X48 Rampage Formula Preview.

Printable View

Show 100 post(s) from this thread on one page

08-02-2009, 09:14 PM
traveler

Quote:

Originally Posted by Ghostleader

@traveler

That 4870X2 brings out a lot of heat and passed through the NB heatsink on the way up and also the NB craves a bit more juice to run the 4870X2 and yes both those things add so the NB runs hotter.

WC is probably the best option to cool the NB but if you plan on stay with air you can take of the heatsink/pipe arrangement that cover the NB and SB and keep the heatsink that cover the VRM, yes the heatsink/pipe is a two part arrangement, one for the NB,SB and one for the VRM.

And then try something like this with a fan for the NB

http://www.thermalright.com/new_a_pa...b/DSC_0175.jpg

and something like this for the SB

http://www.enzotechnology.com/images/cnb_s1l.jpg

So that's what's happening to it. Thanks.

Should I bump the voltage to the NB up a bit? Would that help in some way?

I just touched the heat sink fins on the 4870X2 card. Whew! It's got some heat there! Wonder if there is a way to block the hot air from going past the NB heat sink with either some kind of physical barrier or putting a Corsair Dominator memory fan on the RAM to blow the hot air away from the NB. Has anyone tried this?

Liquid cooling has been under consideration for quite a while. But I did want to get only the CPU set up first under liquid and then add NB and SB blocks later. I used to have lots of room under the 3870X2 for a SB block, but now with the 4870X2 there, there's not as much room as there used to be. Or are there fittings that can efficiently turn the liquid flow 90° entering and leaving a block on the SB?

Nice heat sinks Ghostleader. Thanks for showing them. Since I'm seriously thinking liquid, I don't want to get elaborate "temporary" air cooling solutions for the NB and SB. Yet I do need a temporary solution for now. Is there a SouthBridge heat sink that fits cleanly on the SB, that is, one that has the opposite side parallel screw mount hole tabs? Or is the Asus stock heat sink the only one?

:cool:
__________________
Asus X48 Rampage Formula / Intel Q9450 2.66 @ 3.2 / 2 x 2GB OCZ Reaper HPC PC2 8500 /
Asus Radeon HD4870X2 Tri-Fan 2GB graphics card / 24" DeLL Ultrasharp 2408WFP monitor /
Western Digital Raptor X Hard Drive / Two Western Digital Caviar RE2 WD5001ABYS 500GB 7200 RPM SATA 3.0Gb/s Hard Drives /
Creative SoundBlaster X-Fi Elite Pro 7.1 / Creative Gigaworks S750 7.1 Speaker System /
Lian Li 343B cube case / PC Power & Cooling Silencer 750 Crossfire Edition /
IOGEAR GCS1782 GCS1782 Dual-Link DVI KVMP switch with 7.1 audio
08-03-2009, 01:23 AM
Ghostleader

Quote:

Originally Posted by A-Grey

Nice Memory you've got there Ghostleader.:up:

Thank you

@traveler

Thanks, you´re welcome

If you need a temporary solution, just add a fan on top of the NB heatsink for now.

When you go for WC, don´t wait to add the NB, do it together with the CPU, it doesn´t add much temp to the loop and you will not regret it.

You don´t need WC on the SB it doesn´t get that hot, passive or actice air cooling such as Enzotech CNB-S1L or SLF-1 will do just fine.

I have the CNB-S1L on my Rampage Formula and SLF-1 on my Rampage Extreme and I´m very please with the performance.

You have to modifiy it some though, like this, the holes is to far a part.

