X48 Rampage Formula Preview.

[Alien Grey]

I've already retired my Asus P5E x38@Rampage Formula for good (it is runing E5200 now).

GB-EP45UD3P is great and my best LinX accomplishment is 8 succesfull passes at 4.6GHz.

Q9650@4905MHz

Nice results but why does Cinebench show your Q9650 at 4.5GHz and CPU-Z at 4.9GHz?

[OLDcomer]

Nice results but why does Cinebench show your Q9650 at 4.5GHz and CPU-Z at 4.9GHz?

4.5 is my 24/7. I used EasyTune 6 to set my bench frequencies.

Here you can see how this board is handling high fsb overclocks.

[Alien Grey]

4.5 is my 24/7. I used EasyTune 6 to set my bench frequencies.

Here you can see how this board is handling high fsb overclocks.

Very nice result.


Gigabyte spended a lot of time to build a good BIOS that can handle these high clocks. And we're still waiting for ASUS to open up ours like they promised a year a go.

[Alien Grey]

For what it's worth. My 24/7 Overclock with MOD BIOS 1002.







BIOS Settings:

Ai Overclock Tuner [Manual]
CPU Ratio Setting [9.0]
FSB Strap to North Bridge [400MHz]
FSB Frequency [450MHz]
PCIE Frequency [100MHz]
DRAM Frequency [1199MHz]
DRAM Command Rate [2N]
DRAM CLK Skew on Channel A [Delay 50ps]
DRAM CLK Skew on Channel B [Normal]
DRAM TimingControl [Manual]
Refresh Period [16120T]
DRAM Static Read Control [Disabled]
Ai Clock Twister [Lighter]
Ai Transaction Booster [Manual]
Common Performance Level [06]
Pull-In of CH A/B all disabled
CPU Voltage [1.30000V]
CPU PLL Voltage [1.50V]
North Bridge Voltage [1.27V]
DRAM Voltage [1.80V]
FSB Termination Voltage [1.20V]
South Bridge Voltage [1.05V]
SB 1.5V Voltage [1.50V]
Loadline Calibration [Disabled]
CPU GTL Voltage Reference [0.65X]
NB GTL Voltage Reference [0.67X]
DRAM Controller Voltage REF [Auto]
DRAM Channel A/B Voltage REF [Auto]
CPU Spread Spectrum [Disabled]
PCIE Spread Spectrum [Disabled]
CPU Clock Skew [Delay 300ps]
NB Clock Skew [Delay 200ps]

The BIOS definitely likes CPU Clock Skew Delay 300ps and NB Clock Skew Delay 200ps. But as always I'm going to try for more.

[Alien Grey]

http://vip.asus.com/forum/view.aspx?...page%20Formula


@SoLoR

When you don't like me telling on the ASUS Forum that you build the MOD BIOS 1002 than I'm going to remove it from the post.

[SoLoR]

@SoLoR

When you don't like me telling on the ASUS Forum that you build the MOD BIOS 1002 than I'm going to remove it from the post.

Nah i dont mind... doubt they care much about moding bios, they would probably more care about moded drivers (since Creative is part of this as well) then moded bios.

[Alien Grey]

Nah i dont mind... doubt they care much about moding bios, they would probably more care about moded drivers (since Creative is part of this as well) then moded bios.

Someone was asking for a new SoundMAX driver because the beta 6585 driver is unstable so I replied to him to search this thread but didn't say anything about Creative or mod driver because I don't think that ASUS is going to release a new driver for the Supreme FX II.

[leo27]

Speaking of which I've recently run into some stability issues with 6620.

I've installed it via the update method and everything works fine and then after a few days the sound stopped working and when
I went into the playback devices windows detected sound enhancements causing issues and offered to disable them.
This fixed the sound issue, but now the eax effects etc don't work.

I think the 6600 driver maybe best

EDIT: The same thing just happened with the 6600 driver

[BCBC]

@leo27

Try this;

1. Uninstall Soundmax driver and Sound.Blaster X-Fi.MB
2. Delete "..\ProgramData\Creative\" folder
3. Restart
4. Reinstall respectively...

[OLDcomer]

q9650@4.6GHz 1.42V 8C ambient

[Alien Grey]

q9650@4.6GHz 1.42V 8C ambient

Congrats mate.

[leo27]

Thanks BCBC, I'll try it out.

And OLDcomer stop showing off

[Alien Grey]

I've tried to get it stable at 4.5GHz and DDR 1200MHz with the latest MOD BIOS and stability is worse than ever.

I've tried with different skew combinations and high voltages but it isn't possible to have it running stable for some time.

[greg.m]

Here is my best stable and more important relatively low-voltage settings.


