Question for you Stephen. I just noticed that the ram I ordered for my system was the Mushkin 998805 6GB DDR3 PC3-12800 6-8-6-24 Redline. I guess I had assumed it was 2x3GB as opposed to 2GBX3. Will that be a problem installing 3 DIMMS in the Asus P8P67 Deluxe MB?

Hi,The Mushkin 998805 and 996805 Redline 1600 at 6-8-6-24 sticks are exactly the same, the former being sold three at a time and the later at two at a time. You can use two sticks for 4g or order an additional kit of 996805 for 8g and have one left over. That is what happened to me as I had 6g with the 1366 socket but wanted 8g for the P67 1155. Kind regards,

Hi,The Mushkin 998805 and 996805 Redline 1600 at 6-8-6-24 sticks are exactly the same, the former being sold three at a time and the later at two at a time. You can use two sticks for 4g or order an additional kit of 996805 for 8g and have one left over. That is what happened to me as I had 6g with the 1366 socket but wanted 8g for the P67 1155. Kind regards,

Stephen, have you tried with 2 RAM sticks instead of 4? that might be limiting your overclock a bit and perhaps could help you get to 5GHz with less Vcore

Good question Dazz. Will be eager to see answer.

Hey Corey,Outstanding! I used to fly only tubeliners but I have not even loaded my Level D for probably 2 years. Your video makes it looks like so much fun maybe I need to revisit that.

Stephen, you should make your return to flying tubeliners with the introduction of this bird!

Stephen - what settings are you using in the bios for that 5GHz run?

Another question. I currently OC just when using flight sim. I do this thru Nvidia Control Panel. Everybody with the 2600k seems to be doing it on a bios level. Any way to keep the option?

Hi,The Mushkin 998805 and 996805 Redline 1600 at 6-8-6-24 sticks are exactly the same, the former being sold three at a time and the later at two at a time. You can use two sticks for 4g or order an additional kit of 996805 for 8g and have one left over. That is what happened to me as I had 6g with the 1366 socket but wanted 8g for the P67 1155. Kind regards,

Stephen are you still running the Mushkin at 6-8-6-24 and 1.5v?I'm building my Sandy-based PC today with the same RAM

Stephen are you still running the Mushkin at 6-8-6-24 and 1.5v?I'm building my Sandy-based PC today with the same RAM

Hi David,I envy you! It is so much fun building computers that I am ready to do another one again, or this one over again! Have fun today. Can't wait to see how your new ASUS ROP Maximus Extremus Collosus mother of all motherboards works! Oh yeah, your question...it runs 6.8.6.24 at 1.6 normally and 1.645 at extreme overclocks with nary a peep. You can see how well the memory works in the Passmark benchmark, very fast.Kind regards,

:--)))Yes, I'm having lot of fun building computers too.I have spent few hours cleaning the case and mounting the new rig (this Lian-Li is the worst case as to mounting I've seen in my life but it's soooo unique and beautiful...!) and now I'll finish installing all the lights (...).I'll install Seven a bit later or more probably tomorrow.If I was listening to myself I'l go and buy an SSD drive + some flowers to deal with the wifey, I don't know...Oh yes... I know, I'll listen to myself! :Silly: More news to come!

Did it past the stability test on Prime 95 ?

Hi Dez,I never tried it on Prime 95, just the Intel Burn Test, PassMark, FSX and simultaneous multi applications. It was stable. I am becoming of the opinion that certain benchmarking applications over-overstressing either the GPU or CPU can have significant negative consequences. That is why both Nvidia and AMD has taken steps to prevent full power application on the GPU's and frown upon certain severe treatments for CPU's as well. Just my opinion at the moment. Kind regards,

Stephen,Could you please post your BIOS (UEFI, sorry!) settings @ 5.0?FYI, here is a very interesting Sandy benchmark to read.Very technical and well documented.

Hello StephenWere your figures with HT on? I can only get 4.8Ghz at that voltage. 4.9Ghz just goes over 1.4V with HT on. Not that 4.8 to 5Ghz will make a much difference I suspect other than to pride.

Someone posted in a local forum that he got higher stable clocks using turbo mode instead of cpu ratio (setting the multi for each core)Just in case you guys want to give it a try

...FYI, here is a very interesting Sandy benchmark to read. Very technical and well documented.