Here´s a pic of my Rampage Formula, using a Zalman ZM-WB4 Gold :eek: for the CPU (oh yes it´s a odd choice and it´s old but it works really good, have compared it with a D-Tek Fuzion, not much of a different), for the NB I have a Koolance CHC-122, (an other odd choice but I found it to be very flexible, have used it on several boards), for the SB I use EnzoTech CNB-S1L and for the VRM I use Enzotech BMR-C1.

http://i282.photobucket.com/albums/k...Formula/RF.jpg
08-03-2009, 02:31 AM
Alien Grey

Quote:

Originally Posted by PatRaceTin

I am also found damn VID 1.3000 E8400 (manufactured 04/2009)

4.2Ghz need 1.41xx V
4.5Ghz need 1.58xx V

Send that piece of crap back from where it came.:D
08-03-2009, 08:46 AM
ElAguila

What kind of volts does the NB need to run an oc'd Q9650 and 2 4870 X2's in crossfire X. Right now I am running the NB at 1.5v. The load temps never go above 47C. My system specs are in my sig. I have the stock cooler on the NB with a fan on it and MX2 instead of the stock paste. I have a TRUE on my cpu with a delta fan. My case is very air cooled. It is a ufo cube case and the room temp stays below 25C.
08-03-2009, 10:34 AM
kuebk

NB volts have nothing to do with crossfire or gfx cards...
08-03-2009, 11:38 AM
Ghostleader

Quote:

Originally Posted by kuebk

NB volts have nothing to do with crossfire or gfx cards...

Nothing is a pretty strong word and I don´t agree with you, if I´m not misstaken both PCIe x16 go through the MCH (NB) and according to my own experience I need more juice through the MCH with a HighEnd gfx then with a LowEnd gfx, I think the same goes for the Crossfire.

I´ll soon find out, Crossfire is coming my way.

http://i282.photobucket.com/albums/k...ockDiagram.jpg
08-03-2009, 11:44 AM
kuebk

Since always PCIe lines go through MCH and I've never heard anything about raising MCH voltage to stabilize Crossfire.
08-03-2009, 12:24 PM
Fritz_the_germ

Well kuebk,
it sounds reasonable to me. The data throughput is higher using crossfire. So you should need more voltage to keep your system stable.
08-03-2009, 12:39 PM
Ghostleader

@ElAguila

Forgot about you ElAguila, well, 1,50 sounds alright but it depends on what bus speed, Ai Twister and tRD you run.

I suppose that you are stable with 1,50 so to really find out, lower it to see when you are unstable then you know.
08-03-2009, 12:43 PM
Alien Grey

Quote:

Originally Posted by Fritz_the_germ

Well kueck,
it sounds reasonable to me. The data throughput is higher using crossfire. So you should need more voltage to keep your system stable.

I'm not really sure if you need more voltage on the NB. I think that if you run Prime95 Large FFT's you'll stress your NB as much as you can.

Can two graphics cards in crossfire push more out of your NB then what you can push out of it with your CPU in Prime95 Large FFT's?:shrug:
08-03-2009, 11:48 PM
traveler

Thanks Buds for the help and suggestions.

I found the problem and it was the solution.

My Rampage Formula has a Vantec Spectrum fan in slot six which, besides looking good with the blue LED's, blows air up at the video card. When I replaced the two fan cooling solution of the Visiontek 3870X2 with the cooling solution of the Tri-Fan Asus 4870X2, I thought I would not need the extra air cooling from the Vanrec fan anymore. So I turned it down to the minimum rpm. Guess I was wrong. Apparently the Vantec fan also blew air onto the NB from under the video card.

So, I turned the Vantec Spectrum fan to its maximum rpm. Now my NB temps are coming in at 38°C idle and 48°C for load. It is a bit higher then when I had the 3870X2 card in which was 44°C under load, but I guess with the hotter 4870X2 installed, 48°C is acceptable.

Regarding watercooling, a heatsink on the SouthBridge may be a requirement because there doesn't seem to be any room for hoses and fittings on the SouthBridge waterblock if the 4870X2 (single card or CrossFire) is there and keeps its stock air cooling. Is there anyone who has the Rampage Formula SB under water while keeping the stock air cooling on the video cards?