Ai Overclock Tuner [Manual]
CPU Ratio Setting [9.0]
FSB Strap to North Bridge [Auto]
FSB Frequency [460MHz]
PCIE Frequency [100MHz]

DRAM Frequency [1226MHz]
DRAM Command Rate [2N]
DRAM CLK Skew on Channel A [Auto]
DRAM CLK Skew on Channel B [Auto]
DRAM TimingControl [Manual(5-5-5-15)]
Refresh Period [16120T]
DRAM Static Read Control [Disabled]
Ai Clock Twister [Moderate]
Ai Transaction Booster [Manual]
Common Performance Level [06]
Pull-In of CH A/B all disabled

CPU Voltage [1.33750V] real 1.320V
CPU PLL Voltage [1.50V]
North Bridge Voltage [1.45V] real 1.445V
DRAM Voltage [2.10V]
FSB Termination Voltage [1.36V] real 1.264V
South Bridge Voltage [Auto]
SB 1.5V Voltage [Auto]
Loadline Calibration [Enable]
CPU GTL Voltage Reference [0.62X]
NB GTL Voltage Reference [0.67X]
DRAM Controller Voltage REF [Auto]
DRAM Channel A/B Voltage REF [Auto]

CPU Spread Spectrum [Disabled]
PCIE Spread Spectrum [Disabled]
CPU Clock Skew [Delay 200ps]
NB Clock Skew [Delay 100ps]


I tried with cpu300-nb200 skews and lower voltage and it was stable BUT i got very low L2 cache results. Ithought it was a voltage problem and thats why i tried with exactly the same volt-settings you see above but nothing, same low L2 cache results. I then changed back to my old good cpu200-nb100 skews and i got the results i was used to!
I also had problems with memtest while running my mem @ 2.00V but when i tried with 2.1V it went well. I have to say that one click up from 460 to 461 fsb and i cant find mem-stability!

So from now on these will be my new 24/7 settings.

[Alien Grey]

Here is my best stable and more important relatively low-voltage settings.


Ai Overclock Tuner [Manual]
CPU Ratio Setting [9.0]
FSB Strap to North Bridge [Auto]
FSB Frequency [460MHz]
PCIE Frequency [100MHz]

DRAM Frequency [1226MHz]
DRAM Command Rate [2N]
DRAM CLK Skew on Channel A [Auto]
DRAM CLK Skew on Channel B [Auto]
DRAM TimingControl [Manual(5-5-5-15)]
Refresh Period [16120T]
DRAM Static Read Control [Disabled]
Ai Clock Twister [Moderate]
Ai Transaction Booster [Manual]
Common Performance Level [06]
Pull-In of CH A/B all disabled

CPU Voltage [1.33750V] real 1.320V
CPU PLL Voltage [1.50V]
North Bridge Voltage [1.45V] real 1.445V
DRAM Voltage [2.10V]
FSB Termination Voltage [1.36V] real 1.264V
South Bridge Voltage [Auto]
SB 1.5V Voltage [Auto]
Loadline Calibration [Enable]
CPU GTL Voltage Reference [0.62X]
NB GTL Voltage Reference [0.67X]
DRAM Controller Voltage REF [Auto]
DRAM Channel A/B Voltage REF [Auto]

CPU Spread Spectrum [Disabled]
PCIE Spread Spectrum [Disabled]
CPU Clock Skew [Delay 200ps]
NB Clock Skew [Delay 100ps]


I tried with cpu300-nb200 skews and lower voltage and it was stable BUT i got very low L2 cache results. Ithought it was a voltage problem and thats why i tried with exactly the same volt-settings you see above but nothing, same low L2 cache results. I then changed back to my old good cpu200-nb100 skews and i got the results i was used to!
I also had problems with memtest while running my mem @ 2.00V but when i tried with 2.1V it went well. I have to say that one click up from 460 to 461 fsb and i cant find mem-stability!

So from now on these will be my new 24/7 settings.

Why do you use Ai Clock Twister Moderate when you can use Lighter with less NB Voltage and no difference in memory bandwidth?


I'm still trying to find 24/7 stability at 4.22GHz and DDR 1250MHz with tRD 6 and tREF 16120T. Maybe with longer timings I can always keep it stable so now I'm trying the default timings from the ASUS Rampage Extreme.

1st Information : 5 - 5 - 5 - 15 - 5 - 52 - 12 - 6
2nd Information : 8 - 6 - 6 - 4 - 7 - 4 - 7
3rd Information : 23 - 8 - 1 - 6 - 6

There's a little memory performance loss but that's nothing compared to what I could gain if I can have it 24/7 stable.

[greg.m]

Why do you use Ai Clock Twister Moderate when you can use Lighter with less NB Voltage and no difference in memory bandwidth?

Ill give it a try

[Alien Grey]

Ill give it a try

And try the ASUS Rampage Extreme Memory timings. You'll be surprised how much easier it is to find stability.

[Alien Grey]

This could be my new 24/7 Overclock.

