Our stability testing methods for memory have been discussed at length in the past, so we are going to simply rely on the explanation from previous reviews. Here is a run-down of what we consider to be proper stability testing.

Memory stability, what constitutes stable? What is not considered stable? These questions get hotly debated in enthusiast forums all over the internet like little brush fires on the fringe of an inferno that play havoc with forest fire crews. Everyone has their own opinion about stability, especially when it comes to memory stability. For some, stable means they can do whatever it is on their computer without it crashing, blue-screening, or restarting; whether that means gaming or just surfing the internet. To this user, stable means simply using the computer as they normally would.

Of course, that is not what we would call stable and we do a lot more thorough testing before labeling a memory frequency/timing stable. One of the toughest types of programs on a system has been found to be distributed computing projects such as Rosetta@Home, Folding@Home, World Community Grid, and more. Running 24/7 crunching for one of these great causes is a sure way to find holes in a system if there is truly some instability, unfortunately it takes a considerable amount of time to use them for stability testing so we use the list of programs below to all but guarantee the system to be 24/7 distributed computing stable:

• Multiple loops of 3DMark 01 / 06 (30 minutes of looping the full tests each)
• Dual 32M runs of SuperPi Mod 1.5 (ran at the same time)
• 2 hours of dual Prime95 using Prime95 v25.5 Stress Testing Blend
• 2 hours of dual HCI MemTest Pro in Windows using all available memory

For those use to seeing reviews with a CPU-Z screenshot and a comment that the system was "solid as a rock", you will be quite surprised to see the above testing in the screenshots below. Every overclock of our memory sample listed in this section has gone through this testing. Naturally just because our sample has clocked this well, doesn't mean that every sample will. On the flip side, many samples of this memory will overclock much higher than ours did so it goes both ways. In the end, view the results from this sample as just that, a single sample of many varying kits.

Specification Stability Testing:

The Specification Stability Testing section is just that, to test stability at the specified frequency, timings, and voltage. With memory modules coming out specified to run at the bleeding edge, we wanted to test to make sure they were actually capable of running said specifications. We not only manually set the system up to run the specified settings, but we also measure voltage right from the motherboard to accurately know the actual voltage being supplied to the modules. The voltage that is set in the BIOS is hardly ever what is actually being supplied. Measuring vDIMM from where we do, it ensures that the modules run at their rated voltage, not at an elevated voltage because the BIOS setting was off.

The stability testing used is the exact same we use for testing the stability of the overclocks. This battery of tests has actually been referred to as "...a bit more rigorous than typical applications that most enthusiasts would run on a regular basis." by one memory manufacturer. After hearing that, we know our stability test is adequate since it is too much for some. Of course, the Dominators passed this test with flying colors as we expected despite the motherboard not really liking CL7, we'll see more about this shortly. Let's now move on to the fun stuff.

As we just discussed, all voltages listed in the chart below are marked with an (A). This (A) signifies actual voltage that is being supplied to the modules measured right at the DIMM slot. Most motherboards don't actually supply the voltage that you select in the BIOS so listing that voltage would be useless. This way, we know how the memory performed with the actual voltage being supplied.

That is right folks, there is no lying here, this kit of memory was actually stable at 1045MHz on this 790i motherboard. To say that this motherboard loves this memory at 8-7-6 timings is an understatement. To go even further, the memory was able to run dual 32M SPi instances up to 1060MHz with the same amount of voltage, but wasn't Prime/HCI Memtest stable at that frequency. In all honesty, the 8-7-6 clocks were probably being held back by the motherboard but at the same time, this NVIDIA 790i Ultra SLI chipset is a lot more impressive for memory clocks than any other has been to date. This Corsair Dominator felt like it was going to clock forever at these timings, it really was a blast to test at.

On the other hand, 7-6-5 timings were really disappointing. This again is a motherboard factor we believe. No memory we have tried in the short time we have spent with this motherboard has done well at 7-6-5. It might be CL7 or might simply be the combination of tRCD and tRP at 6-5 with CL7. Either way, we can't blame the memory for these clocks and as soon as the Asus P5E3-Premium comes in, we will be testing this kit at CL7 on that motherboard so look for an update in the discussion thread for this review at that time.

Going down to even tighter 6-6-5 was a bit of a question mark for us after seeing what 7-6-5 did but again, this kit and motherboard combo impressed us like the 8-7-6 clocking. The Corsair Dominator memory really felt at home at this timing set and clocked to a more than acceptable DDR3-1650 at 2.1v (A). These clocks with tRCD and tRP at 6-5 just further point the finger at CL7 on this motherboard as these timings with CL6 were simply incredible.