OK. I followed the 'Clunk guide" to he letter and got a blue screen.System's brand new, followed NickN's guide to th letter and ran both Prime95 and Intel Burn test, which showed it to be stable.THe recommended solution was to ncrease the Vcore.My question is: How much? The recommenedettingsare 120 and I tried 1.30. Just don't know where to go with this. Other question is that I went back, put the BIOS back to where it was and selected the automatic OC feature and am running a stable 4.327, not. How much better am I going to do manually?ThanksDan ColeCPU:Intel Core i5-2500K Sandy Bridge 3.3GHzMOBO: ASUS P8Z68-V PRO LGA 1155PSU: CORSAIR Enthusiast Series TX750M 750W ATX12V v2.31 / EPS12V v2.92 80 PLUS BRONZE CertifiedCASE: Corsair Special Edition White Graphite Series 600T Steel / Plastic ATX Mid TowerMEMORY: CORSAIR Vengeance 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1600COOLER: CORSAIR CAFA70 120mm Dual-Fan CPU CoolerGPU: EVGA SuperClocked 012-P3-1572-AR GeForce GTX 570HD: Western Digital Caviar Black WD1002FAEX 1TB 7200 RPM 64MB Cache SATA 6.0Gb/sSamsung 64 GB SATA II 2.5-Inch SSD 470 SeriesSAMSUNG CD/DVD Burner 22X DVD+R 8X DVD+RW ipb.global.registerReputation( 'rep_post_2112657', { domLikeStripId: 'like_post_2112657', app: 'forums', type: 'pid', typeid: '2112657' }, parseInt('') );

I used the same guide and I also got an blue screen. Did it with annother guide and it´s running stable at 4.6 GHZ.If you want I can give you my settings. I´m not sure whether this will work 100% cause you have a slightly other MOBO than I have.

Other than just to see what you can do, you're not going to see that much differ3e4nce in FSX between what you have at 4.3 and the few points higher that you might get. If the autoclock looks good to you - leave it alone. If you want bragging right - keep going.

Yeah I just want to get the most out FSX + Addons. Don't care about bragging rights. Dan Cole

Every processor is different so I do not know how Clunk can state 1.24-1.27v will work in every case. I read it stated that FSX provides a near linear response to processor speed. If true the difference between 4.3GHz and 4.5GHz = 4.7% so 4.7% times your current frame rate. Overclocking depends on how much risk you are willing to take. The more clock speed wanted means more voltage required means higher risk. Since you are already willing to clock to 4.3GHz and assume that risk, I would venture to suggest that 4.5GHz is not significantly more risk. It is generaly accepted that a conservative SB CPU voltage is 1.375v as long as you are maintaining temperatures on any core of 72c or less. These voltages are as measured externaly using software like CPUZ 1.58 and RealTemp 3.69.1 for temperature. I am at 1.33v and 60c to maintain 4.5GHz Your call; 4.3GHz is a very respectable clock.

Every processor is different so I do not know how Clunk can state 1.24-1.27v will work in every case. I read it stated that FSX provides a near linear response to processor speed. If true the difference between 4.3GHz and 4.5GHz = 4.7% so 4.7% times your current frame rate. Overclocking depends on how much risk you are willing to take. The more clock speed wanted means more voltage required means higher risk. Since you are already willing to clock to 4.3GHz and assume that risk, I would venture to suggest that 4.5GHz is not significantly more risk. It is generaly accepted that a conservative SB CPU voltage is 1.375v as long as you are maintaining temperatures on any core of 72c or less. These voltages are as measured externaly using software like CPUZ 1.58 and RealTemp 3.69.1 for temperature. I am at 1.33v and 60c to maintain 4.5GHz Your call; 4.3GHz is a very respectable clock.

Thanks Gary, That's just what I was looking for. So pardon the newbie question, but higher voltage means higher heat.Rght? And the risk relates to the tempurature, I'm guessing. But if It passes Prime95 and IBT tests, I should be OK. Right? And I assume the risk is mitigated, if I just lower everything a little bit. Right? Thanks. Dan

Thanks Gary, That's just what I was looking for. So pardon the newbie question, but higher voltage means higher heat.Rght? And the risk relates to the tempurature, I'm guessing. But if It passes Prime95 and IBT tests, I should be OK. Right? And I assume the risk is mitigated, if I just lower everything a little bit. Right? Thanks. Dan

Here's a very broad and maybe obvious pointer: Set a maximum Vcore and temperature for yourself, related to how much you care about your hardware's longevity and rather or not you can afford replacing it. My max Vcore is 1.35v. I'll never exceed 1.35v as that's my personal max and extreme overclocks are not my thing. Next is max temp. My personal max temp is 75C under Prime95 load (70C at current clocks). Work from the your personal max voltage (like 1.35v) at a given clock down, taking time between each voltage decrease to test stability. Just my two cents.