Ai Overclock Tuner [Manual]
CPU Ratio Setting [9.0]
FSB Strap to North Bridge [400MHz]
FSB Frequency [469MHz]
PCIE Frequency [100MHz]
DRAM Frequency [1250MHz]
DRAM Command Rate [2N]
DRAM CLK Skew on Channel A [Normal]
DRAM CLK Skew on Channel B [Advance 50ps]
DRAM Timing Control [Manual]
1st Information : 5 - 5 - 5 - 15 - 5 - 52 - 12 - 6
2nd Information : 8 - 6 - 6 - 4 - 7 - 4 - 7
3rd Information : 23 - 8 - 1 - 6 - 6
Refresh Period [16120T]
DRAM Static Read Control [Disabled]
Ai Clock Twister [Lighter]
Ai Transaction Booster [Manual]
Common Performance Level [06]
Pull-In of CH A/B all disabled
CPU Voltage [1.37500V]
CPU PLL Voltage [1.50V]
North Bridge Voltage [1.39V]
DRAM Voltage [1.92V]
FSB Termination Voltage [1.30V]
South Bridge Voltage [1.05V]
SB 1.5V Voltage [1.50V]
Loadline Calibration [Disabled]
CPU GTL Voltage Reference [0.65X]
NB GTL Voltage Reference [0.67X]
DRAM Controller Voltage REF [Auto]
DRAM Channel A/B Voltage REF [Auto]
CPU Spread Spectrum [Disabled]
PCIE Spread Spectrum [Disabled]
CPU Clock Skew [Delay 200ps]
NB Clock Skew [Delay 100ps]

The memory timings from the ASUS Rampage Extreme was exactly what I needed to keep it stable.

[leo27]

Is that with the new BIOS A-Grey because I'm surprised your using the 200ps 100ps clock skews??

[Alien Grey]

Is that with the new BIOS A-Grey because I'm surprised your using the 200ps 100ps clock skews??

Yes. It isn't possible to use 300ps and 200ps with FSB 469MHz. I'm not sure if it is going to be 24/7 stable because today I tested it again and it failed Prime95 very quick.

[spy2520]

You guys settings are slowly but surely getting me stable at fsb484. Used to not run intel burntest for more than a minute now it looks like it takes more than an hour to bluescreen. I'm using much slower ram settings though, with mixed dimms.

[greg.m]

Alien Grey i took your advise about ram settings and now here is my best 460fsb settings:

Ai Overclock Tuner [Manual]
CPU Ratio Setting [9.0]
FSB Strap to North Bridge [Auto]
FSB Frequency [460MHz]
PCIE Frequency [100MHz]

DRAM Frequency [1226MHz]
DRAM Command Rate [2N]
DRAM CLK Skew on Channel A [Auto]
DRAM CLK Skew on Channel B [Auto]
DRAM TimingControl [Manual]
1st Information : 5 - 5 - 5 - 15 - 5 - 52 - 12 - 6
2nd Information : 8 - 6 - 6 - 4 - 7 - 4 - 7
3rd Information : 23 - 8 - 1 - 6 - 6
Refresh Period [16120T]
DRAM Static Read Control [Disabled]
Ai Clock Twister [Lighter]
Ai Transaction Booster [Manual]
Common Performance Level [06]
Pull-In of CH A/B all disabled

CPU Voltage [1.31875V] real 1.304Vload, 1.29Vidle
CPU PLL Voltage [1.50V]
North Bridge Voltage [1.41V] real 1.392V
DRAM Voltage [2.00V]
FSB Termination Voltage [1.28V] real 1.184V
South Bridge Voltage [1.05V]
SB 1.5V Voltage [1.50V]
Loadline Calibration [Enable]
CPU GTL Voltage Reference [0.62X]
NB GTL Voltage Reference [0.67X]
DRAM Controller Voltage REF [Auto]
DRAM Channel A/B Voltage REF [Auto]

CPU Spread Spectrum [Disabled]
PCIE Spread Spectrum [Disabled]
CPU Clock Skew [Delay 300ps]
NB Clock Skew [Delay 200ps]

What i said about skews was bs. The problem i have with l2 cache has to do with my corrupted os. I did some benches with my backup win7-harddrive and i got the right numbers

[spy2520]

Well its good for atleast 4 hours of stress testing. Still no problems yet though, i'll run it over night before i post the settings.

[Alien Grey]

What i said about skews was bs. The problem i have with l2 cache has to do with my corrupted os. I did some benches with my backup win7-harddrive and i got the right numbers

That also counts for me. I'm trying the CPU Clock Skew 300ps and NB Clock Skew 200ps at FSB 469MHz and I've got it running but I still have to find out if it's stable enough.

Well its good for atleast 4 hours of stress testing. Still no problems yet though, i'll run it over night before i post the settings.

I think that when you can do 25 passes IBT or LinX without problems it should be enough and then you should try to run Prime95 for a few hours. Nice overclock with that Q9550.

[greg.m]

What i said about skews was bs. The problem i have with l2 cache has to do with my corrupted os. I did some benches with my backup win7-harddrive and i got the right numbers

BS again i managed to find out what is that which causes these l2 cache differences, it is my g35 logitech headset and now that's for sure Its easy to test it over and over again something i did...

...Couldn't even imagine...