X48 Rampage Formula Preview.

[Ghostleader]

They look promising but do they like to run on the RF.

Prize according to Fudzilla

"If all of these kits are like the one tested, these are some mean overclockers. The kits are both listed at OCZ's website and the 4GB DDR2-1066 XTC Platinum LV kit can be found here, while the 4GB Blade Series LV is here. Both kits are selling at German Alternate.de where the XTC Platinum sells for €63,90 and the fancy Blade heatsinks are a bit more expensive as it sells for €74,90."

Read the rest of the article here.

Tony over at OCZ have tested them on a AMD system though, read it here, start at post 4

[Alien Grey]

There is one thing that we know and that is that the Rampage Formula has no problems to run with 2 X 2 Giga DDR 1200MHz 5-5-5 timings so there's a good chance that they can run at that speed on the Rampage Formula.

[Ghostleader]

There is one thing that we know and that is that the Rampage Formula has no problems to run with 2 X 2 Giga DDR 1200MHz 5-5-5 timings so there's a good chance that they can run at that speed on the Rampage Formula.

I don´t doubt that the RF can handle the speed, more that can the RF run those IC at that speed (Elpida, Samsung, Micron ), probably no problem but you never know.

[Fritz_the_germ]

@greg.m,
thank for posting your oc results. As I see do you have a Q9650 and a ATI 4820X2 video card.
Many friends of mine told me that the X48 chip plus a quad needs a hell of a lot voltage to the nb being prime95 or LinX 0.5.8 stable.
This "hell of a lot voltage" should be about 1.5v and more.
Would you say that´s true?
How much voltage did you take on the NB to get your oc stable?

[Ghostleader]

@greg.m,
thank for posting your oc results. As I see do you have a Q9650 and a ATI 4820X2 video card.
Many friends of mine told me that the X48 chip plus a quad needs a hell of a lot voltage to the nb being prime95 or LinX 0.5.8 stable.
This "hell of a lot voltage" should be about 1.5v and more.
Would you say that´s true?
How much voltage did you take on the NB to get your oc stable?

Depends on how high FSB, high ram speed, low tRD (Performace Level) or aggresive AI Clock Twister you want to run.

If you don´t want to lose to much performance due to the above I would say yes you need more then 1,50v on that NB.


Here´s my Q9650@450 4,05GHz 4x2 ram@1080 555 15 tRD7 Strong with bios 0802.

It´s stable with 50 loop LinX (5+hrs, max problem size and max memory use).

It´s not an over the hill OC but it´s a decent 24/7 OC setup in my opinion but I needed 1,51v (in bios, 1,49v Everest read)

If you want to push for higher FSB, ram speed, lower tRD or more aggresive AI Clock Twister you need higher NB volt.

[Fritz_the_germ]

Thx for your reply Ghostleader,
I am planing to run 450 MHz fsb@600MHz ram speed and a PL of 6.
That means I should fire up 1.35v to the cpu and 1,51v to the nb.
Water cooling might reduce the electromigration too.
(I hope I translate this right; electromigration means that the electrons hit wholes through the tiny transistors so they won´t work anymore. Good cooling shloud prevent that {?})

[greg.m]

Thx for your reply Ghostleader,
I am planing to run 450 MHz fsb@600MHz ram speed and a PL of 6.
That means I should fire up 1.35v to the cpu and 1,51v to the nb.
Water cooling might reduce the electromigration too.
(I hope I translate this right; electromigration means that the electrons hit wholes through the tiny transistors so they won´t work anymore. Good cooling shloud prevent that {?})

Are you talking about e8500 or q9650? If i remember well i could run 450x9@1199 with very low voltage. nbv arround 1.25 and vcore 1.25 both real volts. Ican try once more and come back with the results.

Now i run 480x9@4320@1200pl7 with the volts you see on the screens. Cpu skew@200, nbskew@100. Its prime95small - blend and memtest stable but i havent tried linx and ibt.


and some benches AFTER i set everything in catalyst settings @ factory defaults settings - except clocks, 800-1000. If you go back to my previous screens youll see that i was about 2000 points worst @ vantage with the same cpu and fsb speed!!!

[greg.m]

Now i run 480x9@4320@1200pl7 with the volts you see on the screens. Cpu skew@200, nbskew@100. Its prime95small - blend and memtest stable but i havent tried linx and ibt.

I was wrong not prime95small ftp stable only blend 2 cycles error free

[Fritz_the_germ]

Yo greg.m,
I was talking about the q9650 that should run on my board in future.
My e8500 is running 450MHz@600MHz ram speed very fine right now so I am interessted if the q9650 will do the same job too

[kuebk]

Thx for your reply Ghostleader,
I am planing to run 450 MHz fsb@600MHz ram speed and a PL of 6.
That means I should fire up 1.35v to the cpu and 1,51v to the nb.
Water cooling might reduce the electromigration too.
(I hope I translate this right; electromigration means that the electrons hit wholes through the tiny transistors so they won´t work anymore. Good cooling shloud prevent that {?})

To run my ram stable @ 600mhz I need 1.45v on MCH.

[neo_rtr]

@greg.m,
thank for posting your oc results. As I see do you have a Q9650 and a ATI 4820X2 video card.
Many friends of mine told me that the X48 chip plus a quad needs a hell of a lot voltage to the nb being prime95 or LinX 0.5.8 stable.
This "hell of a lot voltage" should be about 1.5v and more.
Would you say that´s true?
How much voltage did you take on the NB to get your oc stable?

I would say thats very true as iam braking my head to figure it out.
I even push it 1.55v only to shut down the pc once LinX is running