Test Hardware And Setup
| Test Hardware | |
|---|---|
| Processors | Intel Core i7-980X (Gulftown) 3.33 GHz at 3.73 GHz (28 * 133 MHz), LGA 1366, 6.4 GT/s QPI, 12 MB Shared L3Turbo Boost disabled, Hyper-Threading enabled, Power-savings enabled |
| Row 1 - Cell 0 | Intel Core i7-875K (Lynnfield) 2.93 GHz @ 2.93 GHz, 3.2 GHz, 3.73 GHz, LGA 1156, 8 MB Shared L3Turbo Boost disabled, Hyper-Threading enabled, Power-savings enabled |
| Row 2 - Cell 0 | Intel Core i5-655K (Clarkdale) 3.2 GHz @ 2.93 GHz, 3.2 GHz, 3.73 GHz, LGA 1156, 4 MB Shared L3Turbo Boost disabled, Hyper-Threading enabled, Power-savings enabled |
| Row 3 - Cell 0 | Intel Core i3-530 (Clarkdale) 2.93 GHz, LGA 1156, 4 MB Shared L3Power-savings enabled |
| Row 4 - Cell 0 | AMD Phenom II X6 xxxx (Thuban) 3.3 GHz @ 3.3 GHz, 3.7 GHz, Socket AM3, 6 MB Shared L3Turbo CORE disabled, Power-savings enabled |
| Row 5 - Cell 0 | AMD Athlon II X4 640 (Propus) 3 GHz, Socket AM3, No L3Power-savings enabled |
| Row 6 - Cell 0 | AMD Athlon II X2 260 (Regor) 3.2 GHz, Socket AM3, No L3Power-savings enabled |
| Motherboard | Asus Rampage III Formula (LGA 1366) Intel X58/ICH10R, BIOS 0505 |
| Row 8 - Cell 0 | Asus Maximus III Formula (LGA 1156) Intel P55, BIOS 2001 |
| Row 9 - Cell 0 | Gigabyte 890FXA-UD5 (Socket AM3), AMD 890FX/SB850, BIOS F5 |
| Memory | Kingston 6 GB (3 x 2 GB) DDR3-2000, KHX2000C8D3T1K3/6GX @ 8-8-8-24 and 1.65 V |
| Row 11 - Cell 0 | Kingston 4 GB (2 x 2 GB) DDR3-2000, KHX2000C8D3T1K3/6GX @ 8-8-8-24 and 1.65 V |
| Hard Drive | Intel SSDSA2M160G2GC 160 GB SATA 3Gb/s |
| Graphics | AMD Radeon HD 5970 2 GB GDDR5 |
| Row 14 - Cell 0 | AMD Radeon HD 5870 1 GB GDDR5 |
| Row 15 - Cell 0 | AMD Radeon HD 6870 1 GB GDDR5 |
| Row 16 - Cell 0 | AMD Radeon HD 5850 1 GB GDDR5 |
| Row 17 - Cell 0 | AMD Radeon HD 6850 1 GB GDDR5 |
| Row 18 - Cell 0 | AMD Radeon HD 5770 1 GB GDDR5 |
| Row 19 - Cell 0 | AMD Radeon HD 5750 1 GB GDDR5 |
| Row 20 - Cell 0 | AMD Radeon HD 4870 512 MB GDDR5 |
| Row 21 - Cell 0 | AMD Radeon HD 4850 512 MB GDDR3 |
| Row 22 - Cell 0 | AMD Radeon HD 4770 512 MB GDDR5 |
| Row 23 - Cell 0 | Sapphire Radeon HD 4670 512 MB GDDR4 |
| Row 24 - Cell 0 | Sapphire Radeon HD 4550 512 MB GDDR3 |
| Row 25 - Cell 0 | Gigabyte GeForce GTX 580 1.5 GB GDDR5 |
| Row 26 - Cell 0 | Nvidia GeForce GTX 480 1.5 GB GDDR5 |
| Row 27 - Cell 0 | Nvidia GeForce GTX 470 1.25 GB GDDR5 |
| Row 28 - Cell 0 | Nvidia GeForce GTX 460 1 GB GDDR5 |
| Row 29 - Cell 0 | Nvidia GeForce GTS 450 1 GB GDDR5 |
| Row 30 - Cell 0 | Nvidia GeForce GTX 295 896 MB GDDR3 * 2 |
| Row 31 - Cell 0 | BFG GeForce GTX 285 1 GB GDDR3 |
| Row 32 - Cell 0 | Zotac GeForce GTX 275 896 MB GDDR3 |
| Row 33 - Cell 0 | Zotac GeForce GTX 260 896 MB GDDR3 |
| Row 34 - Cell 0 | BFG GeForce GTS 250 1 GB GDDR3 |
| Row 35 - Cell 0 | BFG GeForce 9800 GTX 512 MB GDDR3 |
| Row 36 - Cell 0 | Asus GeForce 8500 GT 256 MB DDR3 |
| Power Supply | Cooler Master UCP-1000 W |
| System Software And Drivers | |
| Operating System | Windows 7 Ultimate 64-bit |
| DirectX | DirectX 11 |
| Graphics Driver | AMD Catalyst 10.10 |
| Row 42 - Cell 0 | Nvidia GeForce Release 260.99 (For all Nvidia cards except GTX 580) |
| Row 43 - Cell 0 | Nvidia GeForce Release 262.99 (For GTX 580) |
All of the graphics cards used in this piece are set to their reference clock rates for the sake of keeping things as even as possible. The only notable specs that vary come from Sapphire's Radeon HD 4670 with GDDR4 memory (most models actually included GDDR3) and Asus' GeForce 8500 GT, which comes overclocked from the factory. With that said, the performance numbers of that card are low enough that we left the higher clock rates in place.
Our benchmark sequence will likely become the topic for debate in the comments section, so I'll address it in as much depth as possible here. Testing in WoW is tricky business. The environments that put most stress on your system are generally going to be busy cities and hectic raids. Neither of these makes for a repeatable benchmark (you try talking nine or 24 other guys into running through and then wiping on the same fight sequence one hundred and some odd times for the sake of a performance evaluation).
So, I went for the next best thing--a point-to-point flight path in Twilight Highlands, one of the new zones opened up in Eastern Kingdoms. After manually watching the frame rate counter and flying over most of the old world content, I'm pretty confident that this is "as good as it gets" for repeatable testing. Just keep in mind that you are likely to see frame rates drop when you get into gnarly raid encounters or when you cram into a city packed with toons ready for a Cataclysm leveling race.
And don't laugh at my UI. The focus is on benchmarking here. I literally transferred my toon over to Echo Isles and started testing.
