Just found out I bought memory with different frequency

I thought I bought the same DDR3 1333 memory, but it turns out that CPU-Z tells me differently. I've been getting some BSODs but they don't seem to point to memory problems. Should I be worried?

No. It is DDR ram (double data rate); 2 x 667 is 1333; as it should be.

Yea, but the frequencies in the timing table section & part numbers are slightly different.

The timings at the highest frequencies are the same at 9/9/10/25/34. The mhz difference between the two sticks is 667 versus 686.

I'd think it would take a heavy duty benchmark to pick up any difference and that you would have about zero chance of noticing any difference in the real world.

The way to avoid that, if you think it is important, is to buy a so-called "matching pair", which typically have consecutive serial numbers and identical part numbers.

I don't know how much faith one can put in CPU-Z either.

Thanks for the input, guys! Reps for you!

the frequencies in the timing table section & part numbers are slightly different.

The motherboard determines the actual running speeds, not the part number. Just because RAM is capable of running faster, if the system can't, the RAM won't. RAM can typically run slower than rated with no problems - if otherwise compatible.

Also, CPUz is generally considered an excellent reporting tool. The only problems I have been aware of is with brand new hardware, but they are constantly releasing updates. Make sure you have the most recent. Also note that many systems toggle down when idle or near idle, and CPUz will reflect that, in particular, with the multiplier reading. So run CPUz, then fire up a couple browser sessions or applications to get the system running full speed, then check CPUz.

@ignatzatsonic - do you have a link to reported problems?

Digerati:

No, I don't. My comment was meant literally. I don't know how much faith one can put in CPU-Z. It may be fantastic, mediocre, or in between. I don't know. I have heard bitching about several monitoring applications (particularly temperature monitors), so I know they aren't perfect, but I have no knowledge about CPU-Z. I use it myself.

Ah, okay. It sounded as if you have doubts about it, or had read something about it. That said, it does not monitor temperatures - and you are correct, there is much documentation about temperature monitors not being accurate. But they also rely on digital sensors, and don't simply count cycles, as CPUz does - so CPUz and temperature monitors are quite different beasts.

Thanks for clarifying.