EVGA P55 Classified 200 LGA1156 Motherboard Review

Test Setup & Methodologies

Test Platform:
Motherboard:	EVGA P55 Classified 200
Processor:	Intel i5 750 ES (918B255) Intel i7 860 (925B478)
Processor Cooling:	Thermalright Ultra-120 eXtreme CU 2 x Scythe Ultra Kaze 120MM 2000RPM 87.6CFM (DFS123812L-2000)
Thermal Paste:	Arctic Cooling MX-2
PCH Cooling:	Stock
NF200 Cooling:	Stock
PWM Cooling:	Stock
Memory:	OCZ Blade 3x2GB PC3-17000 8-9-8-24 (OCZ3B2133LV6GK) OCZ Blade 3x2GB PC3-16000 7-8-7-20 (OCZ3B2000LV6GK) Mushkin Redline Ascent 3x2GB PC3-12800 6-7-6-18 (998692) Corsair Dominator 4x2GB PC3-12800 8-8-8-24 (CMD8GX3M4A1600C8)
Power Supply:	Corsair HX1000W
Video Card:	4 x MSI R4890 Cyclone OC (CCC 9.10) Gigabyte GTX 260OC(NVIDIA GeForce 195.62 WHQL)
Additional Fan:	Scythe Ultra Kaze 120MM 2000RPM 87.6CFM (DFS123812L-2000)
Hard Drives:	Seagate 7200.9 80GB SATAII 8MB cache
OS:	Windows Vista x64 SP2 (with all updates)

Our setup for this review again consists of a couple different processors as well as kits of memory. In addition to this, we have used two different GPU setups, a single GTX 260 and four HD4890's. The reason for each will be known if the various sections but the main purpose was simply to broaden the range of hardware tested for compatibility. The CPU used primarily for this review was the i5 750 but near the time of publication, we acquired an i7 860 so added some screens to the Overclocking & Stability Testing section of the review as well as the Extreme Overclocking section.

Stability Overclocking Methodology

I have made a slight adjustment to my stability testing regiment but nothing drastic. By including a 20 pass run of LinX using all available memory, it became pretty redundant to run both it and OCCT so OCCT has been cut. Needless to say, its replacement does just as fine a job at testing for stability and I feel makes the testing methodology that much stronger.

Unlike many sites that simply post a CPU-Z screen shot of an overclock and call it stable, I rigorously test the stability of the overclocks I post. The methodology of how I go about that can be seen below.

Due to time constraints of a review, stability testing of a number of different overclocks requires a structured and concise testing methodology. It is for this reason that I rely on the virtual recipe below to ensure the overclocks posted are stable. I also use the same methodology for stability testing of certain components at stock as well, such is the case with memory:

• 8 threads of 32M SuperPi Mod 1.5 (ran at the same time using Hyper Pi 0.99b)
• 2 hours of 8 threads of Prime95 using Prime95 v25.7 Stress Testing Blend
• 20 passes of LinX using all available memory LinX 0.6.3
• 2 hours of HCI MemTest Pro in Windows using all available memory
• 6 loops of 3DMark Vantage (over 60 minutes of looping the full test suite)

As mentioned, this is not simply a CPU-Z screen shot stable testing methodology. This particular set of tests is all but guaranteed to provide a completely 24/7 stable computing environment no matter what you use a computer for. All overclocks are also backed up by a screen shot of the above tests completed. Keep in mind, all over clocking shown in this review is to show what our setup achieved and what is possible, but certainly not a guarantee of what a particular component is capable of. There are an endless number of variables when it comes to overclocking and guarantees can never be made.

Benchmark Methodology

My benchmark methodology hasn't changed at all from the last motherboard review but a change was made for this review regarding the 3D/Gaming benchmarks. Instead of comparing the system as has been customary with the same setup as the memory and system benchmarks, I thought it would be fun to test the scaling of CrossFireX on this motherboard. Plus, there are always some questions as to whether EVGA motherboards support CrossFireX because of their lack of advertising about it, so I thought this would lay some of those questions to rest.

Needless to say, the comparison between a single, dual, triple, and quad HD4890 setups are quite fun and exciting. For that section, we stuck with a fairly standard overclock of 4GHz as noted in the charts.

For the rest of the benchmarks, we will rely on the blue bars in the charts to outline the performance of the system at BIOS defaults with the memory timings slightly adjusted to replicate common memory kits. Due to the limited memory ratios of the i5 750 Engineering Sample processor used for the benchmarks, the memory is only running at DDR3-1333 or 667MHz. This should make for some interesting comparisons to the overclocked results with much higher memory frequencies.

For the overclocked results, we will use the red bars to plot performance in the graphs. The overclocked results originate from our Overall Stable Overclock in the Stability Overclocking section that we are about to look at right now. Before that though, have a look at the rundown of how we setup the OS for the benchmarking...

1. Windows Vista x64 w/SP2 is installed using a full format
2. Intel Chipset drivers and accessory hardware drivers (audio, network, GPU) are installed followed by a defragment and a reboot
3. At time of benchmarks the latest drivers were downloaded from their official web sites as the latest drivers, most notable, NVIDIA GeForce Release 195.62 WHQL & Catalyst Control Center 9.10
4. Programs and games are then installed followed by another defragment
5. Windows updates are then completed installing all available updates followed by a defragment
6. Benchmarks are each ran three times after a clean reboot for every iteration of the benchmark unless otherwise stated, the results are then averaged