Asus Rampage Formula (X48) Thread

[hogans]

Originally Posted by Ragsters

What is wrong with getting a vdroop. I just compensate for it by giving more voltage. Just go with whatever cpuz says during idle and load.

What is wrong with it is:

In order to get 1.368v during load to maintain stability at 4230 Mhz. I have to set the voltage in bios to 1.4632v and in turn at idle it is 1.42v.

If LLC was enabled, I wouldn't have to set it as high in bios to get 1.36v during load which in turn would give lower volts at an idle.

See were I am comming from?

[KingerXI]

Are these scores good?

I just got some G.Skill Pi 1100 Mhz memory that I am running at 1092 (410x10 with a 4:3 divider).

Thank you in advance.

[stuntz0rZ]

can any1 tell me what i can do to overclock this a little bit? ive never overclocked b4 but am sure my system can handle it. i have an intel Q9450 and 4gb corsair 1066 mhz ddr2ram. so just wondering whats the best/safest way to speed it up alittle while still keeping it stable.

[ericeod]

Originally Posted by hogans

What is wrong with it is:

In order to get 1.368v during load to maintain stability at 4230 Mhz. I have to set the voltage in bios to 1.4632v and in turn at idle it is 1.42v.

If LLC was enabled, I wouldn't have to set it as high in bios to get 1.36v during load which in turn would give lower volts at an idle.

See were I am comming from?

Loadline calibration defeats the purpose of vdroop. Besides the vdroop, there is also the voltage offsets (commonly called vdrop) which is the difference between bios set vcore and actual windows vcore. All of these safety features are designed to protect the CPU from dangerous spikes in vlotage (shown in the pics below). So enabling LLC with a 45nm chip can be considered dangerous because the 45nm cores are more sensitive to higher voltages.

Here is what is going on:

Anandtech: LLC

Here is the standard vdroop as intended:



Here is with LLC or pencil mod:

No Vdroop means the VRM circuit must work harder at maintaining a constant voltage

In this next case we eliminate Vdroop altogether and examine the chaos that ensues. As illustrated by our model, removing Vdroop does nothing to reduce the magnitude of the idle to full-load transient but does increase the settling time as the VRM must recover to a higher final regulation voltage. As in the case of no Voffset, it is possible to exceed the maximum allowable CPU voltage (VID). Clearly, removing Vdroop gains us nothing and only serves to create problems that are more serious.

So what happens when we remove both Voffset and Vdroop? The answer is simple - bad things. Although the difference between the maximum positive and negative peak overshoot are the same, severe violations to the CPU VID limit occur. If you're asking yourself what's the problem with this, consider the case of a CPU VID of 1.60000V - because the user feels this is the absolute maximum CPU voltage that they will allow. Just how high do you think CPU voltage will go after leaving a heavy load condition? We can't be sure without knowing more of the details, but we can certainly conclude that it will be well in excess of 1.6V. If you've ever run a benchmark only to have your system crash right as it finishes then you have experienced the consequences of this poor setup.


Finally, let's take one last real-world look at the consequences of removing Vdroop. ASUS' implementation of this feature, labeled as Load Line Calibration and included with their latest line of motherboards, is particularly worthy of our attention for a number of reasons. The first is that setting lower voltages with this option enabled actually results in a condition in which the CPU voltage under load is higher than the idle voltage. Imagine our confusion as we desperately struggle to understand why our system is Prime95 stable for days yet continues to crash under absolutely no load. What's more, in spite of the absence of droop and for reasons unknown, enabling this feature artificially raises our CPU's minimum stable core voltage at 4.0GHz from 1.28V to about 1.33V. As a result, our system uses more power under load than is otherwise necessary. Our efforts to reduce our processor's supply voltage backfired - instead of lowering the system's total power consumption we managed to affect a 20W increase.

[Ghostleader]

Originally Posted by KingerXI

Are these scores good?

I just got some G.Skill Pi 1100 Mhz memory that I am running at 1092 (410x10 with a 4:3 divider).

Thank you in advance.

Average, you can do better at bus 410 if you disable DRAM Static Read Control, set Ai Clock Twister to Strong, Transaction Booster to Manual and Common Performance Booster to 6, you might need NBv ~1,51 -1,57 to have it stable though.

You can also try bus 458, multi 9 (~4,12GHz) and ram at 1100 (strap 333) and DRAM Static Read Control to Disabled, set Ai Clock Twister to "Strong, Transaction Booster "Manual" and Common Performance Booster to 7.

Check out the difference between bandwidth and latency below.

Originally Posted by stuntz0rZ

can any1 tell me what i can do to overclock this a little bit? ive never overclocked b4 but am sure my system can handle it. i have an intel Q9450 and 4gb corsair 1066 mhz ddr2ram. so just wondering whats the best/safest way to speed it up alittle while still keeping it stable.