:cool:
__________________
Asus X48 Rampage Formula / Intel Q9450 2.66 @ 3.2 / 2 x 2GB OCZ Reaper HPC PC2 8500 /
Asus Radeon HD4870X2 Tri-Fan 2GB graphics card / 24" DeLL Ultrasharp 2408WFP monitor /
Western Digital Raptor X Hard Drive / Two Western Digital Caviar RE2 WD5001ABYS 500GB 7200 RPM SATA 3.0Gb/s Hard Drives /
Creative SoundBlaster X-Fi Elite Pro 7.1 / Creative Gigaworks S750 7.1 Speaker System /
Lian Li 343B cube case / PC Power & Cooling Silencer 750 Crossfire Edition /
IOGEAR GCS1782 GCS1782 Dual-Link DVI KVMP switch with 7.1 audio
08-04-2009, 01:17 AM
Ghostleader

@kuebk and A-Grey

What I have found out is that when ever I´m stable with P95 small/large FFT and blend, LinX/IBT and HCI Design Memtset and do some for exemple Crysis, Crysis Warhead or FarCry 2 gaming with my 4870X2 I do need some extra voltage through my X48, not much ~0,02 - 0,04, this is when I compare it with my 9800GTX.

I usually do run Crysis benchmark Tool, Harbor Assualt @ very high, loop for 2hrs after the above tests and even though I´m stable with the above tests Crysis often freezes and I need to raise the NBv and then it´s fine.

Don´t know if I do some this wrong or what but that´s the way it´s for my. :shrug:

Quote:

Originally Posted by traveler

Thanks Buds for the help and suggestions.

I found the problem and it was the solution.

My Rampage Formula has a Vantec Spectrum fan in slot six which, besides looking good with the blue LED's, blows air up at the video card. When I replaced the two fan cooling solution of the Visiontek 3870X2 with the cooling solution of the Tri-Fan Asus 4870X2, I thought I would not need the extra air cooling from the Vanrec fan anymore. So I turned it down to the minimum rpm. Guess I was wrong. Apparently the Vantec fan also blew air onto the NB from under the video card.

So, I turned the Vantec Spectrum fan to its maximum rpm. Now my NB temps are coming in at 38°C idle and 48°C for load. It is a bit higher then when I had the 3870X2 card in which was 44°C under load, but I guess with the hotter 4870X2 installed, 48°C is acceptable.

Regarding watercooling, a heatsink on the SouthBridge may be a requirement because there doesn't seem to be any room for hoses and fittings on the SouthBridge waterblock if the 4870X2 (single card or CrossFire) is there and keeps its stock air cooling. Is there anyone who has the Rampage Formula SB under water while keeping the stock air cooling on the video cards?

No problem, you´re welcome, nice to here that you solved your problem :up:
08-04-2009, 02:30 AM
Alien Grey

Quote:

Originally Posted by traveler

Regarding watercooling, a heatsink on the SouthBridge may be a requirement because there doesn't seem to be any room for hoses and fittings on the SouthBridge waterblock if the 4870X2 (single card or CrossFire) is there and keeps its stock air cooling. Is there anyone who has the Rampage Formula SB under water while keeping the stock air cooling on the video cards?

There is a solution for that problem. I know that there is a watercooling heatsink that can cool the NB and SB in one loop.

Grnfinger has one of these coolers. I can't seem to find that post anymore with the picture.:mad: I don't know if I've seen it in this thread or in the Maximus II thread.

-------------------

Here's the picture with the solution to watercool the NB and SB in one loop on the ASUS Rampage Formula: http://www.xtremesystems.org/forums/...postcount=3660
08-04-2009, 02:49 AM
Armored_Kech1

Ordering some improved cooling for my NB now (thanks Ghostleader). I decided to upgrade my graphics card too. Now I can't decide between the 4890 and GTX 275 (it's driving me nuts). I game at 1680x1050. Both have aftermarket coolers that I know nothing about (noise, cooling performance).

Gainward GTX 275
http://www.overclockers.co.uk/pimg/GX-102-GW_200.jpg

ASUS 4890
http://www.overclockers.co.uk/pimg/GX-209-AS_400.jpg
08-04-2009, 03:35 AM
Ghostleader

Quote:

Originally Posted by Armored_Kech1

Ordering some improved cooling for my NB now (thanks Ghostleader). I decided to upgrade my graphics card too. Now I can't decide between the 4890 and GTX 275 (it's driving me nuts). I game at 1680x1050. Both have aftermarket coolers that I know nothing about (noise, cooling performance).