Thanks Gary, That's just what I was looking for. So pardon the newbie question, but higher voltage means higher heat.Rght? And the risk relates to the tempurature, I'm guessing. But if It passes Prime95 and IBT tests, I should be OK. Right? And I assume the risk is mitigated, if I just lower everything a little bit. Right? Thanks. Dan

You are correct volts makes heat, more technical and unecessary for this discussion but some tech-nut will jump on me if I do not qualify that by saying volts =watts=heat.Temperature is the main enemy however voltage and frequency in themselves in the absense of heat also are cause for concern. A CPU is a switch, a very fast switch and the faster you switch it on and off (frequency) the faster it wears out. Transisters (contained in the CPU) in addition to being the switches, are sensitive to voltage as well. Zero risk (or minimum risk) occurs at zero overclock. Zach makes good points above. Temperature at or below 72c (my limit) volts max 1.375v (my limit although 1.35v is much better) Pass 1-round of Intel burn test, pass 1-hour of OCCT, pass 1/2-hour of Prime95 max 72c on any single core and pass 2 complete rounds of mem86. That is for the initial clock. My fast checks once I pass the former, when I am playing with settings and just need a quick read are 1-round of Intel burn, 10-minutes of Prime, 10-minutes of OCCT. Most unstable clocks won't make the fast test. I would not lower anything myself, you should be able to run 4.5GHz well under the values stated. If you are really going to lose sleep leave it at 4.3 You have 3-years warranty and Intel will never know that you overclocked it unless you tell them. You are guaranteed 3-years anyway.

You are correct volts makes heat, more technical and unecessary for this discussion but some tech-nut will jump on me if I do not qualify that by saying volts =watts=heat.Temperature is the main enemy however voltage and frequency in themselves in the absense of heat also are cause for concern. A CPU is a switch, a very fast switch and the faster you switch it on and off (frequency) the faster it wears out. Transisters (contained in the CPU) in addition to being the switches, are sensitive to voltage as well. Zero risk (or minimum risk) occurs at zero overclock. Zach makes good points above. Temperature at or below 72c (my limit) volts max 1.375v (my limit although 1.35v is much better) Pass 1-round of Intel burn test, pass 1-hour of OCCT, pass 1/2-hour of Prime95 max 72c on any single core and pass 2 complete rounds of mem86. That is for the initial clock. My fast checks once I pass the former, when I am playing with settings and just need a quick read are 1-round of Intel burn, 10-minutes of Prime, 10-minutes of OCCT. Most unstable clocks won't make the fast test. I would not lower anything myself, you should be able to run 4.5GHz well under the values stated. If you are really going to lose sleep leave it at 4.3 You have 3-years warranty and Intel will never know that you overclocked it unless you tell them. You are guaranteed 3-years anyway.

So initially, try v 1.35v, running at 4.5, with temps no higher than 72c . If that works, run the stress tests, If they're good, try a lower voltage at .01 increments? If it doesn't work, reduce the speed to 4.4? And keep reducing until it does work? By the way, does the I5 require more voltage than the I7 for the same frequency? Thanks.

So initially, try v 1.35v, running at 4.5, with temps no higher than 72c . If that works, run the stress tests, If they're good, try a lower voltage at .01 increments? If it doesn't work, reduce the speed to 4.4? And keep reducing until it does work? By the way, does the I5 require more voltage than the I7 for the same frequency? Thanks.

i5 and i7 are the same voltages. I would try: Load Line Calibration = Enabled High if this wasn't already try it again at 1.3v if it was and you failed at 1.3v try 1.31, 1.32... 1.35v remember this is as read in CPUZ not the actual bios settings. If it fails at 1.35v drop down to 44 if it fails at 44 use the overclock tool and run at 4.3 you ar done.

i5 and i7 are the same voltages. I would try: Load Line Calibration = Enabled High if this wasn't already try it again at 1.3v if it was and you failed at 1.3v try 1.31, 1.32... 1.35v remember this is as read in CPUZ not the actual bios settings. If it fails at 1.35v drop down to 44 if it fails at 44 use the overclock tool and run at 4.3 you ar done.

Interesting what you say about the voltage reding in CPUZ vs. the bios. They seem alway to be lower in CPUZ.

Other question is that I went back, put the BIOS back to where it was and selected the automatic OC feature and am running a stable 4.327, not. How much better am I going to do manually?ThanksDan Cole

IMHO, your ASUS MB does a horrible job of automatically overclocking. For one thing, it just doesn't know the proper timings and settings for your memory and you'll eventually get BSOD's. Most of the time it just puts timings and voltages at Automatic and that may or may not work with your installed memory. It's a good place to start though but understand the memory voltages and timings might need to be tweaked to really get a stable overclock. I got my system up to 4.5 with an automatic overclock but had BSOD's as I previously discussed. You can look at the pinned thread in this forum which shows pictures of ASUS bios settings of several people who have successfully overclocked their system. Best regards,Jim

Interesting what you say about the voltage reding in CPUZ vs. the bios. They seem alway to be lower in CPUZ.

There are a number of reasons for that and here are a few: 1) Software is not as accurate as a digital volt/ohm meter (DVM or DVOM). 2) Vdrop can have an effect. 3) Vdroop can have an effect. 3) voltage offset can have an effect.I stay well below whatever most consider acceptable voltage limits and therefore do not require the accuracy of a DVOM. I agree with Jim above in that manual setting is prefered to automatic but prefer a more reliable source of information which predomently comes from understanding what it is you are setting and why. That Clunk guide is pretty reliable. This is good too: http://hardforum.com/showthread.php?t=1578110 See this also: http://www.anandtech.com/show/2404/5

And this freeware will provide you with all the timing/voltage settings for your memory - http://www.gtopala.com/. Best regards,Jim

Interestingly, the autooverclock seems to work ok with none of the problems mentioned. But since I want to do it manually and get a blue screen with a bios vcore of 1.35 with a multiplier of 45, with temps in the 50's and low 60's. I'm still working on it. BTW in the auto OC, its running at 4.3 and CPUZ shows the vcore at between 1.344 and 1.36, with temps also in the 50's and 60's. Would like to get to 4.5, but I guess that means increasing vcore higher than 1.35, right? I am going to try and set the VRM frequency to 350 and see if that helps. Dan