Hi and welcome to OCN

First, fill in your system spec , go to "User CP" top of the page, then "Edit System" under "System Information" to the right in "Your Control Panel".

That way it´s easier for others to help.


You can start with a moderate OC at bus 400, CPU Clock 3,2GHz and ram @ 1066 555 15.

You can try the suggestion below but since this´s a suggestion some volt values have to be alter to suite your system.

*Extreme Tweaker*
Ai Overclock Tuner               [Manual]
CPU Ratio Setting                [8]
FSB Strap to North Bridge        [400]
FSB Frequency                    [400]
PCIE Frequency                   [100]

DRAM Frequency                   [1066]

DRAM Timing Control              [Manual]
 1st Information
  CAS# Latency                   [5]
  RAS# to CAS# Delay             [5]
  RAS# PRE Time                  [5]
  RAS# ACT Time                  [15]
 DRAM Static Read Control         [Disabled]
Ai Clock Twister                 [Moderate]
Ai Transation Booster            [Manual]
  Common Performance Booster     [7 or 6] More NBv with 6 then with 7

 
CPU Voltage                      [1,3625]
CPU PLL Voltage                  [1,50]
North Bridge Voltage             [1,45]
DRAM Voltage                     [What your ram need for 1066 Cas5]
FSB Termination Voltage          [1,30]
South Bridge Voltage             [1,05]
SB 1,5V Voltage                  [1,50]
Loadline Calibration             [Disabled]
CPU GTL Voltage Reference        [0,65]
NB GTL Voltage Reference         [0,67]

CPU Spread Spectrum              [Disabled]
PCIE Spred Spectrum              [Disabled]
CPU Clock Skew                   [Delay 100ps]
NB Clock Skew                    [Normal]

*Advanced settings*
CPU Ratio Setting                [8]
C1E Support                      [Disabled]
CPU TM Function                  [Disabled]
Vanderpool Technology            [Disabled]
Execute Disable Bit              [Enabled]
Max CPUID Value Limit            [Disabled] 

Rest is on auto

[KingerXI]

Ghostleader... thanks for your post. I changed the settings to 460x9, and the memory benchmarks improved significantly. READ 9261, WRITE 9760, COPY 9665, LATENCY 55.7

I haven't tested the system for stability, but I it booted into Windows no problem. Add Rep!

[Ghostleader]

@KingerXI

Nice, happy to help

[KingerXI]

Ghostleader - I know it is blasphemy, but I just picked up a new E0 Q9550, and I have read that people are easily getting it to 4 Ghz on the Gigabyte EP45-UD3P board, so I am thinking about punting the X48 Rampage Formula. I have had limited success getting my E8600 above 4.1Ghz with this board, and I figure if I can get 4 cores running at the same speed, it will have to perform better. For those who have a MicroCenter close to them, they are selling the Q9550 for $170. What are your thoughts? Thanks in advance.

[Jeffrey]

Removed by user. Found my answers.

[Ghostleader]

Originally Posted by KingerXI

Ghostleader - I know it is blasphemy, but I just picked up a new E0 Q9550, and I have read that people are easily getting it to 4 Ghz on the Gigabyte EP45-UD3P board, so I am thinking about punting the X48 Rampage Formula. I have had limited success getting my E8600 above 4.1Ghz with this board, and I figure if I can get 4 cores running at the same speed, it will have to perform better. For those who have a MicroCenter close to them, they are selling the Q9550 for $170. What are your thoughts? Thanks in advance.

EP45-UD3P is a very good MB, specially for OC quad, it´s easier to OC quad passed bus 450 on the UD3P then the RF but it´s much more sensitive then the RF, that´s is due to the nature of the P45 chipset.

In my opinion the RF is easier to OC up to a surten busspeed, it´s more robust, rigid, not so sensitive and more forgiven.

With the UD3P you most get (for a example) the GTL ref (CPU ref and MCH ref) right, one step up or down it won´t even boot, it´s almost the same with the FSB and NB voltage (CPU Termination and MCH Core) but once you get it right, it should run a quad at bus 500 rock solid stable and compare to RF, it´s almost impossible to get a quad stable at 500.

Get 4,0GHz at bus 500 from that Q9550 shouldn´t be that difficult and if you get lucky you can get that Q9550 to run 4,25GHz or higher but you need to have a very good CPU cooler to take the heat from the quad at that speed though.


I´m a bit of puzzled why you can´t get that E8600 stable with more then 4,1GHz, shouldn´t be that difficult to reach 4,25GHz with a busspeed @ 500, maybe it´s due to your CPU cooler, Arctic Cooler 7, right, you may need to improve on that one.


Good luck if you choose to go with the UD3P but be prepare to do a lot of trail and error if you want to reach higher then 4,1 GHz with a Q9550.

If you need help, you can seek it in the Gigabyte EP45-UD3P Owners Guide or feel free to PM me and I will try to help you the best I can.