Thanks, you´re welcome, glad that I inspired you to take the step. :)

I understand, that´s a tough call, can´t really help you there, I think you gonna be pleased either way.

I choose two Sapphire Toxic HD 4890 for myself and I´m wait for them to arrive. :D

http://www.sapphiretech.com/images/p...90525_3429.jpg
08-04-2009, 06:34 AM
Armored_Kech1

Thanks again Ghostleader. I think I'll go for a Sapphire Vapor-X 4890. I figured I could always use my 8800GT for physx should the need arise. I can't find the toxic/atomic versions in stock anywhere, just the regular vapor-x which I expect will perform adequately (I know it doesn't support software voltage control but I'm not really bothered with that).
http://www.overclock3d.net/gfx/artic...112449516l.jpg
08-04-2009, 09:08 AM
Roger_D25

I think your making the right choice Armored Kech1 given the resolution you plan to use (and I assume the amount of money you want to spend), in my opinion the 4890's are amazing cards for the money at more reasonable resolutions!

A-Grey - I want to apologize for not doing a better job of pushing Asus to make those key BIOS changes you have been discussing for months now (I did receive all your PM's but only contacted Asus 3 times). Its very easy to get frustrated when trying to get Asus to make changes on an older motherboard now that they have moved on to LGA1366 (I now believe its a lost cause)! With that all said do you plan to continue working with the Rampage Formula or do you plan to make the switch to i7 in the near future? Even though I haven't been pushing Asus lately I have been continuing to push the motherboard with limited results!
08-04-2009, 10:05 AM
kuebk

Quote:

Originally Posted by Ghostleader

@kuebk and A-Grey

What I have found out is that when ever I´m stable with P95 small/large FFT and blend, LinX/IBT and HCI Design Memtset and do some for exemple Crysis, Crysis Warhead or FarCry 2 gaming with my 4870X2 I do need some extra voltage through my X48, not much ~0,02 - 0,04, this is when I compare it with my 9800GTX.

I usually do run Crysis benchmark Tool, Harbor Assualt @ very high, loop for 2hrs after the above tests and even though I´m stable with the above tests Crysis often freezes and I need to raise the NBv and then it´s fine.

Don´t know if I do some this wrong or what but that´s the way it´s for my. :shrug:

I would rather say that 4870x2 increased temps near mch so you had to increase it's voltage.
08-04-2009, 10:12 AM
Ghostleader

Quote:

Originally Posted by Armored_Kech1

Thanks again Ghostleader. I think I'll go for a Sapphire Vapor-X 4890. I figured I could always use my 8800GT for physx should the need arise. I can't find the toxic/atomic versions in stock anywhere, just the regular vapor-x which I expect will perform adequately (I know it doesn't support software voltage control but I'm not really bothered with that).

Nice choice, an other awesome piece of machinery from Sapphire.

Sorry guys for the :off:, better get back before someone :flame: us.

Armored_Kech1, what kind of cooler did you get for the NB?

Quote:

Originally Posted by kuebk

I would rather say that 4870x2 increased temps near mch so you had to increase it's voltage.

Don´t think so, I cool the MCH with water, never over 35C.
08-04-2009, 10:23 AM
kuebk

Weird Ghostleader, anyway has anyone heard anything new about bringing sli to x48?
08-04-2009, 01:52 PM
Alien Grey

Quote:

Originally Posted by Roger_D25

A-Grey - I want to apologize for not doing a better job of pushing Asus to make those key BIOS changes you have been discussing for months now (I did receive all your PM's but only contacted Asus 3 times). Its very easy to get frustrated when trying to get Asus to make changes on an older motherboard now that they have moved on to LGA1366 (I now believe its a lost cause)! With that all said do you plan to continue working with the Rampage Formula or do you plan to make the switch to i7 in the near future? Even though I haven't been pushing Asus lately I have been continuing to push the motherboard with limited results!

