Testing Method & Test Setup

Knowing about Hynix CFR's capabilities from recent reviews, we armed ourselves with an Ivy Bridge testing platform that should allow our memory to show every last bit of its overclocking potential.
To make sure that our figures represent the sort of stability safe to use ever day, we are going to run each setting until we get a 150% pass of eight 750MB instances of HCI Memtest that is considered one of the toughest memory stress-tests around.

Motherboard	ASUS Maximus V Gene (BIOS 1204)
CPU	Intel Core i7-3770K @ 4.0 GHz
Graphic card	ASUS GTX 580
Memory	Patriot Extreme Masters Viper 3 Limited Edition PVI38G213C1K
HDD	Intel SSD 330 120 GB
PSU	Seasonic Platinum 1000 Watts
OS	Windows 7, 64 bit SP1

Results

Our processor's IMC isn't HCI Memtest stable above 1'330 MHz (DDR3-2660) on the memory. Therefore we stopped the testing at 1'330 MHz.




Firstly, we've seen almost no scaling at all when altering voltages and using “straight” timings like 11-11-11 (what the kit is specced at in fact). This could be partially explained by the fact that most ICs these days have so-called “walls” using certain tRCD and tRP values which stop memory from scaling from voltage with a fixed tCAS value.
Secondly, we notice an emerged pattern that in order to maintain linear scaling from voltage at least up to 1.65V, tRCD must be equal to tCAS+2 and tRP to tCAS+1 which we highlighted in bold on the results’ table. Similar behavior is characteristic for nearly all memory chips currently available, not only Hynix 2Gbit CFR, so it explains the presence of lots of kits with “uneven” timings on the market. We don't understand why Patriot is keeping on using straight timings.
Thirdly and most shockingly, we see that our sample has failed to achieve rated speeds topping out at 1'060MHz (DDR3-2120) using 11-11-11-27 timings at 1.5V.
We had the same issue last time with another Patriot kit, we found that the culprit is the 1T command rate we use. Setting it to 3T instead - as it is prescribed in the XMP - made things stable up to 1080MHz, which can be classified as a specification pass. Still, if memory is barely capable of doing rated speed, it won’t leave much headroom for the annual 2-5MHz overclockability loss caused by normal degradation. In this aspect, Patriot are just begging for high return rates unless they tighten up the binning procedure.

Below two screenshots showing the memory performance difference between CR 1T and 3T. Nothing to worry about.

Qui sotto un paio di screenshot delle differenti performance con il command rate impostato a 1T e 3T, niente di cui preoccuparsi!

Page 1 - Introduction	Page 4 - Results
Page 2 - Closer Look	Page 5 - Conclusion
Page 3 - Photo Gallery

Discuss this article in the forums