wevsspot Posted August 9, 2010 Posted August 9, 2010 (edited) Shurman - remember that Core i7 has the memory controller on die, so you are probably starving your processor for power running at 3.8Ghz and 1.18 vcore. Here are the things I would change that should allow you to run your memory at or near it's specified rating; CPU Voltage = 1.2625 CPU PLL Voltage = 1.86 QPI DRAM Core Voltage = 1.30000 IOH Voltage = 1.30 IOH PCIE Voltage = 1.60 Also, need to know the part number of the G.Skill RAM kit you are running to make sure I recommend the correct DRAM voltage Edited August 9, 2010 by wevsspot Share this post Link to post Share on other sites More sharing options...
Shurman292 Posted August 9, 2010 Posted August 9, 2010 (edited) Shurman - remember that Core i7 has the memory controller on die, so you are probably starving your processor for power running at 3.8Ghz and 1.18 vcore. Here are the things I would change that should allow you to run your memory at or near it's specified rating; CPU Voltage = 1.2625 CPU PLL Voltage = 1.86 QPI DRAM Core Voltage = 1.30000 IOH Voltage = 1.30 IOH PCIE Voltage = 1.60 Also, need to know the part number of the G.Skill RAM kit you are running to make sure I recommend the correct DRAM voltage The part number for the ram is F3-12800CL9T-6GBNQ. I knew 1.18 seemed pretty low but it passed all the stress tests I could throw at it with very good temperature readings. I will give this a shot asap! --Edit-- I just tried running those settings and it immediately booted up with only 4gb back to 1066 for now I suppose Edited August 9, 2010 by Shurman292 Share this post Link to post Share on other sites More sharing options...
PingoPongo Posted August 9, 2010 Posted August 9, 2010 (edited) Not sure but maybe this helps. Edited August 9, 2010 by PingoPongo Share this post Link to post Share on other sites More sharing options...
wevsspot Posted August 9, 2010 Posted August 9, 2010 Shurman, please use the template below and fill in all the variables for your system. **************************** AI Tweaker **************************** Ai Overclock Tuner........... = [Manual] CPU Ratio Setting............ = [20.0] Xtreme Phase Full Power Mode. = [Enabled] BCLK Frequency............... = [200] PCIE Frequency............... = [100] DRAM Frequency............... = [1603 Mhz] UCLK Frequency............... = [3209 Mhz] QPI Link Data Rate........... = [Auto] CPU Voltage Control.......... = [Manual] CPU Voltage.................. = [1.35000] CPU PLL Voltage.............. = [1.80] QPI/DRAM Core Voltage........ = [1.30000] IOH Voltage.................. = [1.10] IOH PCIE Voltage............. = [1.50] ICH Voltage.................. = [1.10] ICH PCIE Voltage............. = [1.50] DRAM Bus Voltage............. = [1.66] DRAM DATA REF Voltage on CHA. = [Auto] DRAM CTRL REF Voltage on CHA. = [Auto] DRAM DATA REF Voltage on CHB. = [Auto] DRAM CTRL REF Voltage on CHB. = [Auto] DRAM DATA REF Voltage on CHC. = [Auto] DRAM CTRL REF Voltage on CHC. = [Auto] Load-Line Calibration........ = [Disabled] CPU Differential Amplitude... = [Auto try up to 1000mv for stability but may vary between individual cpus] CPU Clock Skew............... = [Normal] CPU Spread Spectrum.......... = [Disabled] IOH Clock Skew............... = [Normal] PCIE Spread Spectrum......... = [Disabled] C1E Support.................. = [Enabled or Disabled] Hardware Prefetcher.......... = [Enabled] Adjacent Cache Line Prefetch. = [Enabled] Intel® Virtualization Tech. = [Disabled] CPU TM Function.............. = [Disabled] Execute Disable Bit.......... = [Enabled] Intel® HT Technology....... = [Enabled] Active Processor Cores....... = [All] A20M......................... = [Disabled] Intel® SpeedStep Tech.. = [Enabled or Disabled] Intel® C-STATE Tech........ = [Disabled] ******************* DRAM Timing Control ******************* 1st Information: DRAM CAS# Latency............ = [9] DRAM RAS# to CAS# Delay...... = [9] DRAM RAS# PRE Time........... = [9] DRAM RAS# ACT Time........... = [24] DRAM RAS# to RAS# Delay...... = [AUTO] DRAM REF Cycle Time.......... = [110] DRAM WRITE Recovery Time..... = [AUTO] DRAM READ to PRE Time........ = [AUTO] DRAM FOUR ACT WIN Time....... = [AUTO or 31 and higher] DRAM Back-To-Back CAS# Delay. = [Auto] 2nd Information: DRAM Timing Mode............. = [2N] DRAM Round Trip Latency on CHA = [AUTO] DRAM Round Trip Latency on CHB = [AUTO] DRAM Round Trip Latency on CHC = [AUTO] 3rd Information: DRAM WRITE To READ Delay (DD) = [AUTO] DRAM WRITE To READ Delay (DR) = [AUTO] DRAM WRITE To READ Delay (SR) = [AUTO] DRAM READ To WRITE Delay (DD) = [AUTO] DRAM READ To WRITE Delay (DR) = [AUTO] DRAM READ To WRITE Delay (SR) = [AUTO] DRAM READ To READ Delay (DD) = [AUTO] DRAM READ To READ Delay (DR) = [AUTO] DRAM READ To READ Delay (SR) = [AUTO] DRAM WRITE to WRITE Delay (DD)= [AUTO] DRAM WRITE to WRITE Delay (DR)= [AUTO] DRAM WRITE to WRITE Delay (SR)= [AUTO] Express Gate: Disabled USB Legacy: Disabled for memtest86+ v4.10 tests Share this post Link to post Share on other sites More sharing options...