I'm not going to make the switch to i7. The high prices for new hardware in Belgium are making the decission for me very easy.
I don't want to spend that amount of money on hardware that has such a small performance gain. If there's going to be any at all with the luck that I've got lately with buying new hardware.

If you look a few posts back you'll see that I've bought a Q9650. The CPU is a piece of crap and I already removed it from my board after two days.

This isn't the CPU that can replace my Q9450. now I have to wait for Intel to swap the Q9650 for a good Q9650.

There's an interesting thread on the ASUS Rampage Formula forum about flashing the Rampage Formula with the Rampage Extreme BIOS.

I'm not going to try it because I think it will only destroy the board but it's interesting that more guys want these extra settings in the BIOS and that it wasn't me that was complaining again.:D
08-04-2009, 01:55 PM
Alien Grey

Quote:

Originally Posted by Ghostleader

Don´t think so, I cool the MCH with water, never over 35C.

I think that a voltage raise of 0.02V or 0.04V isn't actually a voltage raise. There's a good chance that if you stress these GFX cards that the load on the PSU makes the voltage drop a bit on the NB and needs to compensate with one or two steps higher voltage in the BIOS.:shrug:
08-04-2009, 11:29 PM
OLDcomer

Quote:

Originally Posted by A-Grey

...
There's an interesting thread on the ASUS Rampage Formula forum about flashing the Rampage Formula with the Rampage Extreme BIOS.

I'm not going to try it because I think it will only destroy the board but it's interesting that more guys want these extra settings in the BIOS and that it wasn't me that was complaining again.:D

10x for the info A-Gray. Any info for the tref bios feature or 50ps step CPU/NB skews?

btw: I recently made some benchmarks with my new q9650 and submitted them to hwbot.org.

16th @ PCMark 2005
8th @ SuperPi 32m
12th @ PiFast
12th @ wPrime 1024m

Not bad for a x38 board flashed to x48.
08-05-2009, 12:45 AM
Alien Grey

Quote:

Originally Posted by OLDcomer

10x for the info A-Gray. Any info for the tref bios feature or 50ps step CPU/NB skews?

btw: I recently made some benchmarks with my new q9650 and submitted them to hwbot.org.

16th @ PCMark 2005
8th @ SuperPi 32m
12th @ PiFast
12th @ wPrime 1024m

Not bad for a x38 board flashed to x48.

No news, it's been months since I've wrote an email to ASUS. They probably think that I don't use that board anymore.:D

This can change of course when I have some fun again with my Q9650.

Very nice clocks with your X38 flashed with the Rampage BIOS.:up:
08-05-2009, 02:35 AM
Ghostleader

Quote:

Originally Posted by A-Grey

I think that a voltage raise of 0.02V or 0.04V isn't actually a voltage raise. There's a good chance that if you stress these GFX cards that the load on the PSU makes the voltage drop a bit on the NB and needs to compensate with one or two steps higher voltage in the BIOS.:shrug:

Could be, could be but I think that my Thermaltake Toughpower 1000W can handle that stress though and I still believe that the data throughput that Crysis produce is more then Prime95 or LinX can produce and therefore I have to increase the volt.

Also when I benchmark Crysis my water temp (only CPU and MCH under water, GFX stock air) increases by 2-3C if I compare it with LinX (max problem size, max memory use) even though the CPU temp is lower with Crysis then with LinX.

Go figure :shrug:
08-05-2009, 02:50 AM
Alien Grey

Quote:

Originally Posted by Ghostleader

Could be, could be but I think that my Thermaltake Toughpower 1000W can handle that stress though and I still believe that the data throughput that Crysis produce is more then Prime95 or LinX can produce and therefore I have to increase the volt.

Also when I benchmark Crysis my water temp (only CPU and MCH under water, GFX stock air) increases by 2-3C if I compare it with LinX (max problem size, max memory use) even though the CPU temp is lower with Crysis then with LinX.

