Results 1 to 7 of 7
Thread: Overclocking an i7 920
04-26-11, 10:20 AM #1
- Join Date
- Apr 2011
Overclocking an i7 920
Motherboard - GA-EX58-UD3R
RAM = http://www.compusa.com/application [...] u=C13-6152
I have 3 sticks of those running tripple channel.
CPU Clock Ratio 20x
CPU Frequency 4.00GHz (200x20)
QPI Clock Ratio Auto (I had it set to x36 then changed it to Auto and it has made a difference so far)
QPI Link Speed 7.2GHz
Uncore Clock Ratio x17
Uncore Frequency 3400MHz
BCLK Frequency 200
System Memory Multiplier 8.0
CPU VCore 1.325
QPI/Vtt Volts 1.315
IOH Core 1.30 (This setting I never changed until yesterday and I think it made a difference. This setting was mentioned in the article from Voicy but not in the guide provided by Gigabyte)
DRAM Volts 1.66v
CAS - somehow by default it was at 8/20 I changed it to 8/24 and tried 9/24
As of right now, I ran Prime95 for 2 hours and no errors, I also set Prime95 to test with 1 thread while setting the memory to 6100MBs and ran for over an hour with no errors.
I then Ran LinX with the default settings of 16331 for the problem size and 2048 for the memory but increased it to run 10 times and ran with no errors.. This was the program I was getting errors on earlier this week.
I then ran Memtest in windows. I ran 2 instances at the same time using 1500MB of RAM in each instance and it ran for over 200% coverage with no errors.
I ran Fraps, GPU-Z to record highest temps, Real Temps to record max temps and played MWII for well over an hour and no crashes at all.
Everything seem fine and dandy until yesterday, not even 4 rounds deep into MWII MP, it locks up, no BSOD, no hard crash like before, but a lock up with this loud repetitive sound. I had to do a hard boot and then I ran those tests again.
This time Prime95 had one out of 8 workers stopped. it ran 6 tests in 11 minutes with 1 error.. Ran 2 different tests in OCCT and they passed, then tried Prime again and the same worker stopped.
These are the 2 sources I used to help with the OC
http://www.overclockers.com/3-step [...] -i3-i5-i7/
http://www.tweaktown.com/news/1082 [...] index.html
The gigabyte link has a "grab a guide"
Can anyone tell me if my MoBo isn't meant to have the RAM OC'd? is it the IMC arguing with the RAM frequency? Does the CPU and the RAM not getting along to cause these lockups and BSODs before?
Please advise... I am at a loss.
04-26-11, 11:08 AM #2
- Join Date
- Apr 2004
What is your memory and timings. Your link goes nowhere. Try multiplier 21 X 180 then increase in steps to 190 to achieve same results. (turbo off)
Reduce Vcore (I run mine 21 x181 1.2 Volts 24/7), (leave qpi and ioh at auto) Have you entered the proper memory timings?
04-26-11, 11:34 AM #3
- Join Date
- Apr 2011
By default the memory timing was set to auto which showed 8-8-8-20 in blue. I changed it to 8-8-8-24 and even tried 9-9-9-24
Now will 21 x 180/181 get me to 4.0GHz?
My RAM is
Here is the link again - Corsair Dominator PC10666 RAM - 4GB, (2x2GB) DDR3, 1333MHz, but add another stick so its 3x2GB
04-26-11, 12:06 PM #4
With a current BCLK of 200 using the x8.0 DRAM multiplier (System Memory Multiplier), the effective DRAM frequency is DDR3-1600, which is well above what the modules are rated to run at. Since your RAM is DDR3-1333, rated for timings of 9-9-9-24 at that frequency, lower the DRAM multiplier from x8.0 to x6.0 for an effective DDR3-1200. Manually adjust the four primary DRAM timings, and leave all of the sub-timings on Auto. And a IOH Core voltage of 1.30V is unnecessarily high, so drop it to ~1.20V or Auto (1.10V) for a 4.0GHz clock. Also lower the Uncore Clock Ratio to x16 = 3200MHz, and QPI/VTT voltage from 1.315V to 1.30V. After making the above changes, re-run Memtest86+ to check for stability.
04-26-11, 02:01 PM #5
- Join Date
- Apr 2011
I thought my RAM was overclockable? I thought the dominator series was able to be OC'd? that is where I am confused... also the 'grab a guide' from the tweaktown link, the guy is using the same speed as I, only difference is he has the XMS series RAM and the extreme version of my MoBo.
and here is why -
"For Uncore Frequency, the BIOS allows users to change the ratio to what they want, but the i7 processor still has some limits,
because it still needs to follow some logic formula from Core i7. The formula is, Uncore speed needs to be 2x speed of memory or
ratio of 2x+1. In theory, the higher the Uncore speed, the better the performance because it gives memory and other components
inside the processor a boost and more bandwidth. I use the same way to avoid Uncore settings causing instability, so the safe way is
to force the Uncore to run at 2.4GHz (2x of DDR3 speed). But, I soon figure out that the i7 canít accept this ratio due to a logic bug,
so I raise the ratio to 2x+1. Uncore is now running at 2.6GHz. The same for memory, so I now know that Uncore will not cause the
4GHz overclock to fail. So recapping, I set Bclk to 200MHz, CPU is 4GHz, Uncore 2.6GHz and DDR3 at 1200MHz. Uncore and DDR3 is
lower than default rated speed, so I know we are giving them enough voltage to operate."