Excellent! Many thanks for posting! :-)OC VOLTAGE WARNING: When you OC your SB "to the max" you are all assuming the Vcore (= CPU Vcc) reported by your MB is 100% accurate. Well, it is not! The measurement error varies according to the MB manufacturer/model and is reported in the review to be an overvolt (actual voltage above measured voltage) of 60 mV for ASUS and an overvolt of 56 mV for GBT, as measured under full load at the Vcore CPU socket pins in the article linked by David Roch (Search for "CPU Vcore monitoring"). So even though you think you are within safe limits for Vcore, in practice you might actually be slightly over, causing your SB to fail after some time (Worst case: Before your SB 2011 kit arrives at your doorstep!). So overclockers, beware!Note: The preceding warning may be total nonsense if the Vcore reported by the BIOS is measured inside the CPU: Then it follows that as load (current) increases, voltage drops accross the CPU socket pin connector resistances, so the CPU signals the external voltage controller to raise Vcore to account for that drop and maintain Vcore constant inside the CPU. Perhaps someone on the know can clarify?The Asus Maximus IV vs. the Gigabute P67A-UD7review cites several important points:

1. The GBT NB's 24-phase Vcore power supply is needlessly complex, offers no benefits, makes the board more expensive and actually makes controlling Vcore more difficult, as evidenced in the Vcore noise measurements with transiemts of up to + or - 115 mV under load, vs. only 75 mV for the ASUS board, also under load.
2. The GBT MB DIMM sockets are closer to the CPU so less clearance is available for heat-sink-endowed DIMMs. But note heatsinks are now longer needed for DIMMs as heat output has decreased with latest generation DIMMs.
3. GBT's Award BIOS uses standard ASCII text screens to set BIOS parameters and is superior to the ASUS UEFI needlessly-complex, laggy and uselessly "flashy" set-up screens, even lacking in some needed basic functionality.
4. ASUS use of 2 BIOS chips with an on-board LED indicating which one is active was deemed useful.

Conclusion:

ASUS ROG board and Gigabyte UD7 for Sandy Bridge processors marketed at same level as current flagship motherboard, but in real battle for benchmarks and usability ASUS solution is quite ahead. To be honest, I'm not an ASUS fan, but this time their solution works better . Both boards are working perfect for 24/7 use, and allowed me to reach exact top level from my single D1 ES 2600K sample, but Maximus made that easier.

Note I am only transcribing what I think I understood from a rather technically complex review. If you have any doubts, please refer to the original article. If you find any mistakes in my understanding/transcription, please do correct them :-)Cheers,- jahman.

Stephen,Could you please post your BIOS (UEFI, sorry!) settings @ 5.0?FYI, here is a very interesting Sandy benchmark to read.Very technical and well documented.

Hi David,Sorry I took so long to answer.When I loaded Windows 64 and the ASUS Drivers something went wrong with the Marvell driver suite that disabled the Windows from seeing the two drives plugged into them although they were seen in BIOS. The P67 chipset's 2-6g/s and 4-3g/s ports were fine and allowed 4 drives including CD/DVD to function normally. But no matter how I moved my 5 plus CD/DVD drives around the Marvell plugs would not work and one drive was always missing. I know my way around computers quite well and was stumped after trying all the usual sweeps. reloads, fixes and swap outs. Finally I just decided to reinstall windows and start over, which I just finished. Now everything is back to normal, everything works as it should, and the computer is none the worse for wear, but I stayed up half the night troubleshooting. I say all that to explain that all my overclock BIOS settings saved in the OC profiles were lost, including the one run for the tests at 5 GHz with hyperthreading. However I kept notes and can (I hope) reproduce them (fairly) accurately from memory. Please be careful if you use them because I am not entirely certain with some of them.For the PassMark, FSX, Intel Burn Tests at 5.0g stable found here:PASSMARK TESTSLoad Line Calibration = HighVRM Frequency = AutoPhase Control = ExtremeDuty Control = T.ProbeCPU Current Capability = 130%CPU Voltage = Manual ModeCPU Manual Voltage = 1.40Dram Voltage = 1.65 (may have been 1.6 volts)VCCSA Voltage = 1.125VCCIO Voltage = 1.25CPU PLL Voltage = 1.9PCH Voltage = AutoDram Data Ref Voltage on CHA = Auto?CPU Ratio = 33BCLK = 100Turbo Multiplier = 50Enhanced Intel Speedstep Technology = EnabledIntel Hyperthreading Technology = EnabledTurbo Mode = EnabledTurbo Mode Parameters = All AutoRAM Settings = 6-8-6-24Intel adaptive Thermal monitor = disabledActive Processor Cores - AllLimit CPUID Maximum = DisabledIntel Virtualization Technology = DisabledCPU CIE = DisabledCPU C3 Report DisabledCPU C6 Report = DisabledKind regards,