Terran Conflict
Here is a list of the professional applications we used for our testing:
Cinebench R10 x64
A 12-core system laid out on the desk like this sure ain't pretty - but it is extremely fast in multithreaded applications
To put the Opteron 2435, or more precisely a pair of them through their paces, we partnered them with a Supermicro H8DAE-2 motherboard and 16GB of PC2-6400 Crucial ECC registered DDR2. These test components were kindly provided by Boston, the UK distributor of Supermicro products. For comparative purposes, we ran the same benchmarks that were conducted on the Opteron 2382 and the Xeon W5580 a few months ago. The benchmark graphs also show the results from a system with a pair of the fastest previous-generation Xeon X5482s installed.
Graphics Card: Zotac GeForce GTX 260 AMP Hard Drive: Samsung SpinPoint P120S 250GB SATA II Power Supply: Enermax Galaxy 1kW, Supermicro 800W Drivers: Nvidia GeForce 185.85 Operating System: Windows Vista Ultimate x64 SP1
We tried to keep the test kit as similar as possible between all three systems (same OS, HDD and graphics card), but had to use different RAM as follows:
While there is a disparity in total memory available to the test systems, we felt it best to make this clear at the outset, and we felt that on balance, it's better to include more results from more systems in the article. This is especially the case given that most of the benchmarks we used focus on CPU performance, and so simply don't benefit from having more than 4GB of system memory.