I copied and pasted that from the grab a guide from the tweaktown link.
right now Im running at 1200 because I can't seem to get 1600 to run stable with no errors. If I set the Uncore freq to 2400, it won't boot at all....
So my question now based off your response, am I not able to run at both 4.0 and 1600Mhz?
04-27-11, 12:01 PM #6
I made a mistake in my above post as you noticed, as I based my reply on the x8 DRAM multiplier (x16 Uncore Ratio) instead of x6 at DDR3-1200 (x12 Uncore = 2.4GHz, or x13 / x12+1 = 2.6GHz) at a 200 BCLK. The DDR3-1333 modules that you have are rated at 9-9-9 and 1.65V, which are already loose as is at that low a frequency, so in order to hit DDR3-1600 you may need to bump the DRAM voltage to upwards of 1.70V and / or loosen the timings even more.
04-27-11, 12:17 PM #7
- Join Date
- Apr 2011
Thanks Redduc, I might give this a try this evening, however based on this, it seems like you are saying I can run at 4.0 and 1600MHz together? btw, you are right about the timing of my RAM but it runs 1.5v at stock, not 1.65
I read this on another board yesterday and I want to get your opinion on this. I will copy and paste a link and an excerpt from the article, please let let me know what you think.
"3. What the heck is the deal with the 20x multiplier then?
This one is a real kicker (in a funny way). Yes, 20x multiplier is special, in more than one way, but mostly in the way people are trying to use it. You see, there is nothing wrong with the 20x multiplier. Indeed, it works, just like all other multipliers, perfectly fine. It is the DRAM multiplier and BCLK that people are combining with it that causes problems when combined with the QPI multiplier and the Uncore speed. At 20x200MHz, our default memory multiplier is 8 and our memory is at 1600MHz. The Uncore speed becomes 16x200MHz or 3.2GHz (or 2xDRAM at minimum). Therefore, remembering our (safe) ratios of 8:9 Uncore to QPI we see that the QPI link speed must be a minimum of 3.6GHz or higher. In case of 20x200MHz the Uncore and QPI are exactly in this ratio and with the added stress on the core, cache, IMC, DRAM, etc. this becomes a problem. At this point clock oscillations become much more prominent and if the QPI link speed falls below the 9:8 ratio to Uncore at any point in time combined with the added stress on the components the system can and likely will become unstable. Hence, at 20x200MHz with a default memory, Uncore and QPI multipliers we cannot really have a 100% stable system. So you say "just up the QPI link speed then". Not so fast. Remember our multipliers for QPI? They start at 18x, and the next is 22x, and so on. Unfortunately, 22x200MHz results in QPI link speed of 4.4GHz which results in no POST. Therefore, this cannot be done. Indeed, the highest reliable BCLK for default multipliers when using 20x core multiplier is 181MHz which with a 22x QPI multiplier would result in 3.982GHz QPI link. Even if you could set QPI multiplier to 20x, this would still not work for 99% of the boards out there as your QPI link would still be 4.0GHz. So what is the solution you ask? Memory multiplier. It needs to be lower. 1600MHz DRAM is approaching the maximum stress point for 20x200MHz core settings as it pushes on the limits of Uncore and QPI too much. By using a lower memory multiplier (say 6x if possible) the DRAM speed would become 6x200MHz or 1200MHz, Uncore could then be set at 2.4GHz with a lower multiplier which would then allow the QPI link to stay at 18x and well below the QPI link limits but still in a stable ratio to the Uncore.
So there you go. A couple of answers to some of the most painful Core i7 questions. Unfortunately, if you can read between the lines, this does not bode well for anyone owning a Core i7 940 or a 965 EE. For all intents and purposes Core i7 920 performance and Core i7 940 performance are identical. Core i7 965 EE has more flexibility but only because of its unlocked multiplier which allows it to use lower BCLK speeds and DRAM multipliers to achieve higher clocks. Otherwise, it is no different than a Core i7 920 or a Core i7 940. Like I said before, this may also be a chipset imposed limitation, and the processors themselves may be a lot better but at the moment, the state of the affairs is like this.
I have put a lot of thought and work into this so please if you want to comment do it constructively. I welcome any feedback and I hope that this will help a lot of people. It has helped me now to have a rock solid 4-core/8-thread, full 64-bit instruction set 4GHz air-only overclock on the Core i7 940 combined with a nice 1800MHz on my Corsair Dominator triple-channel sticks while keeping all of the voltages within Intel's recommended settings (i.e., DRAM is below 1.65V, QPI/DRAM is below 1.4V, and Vcore is below 1.39V under 100% load)."
Source = http://www.xtremesystems.org/forums/...&postcount=877