weird problem with gigabyte P67A-UD4 B3 REV. 1.1

[hclarkjr]

i recently built a sandy bridge system using these 2 sets of 8gb gskill memory - F3-12800CL9D-8GBRL and F3-12800CL9D-8GBXL. bios says they are running at the rated 1600 via XMP profile 1 but CPU-Z and AIDA64 both say one set is running at 1600 and one set at 1333. XMP shows as enabled and dual channel in CPU-Z. I am currently stress testing with prime95 and have had zero errors for over an hour so far. has anybody ever experienced this? system specs are in signature. thanx

[Chike]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

If you are looking at the SPD it doesn't show what speed they run at.
In CPU-Z it's the memory tab that does.

EDIT: please post(attach) a CPU-Z text report (CPU-Z -> About tab -> Save Report (.TXT))

[hclarkjr]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

ok here is the attachment

[Chike]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

Memory runs at DDR-1600(900MHz clock).
It may be better for stability to use each set in the same channel (e.g. one in slots 1 & 2, and the other slost 3 & 4)

[hclarkjr]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

Memory runs at DDR-1600(900MHz clock). It may be better for stability to use each set in the same channel (e.g. one in slots 1 & 2, and the other slost 3 & 4)

the manual says slots 1 and 3 and 2 and 4 for matching sets correct?

[Chike]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

2. When enabling Dual Channel mode with two or four memory modules, it is recommended that
memory of the same capacity, brand, speed, and chips be used for optimum performance.

Unfortunately you don't have match sticks.
For instanse, the BXL modules need 162ns tRFC at DDR-1600 while the BRL modules only need 110ns. Each channel has it's own controller so populating each channel with the same RAM is the better option.

[hclarkjr]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

Unfortunately you don't have match sticks. For instanse, the BXL modules need 162ns tRFC at DDR-1600 while the BRL modules only need 110ns. Each channel has it's own controller so populating each channel with the same RAM is the better option.

ok, i will try it

[Chike]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

Check channel advaced timings after you do to see if there any difference, you might need to set it manually if the BIOS is not smart enough to do it.

[hclarkjr]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

Check channel advaced timings after you do to see if there any difference, you might need to set it manually if the BIOS is not smart enough to do it.

i checked in bios, all the timings match and everything says 1600. i think the software is on crack

[Chike]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

I emant channel settings, specifically tRFC.
The software did show 1600:

Northbridge Intel Sandy Bridge rev. 09
Southbridge Intel P67 rev. B3
Graphic Interface PCI-Express
PCI-E Link Width x16
PCI-E Max Link Width x16
Memory Type DDR3
Memory Size 16384 MBytes
Channels Dual
Memory Frequency 802.4 MHz (1:6)
CAS# latency (CL) 9.0
RAS# to CAS# delay (tRCD) 9
RAS# Precharge (tRP) 9
Cycle Time (tRAS) 24
Command Rate (CR) 2T
MCHBAR I/O Base address 0x0FED10000
MCHBAR I/O Size 19456

[hclarkjr]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

i ran prime95 for almost 4 hours just now and did not get 1 error so i think everything is fine. just going to ignore the software and believe the bios

[hclarkjr]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

thanx for you help and advice too

[hclarkjr]

Re: weird problem with gigabyte P67A-UD4 B3 REV. 1.1

update, started getting blue screens yesterday and had to pull the 2 sticks from system. all is well again after pulling them out, can not figure out how it ran fine for 6 days then craps out on me. i ordered another set of the ripjaws x to match the ones in system and will sell the other set.