ebarone Posted August 9, 2010 Posted August 9, 2010 Is there a known way to reproduce your error? As in, do you know that setting ___ in the BIOS will always replicate the error right away? I think what you should do is try to determine the exact piece of hardware that is causing the problem, be it the motherboard or a particular stick of RAM. First, use a single stick in the first slot, and see if it boots under an overclock setting that you know is bad. Then put the same stick into the middle slot, again see if it boots. Then into the third slot, repeat. If you boot all three times you can probably say its a memory stick problem, then go about finding the bad stick (its one of the other two). If at any point you fail to boot, put a different stick into the same slot, and try again. If it fails, then you have a bad memory slot. If it succeeds, you found your bad memory stick. If all that doesnt yeild anything, then its probably a voltage/current issue somewhere. Share this post Link to post Share on other sites More sharing options...
Shurman292 Posted August 9, 2010 Posted August 9, 2010 (edited) Razorback, thanks for the reply! I will try these values out when I get home after work tonight. Just wondering: is the even CPU multiplier of 20 as stable as the odd ones like 19 or 21? I have not tried any of these other than 21 and I read somewhere that the odd ones were most stable. --Edit-- I just implemented all of the settings you provided. For now, it is working like a charm! Long-term stability with 6gb remains to be seen but it is all running as it should be. Now, however,I am slightly concerned about my temps because of the voltage increase. Running linx for a short period of time boosts my core temp up to around 83-84C with 1.35v on the cpu. I turned down my voltage to 1.325 and ran 3 linx passes (using all memory) without errors, however my core temp was up to 87C at one point! Yikes... Also, 1.30v gave me a BSOD after just one pass of linx. Edited August 10, 2010 by Shurman292 Share this post Link to post Share on other sites More sharing options...
Shurman292 Posted August 10, 2010 Posted August 10, 2010 Well, booted up this morning with 4gb so I turned cpu voltage back up to specified 1.35v. Although the linx tests proved hot still, I loaded the most cpu intensive application I could think of (Flight simulator x) and the temps only got up around 70-74C under full load. Idle temps seem to be around 43C. Are there any suggestions about how to tame these temps? I will have to wait a few days again to call it stable for 6gb at these settings. Share this post Link to post Share on other sites More sharing options...
wevsspot Posted August 10, 2010 Posted August 10, 2010 Shurman - try the settings I point out below; CPU Voltage = 1.32500 Load-Line Calibration........ = ENABLED CPU Differential Amplitude... = [Auto try up to 1000mv for stability but may vary between individual cpus] CPU Clock Skew............... = [Normal] CPU Spread Spectrum.......... = [Disabled] IOH Clock Skew............... = [Normal] PCIE Spread Spectrum......... = [Disabled] C1E Support.................. = ENABLED Hardware Prefetcher.......... = [Enabled] Adjacent Cache Line Prefetch. = [Enabled] Intel® Virtualization Tech. = [Disabled] CPU TM Function.............. = [Disabled] Execute Disable Bit.......... = [Enabled] Intel® HT Technology....... = DISABLED Active Processor Cores....... = [All] A20M......................... = [Disabled] Intel® SpeedStep™ Tech.. = ENABLED Intel® C-STATE Tech........ = ENABLED I know you said you tried 1.32500 vcore before and got a BSOD - but that is why we are going to try it with Load Line Calibration Enabled. With LLC enabled your cpu core voltage will jump about .2v under load versus droop under load. Common for most boards, but especially Asus. If it BSODs you can always go back to 1.35000 vcore and still keep LLC enabled. I've enabled/disabled a couple other options that will bring your idle temps down and reduce power consumption/heat without affecting your overclock. Remember - the temps you are seeing while stress testing are 100% loads and most likely will never be reached in your day to day computing. Share this post Link to post Share on other sites More sharing options...
