|CPU:||Intel Core i7 920|
2.66 GHz, 2 MB Cache
|Cooling:||Noctua U12P 1336 Edition|
Kindly supplied by Noctua
|Motherboard:||Foxcoon Bloodrage X58, BIOS P07|
Kindly supplied by Foxconn
|Video Card:||PowerColor Radeon X800GTO 16 256MB|
|Harddisk:||Samsung P80 80 GB|
|Power Supply:||Deluxe 500W|
|Software:||Windows XP SP2, Catalyst 9.3|
Performance & Overclocking
Before starting to benchmark the kit, CPU-Z was used to find out what SPD programming comes with these modules. As you can see Team Group has stored various lower speeds with relaxed timings in the kit. This should help with starting up the system after a failed OC. Nonetheless, there is also a perfectly spec'ed XMP profile present.
This time around I was eager to see if this would be the first kit on our test bench that would start at CL5-5-5-15 and 1.5V at 1333 MHz. And I am happy to report, that it did. At CL5 we were able to push the memory between 701 MHz (1402 MHz) at 1.5V and 721 MHz (1442 MHz) at 2.0V. So at this setting the memory does not scale all that well, only giving us small gains. Relaxing the timings to 6-6-6-18, the memory managed to break the 800 MHz (1600 MHz) barrier with 1.5V, but just barely as you can see in the graphs below. With JEDEC voltage settings 803 (1606 MHz) was the end of the line at 1.5V. Raising the voltage with this latency setting yielded much greater performance benefits than with tighter timings. The Xtreem kit managed to run at 888 MHz (1776 MHz) with 1.65V - the maximum voltage one should use according to Intel specifications. Pushing the memory beyond that we hit a wall at 902 MHz (1804 MHz) at 1.8V. Further raising the voltage did not allow the kit to go any higher. The next step consisted of a CL7-8-7-20 setting. This is the memory's advertised latency. At this setting it managed to show huge gains between when the voltage was raised. Right out of the gates - at 1.5V the memory ran flawlessly at 921 MHz (1842 MHz). A mere 0.1V more almost managed to break the 1000 MHz (2000 MHz) barrier, but alas the highest stable setting turned out to be 994 MHz (1988 MHz). At 1.65 V the memory did not only manage to run at the advertised speed with ease, but also allowed us to push our testing system to the limit: booting at 1033 MHz (2066 MHz) and running benchmark stable at 1025 MHz (2050 MHz). The Foxconn Bloodrage is easily capable of around 215-220 MHz Baseclk, thus the CPU is holding the system back. Still anyone wanting to push their system well beyond the usual boundaries of everday overclocking is well served with the Team Group Xtreem 2000 MHz kit.
As far as voltage scaling is concerned we managed to hit a few walls. At CL5 the memory only budged a few MHz with every bump in Voltage. At CL6, we managed very nice gains up to 920 MHz (1840 MHz) before hitting a wall at 1.8V. Additional voltage did not help at all at this point. Raising the settings to CL7-8-7 the memory managed to jump big until we hit the wall of our testing station. While we were able to bench at 1025 MHz (2050 MHz), the memory booted with 1033 MHz (2066 MHz). At this point the i7 920 is holding us back, so the memory should be able to do much more. As you can see in our benchmarks above, the kit managed 2050 MHz at CL8 with JEDEC voltage, thus there should be even more headroom here. Remember, these benchmarks only represent what our sample is capable of. Your mileage may vary!