|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Best viewed @ 1024 x 768 and higher
Installation and Testing: Stock and overclocked configurations were used to test the system to see how well it performs to specs and beyond. The system was also tested for stability using Orthose, Prime95, and MemTest. It's no surprise, once again, to see a perfect 6 hour burn in come and go without errors or issues. After allowing the system to idle for a couple hours, benchmarks were taken using SANDRA PRO XI, Fear, and Battlefield 2142.
After installing the motherboard, the BIOS was cleared in order to start from a clean slate. I also happened to have a liquid cooling system already configured to use just as I had with both the previous 650-SLI systems. This should keep results and conditions comparable. After the system booted, it was straight to the BIOS to see what's new and how it's laid out. The BIOS layout is smaller than it's bigger 680i brother. There are a few less settings and is easier to understand as you navigate it. The main changes fall in line with the A.I. and ASUS OC features where you can set, create, or select overclocking profiles. You can set the BIOS to configure an overclock based on a percentage and let the system set itself. Sometimes this works and sometimes it doesn't. If the memory is left to default frequencies and lower timings, it sometimes won't accept the OC you designate. So, setting your memory timings to 5-5-5 will allow the BIOS to better achieve a stable system.
The BIOS on the
P5N-E SLI is a
mature, well built one. Sometimes, you can tell when a BIOS is still very new
based on its quirks and settings not working when you've had it working on
similar boards. This BIOS had a recent revision on 6/26/2007 which changed a bit
of the board's behavior since it came with a 1/7/2007 BIOS. The biggest update
is probably the guaranteed 1333FSB processor support with a few minor tweaks to
guarantee a stable platform. Stock performance benchmarks were extremely close to the Abit FP-IN9. As a matter of fact, they were so close that benchmarks were performed twice. This is where I discovered something about the board's default BUS settings which effect memory performance. After running SANDRA Memory Bandwidth a second time, the score actually increased. The last time this was seen it was due to the FP-IN9's factory overclocked Bus. So, a third, fourth, and fifth test were performed with each score increasing each time. The memory bandwidth finally topped out at a maximum 7776 M/B per second which isn't too shabby at all. This was enough to pass up the FP-IN9 which scored a maximum 7507 M/B. 3D Performance can improve greatly as you start raising or overclocking the memory frequencies in the Unlinked BIOS setting. Having an overclocked Bus means higher frames as with a processor FSB at 1270MHz even with 5-5-5 memory timings. Here you can see how the two boards measure up. Since the FP-IN9 SLI can squeeze out a little more in terms of memory and Front Side Bus frequencies, it edges past the P5N-E SLI, but not by a great margin. Does anyone really notice a few frames difference in games? Not really. Pushing the system further seemed pretty easy thanks to the newest BIOS update. The board was then overclocked using both Linked and Unlinked BIOS memory to CPU settings. Unfortunately, overclocking the P5N-E SLI didn't quite match the FP-IN9 which achieved a 1300MHz Front Side Bus frequency. The most I could coax out of the P5N-E SLI was 1250MHz using most of the voltage options. Increasing the chipset, processor, and memory voltages to a safe maximum allowed the system to operate at 1270MHz which only proved to be 95 percent stable as there were a few small system quirks that weren't originally visible. With some more extreme tweaking, that frequency might just be achievable on your motherboard since no two are perfectly alike. << A Closer Look | More Testing & Conclusion >>
Got News? Send 'em in!
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
| |
|