Shurman292 Posted August 10, 2010 Posted August 10, 2010 Shurman - try the settings I point out below; CPU Voltage = 1.32500 Load-Line Calibration........ = ENABLED CPU Differential Amplitude... = [Auto try up to 1000mv for stability but may vary between individual cpus] CPU Clock Skew............... = [Normal] CPU Spread Spectrum.......... = [Disabled] IOH Clock Skew............... = [Normal] PCIE Spread Spectrum......... = [Disabled] C1E Support.................. = ENABLED Hardware Prefetcher.......... = [Enabled] Adjacent Cache Line Prefetch. = [Enabled] Intel® Virtualization Tech. = [Disabled] CPU TM Function.............. = [Disabled] Execute Disable Bit.......... = [Enabled] Intel® HT Technology....... = DISABLED Active Processor Cores....... = [All] A20M......................... = [Disabled] Intel® SpeedStep™ Tech.. = ENABLED Intel® C-STATE Tech........ = ENABLED I know you said you tried 1.32500 vcore before and got a BSOD - but that is why we are going to try it with Load Line Calibration Enabled. With LLC enabled your cpu core voltage will jump about .2v under load versus droop under load. Common for most boards, but especially Asus. If it BSODs you can always go back to 1.35000 vcore and still keep LLC enabled. I've enabled/disabled a couple other options that will bring your idle temps down and reduce power consumption/heat without affecting your overclock. Remember - the temps you are seeing while stress testing are 100% loads and most likely will never be reached in your day to day computing. First I tried these settings (though keeping HT enabled) and it was quickly on its way to 100C before I shut linx down for fear of burning it up. Then I turned the CPU voltage down to 1.330 and disabled LLC. Ran linx again and the results are attached. It reached near 90C, but as you said, these temps will not be seen in day-to-day usage. I noticed that the idle temps did decrease to about 36-38C which makes me feel better. Hopefully this will be my final configuration! One last thing: would turning the multiplier down to 19 (3.8ghz) decrease temps or mess with the ram in any way? Thank you so much for all your help! Share this post Link to post Share on other sites More sharing options...
CowKing Posted August 10, 2010 Posted August 10, 2010 It seems as though that if you want to keep those settings you'll need a better heatsink i.e. NH-D14 or the Megahalems Share this post Link to post Share on other sites More sharing options...
wevsspot Posted August 10, 2010 Posted August 10, 2010 While the Zalman 9900 isn't a Noctua or Meghelam, if properly installed it should be capable of keeping your CPU down in the mid 80s under full load. Before proceeding I'd remove the Zalman heatsink and check the flatness of the heatsink base as well as the flatness of the cpu IHS. You can do that with a razor blade or metal engineer's ruler. Make sure they are perfectly flat. If you see any light between the edge of the razor blade and the surface you may have to fine lap the processor IHS or heatsink base to get a better mating surface. When you remove the heatsink you should also check for proper dispersion of the thermal paste you used. It shouldn't be globbed on and there should be a footprint around the entirre heatsink and cpu where the two surfaces were mating. A bare patch or improperly spread thermal paste could indicate bad flatness or improperly applied thermal paste. For correct thermal paste application you should clean both surfaces with 90% pure rubbing alcohol or lighter fluid. Clean well using a lint free cloth. When reapplying the thermal paste, use a small rice sized dab right in the middle of the cpu IHS. Some actually apply the thermal paste to the heatsink base, but I prefer to apply it on the cpu. Before re-tightening the heatsink rotate the heatsink back and forth to help settle the thermal paste, then tighten down the heatsink retention mechanism. Dropping your cpu multiplier to 19 and reducing your cpu voltage will help with temperatures. But remember that when you do that your cpu operating frequency will drop from 4.0Ghz to 3.80Ghz. Share this post Link to post Share on other sites More sharing options...
Shurman292 Posted August 11, 2010 Posted August 11, 2010 Alright, I spent a few hours fiddling with it this morning and came out at 3.8ghz (backed it down to x19 and 1.275v). My Tmax is around 74C when under full load. I am not comfortable running my pc at 4.0ghz until I get time to lap the heatsink and reapply thermal paste (or get a new cooler altogether) so 3.8ghz will have to do. Hopefully it has enough power running to it to keep the 6gb which was the problem in the first place! Share this post Link to post Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now