Go figure :shrug:

You are right. Crysis is probably producing more data throughput.

Are you using your E8600 CPU? That could make sense.

If you use your Q9650 the data throughput is going to be twice that high for the same CPU speed.
08-05-2009, 07:39 AM
Ghostleader

ASUS Rampage Formula BIOS 0902 / E8600 / 4x2 OCZ Blade PC-9600 / Sapphire HD 4870x2

Quote:

Originally Posted by A-Grey

You are right. Crysis is probably producing more data throughput.

Are you using your E8600 CPU? That could make sense.

If you use your Q9650 the data throughput is going to be twice that high for the same CPU speed.

It´s with my E8600.

Gonna test later with my Q9650, 4x2 Blades and those Sapphire Toxic HD 4890 in Crossfire, it gonna be a tough test for the RF, looking forward to it.

Right know I have done some test with the following gear:

ASUS Rampage Formula BIOS 0902 / E8600 / 4x2 OCZ Blade PC-9600 / Sapphire HD 4870x2 with Windows 7 RC 64bit.

I can´t make 4x2 of those Blades stable @ 1200MHz with 1,80, they need ~1,85 (1,84 set in bios), if I run 2x2 there´s no problem with 1,80 though.

Couldn´t go lower then 1,55v through the MCH to have it stable.

Here´s some screenies with my E8600@500 4,25GHz and 4x2 OCZ Blade PC-9600@1200 555 15 tRD7 Strong.

~5 hrs P95 blend (4x1600mb)
~5 hrs LinX (25 loop, max problem size and max memory use)
~5 hrs HCI Design Memtest (~500% with 8x900mb)
~3,5 hrs Crysis Benchmark Tool Assualt Harbor 90 loop

Stable enough for my.

http://i282.photobucket.com/albums/k...RD7StrongS.jpg

http://i282.photobucket.com/albums/k...tRD7Stro-2.jpg

Oh yes, she made it through all 25

http://i282.photobucket.com/albums/k...tRD7Stro-1.jpg

http://i282.photobucket.com/albums/k...tRD7Stro-3.jpg

http://i282.photobucket.com/albums/k...tRD7Stro-4.jpg
08-05-2009, 09:30 AM
kuebk

Ghostleader how much V does your e8600 require for stable 4250mhz?

Has anyone tried to cool mch/ich with hr-05? Does it cool much?
08-05-2009, 10:06 AM
Alien Grey

@Ghostleader

It's completely normal that you have to raise the memory voltage a bit when using 4 X 2 Giga.

It's amazing that you can keep it stable at FSB 500MHz and DDR 1200MHz with 8 Giga, Ai ClockTwister Strong and tRD 7.:up:

I'm curious if you can do that with your Q9650 too.
08-05-2009, 10:09 AM
Alien Grey

Quote:

Originally Posted by kuebk

Ghostleader how much V does your e8600 require for stable 4250mhz?

Has anyone tried to cool mch/ich with hr-05? Does it cool much?

My SB is cooled with a HR-05 IFX and a 80 mm fan. I've never seen the temperature over 40° Celsius with the fan running at low speed.

My room temperature is around 25° Celsius these days.
08-05-2009, 10:26 AM
Ghostleader

Quote:

Originally Posted by kuebk

Ghostleader how much V does your e8600 require for stable 4250mhz?

Has anyone tried to cool mch/ich with hr-05? Does it cool much?

~1,33 under load to be stable with 4x2Gb ram in LinX, little less with 2x2Gb (less problem size -> less stress), ~0,10 less with P95 small FFT.

Quote:

Originally Posted by A-Grey

@Ghostleader

1: It's completely normal that you have to raise the memory voltage a bit when using 4 X 2 Giga.

2: It's amazing that you can keep it stable at FSB 500MHz and DDR 1200MHz with 8 Giga, Ai ClockTwister Strong and tRD 7.:up:

3: I'm curious if you can do that with your Q9650 too.

1: Didn´t know that, do you have a short explaination why it need more jucie?

2: Thank you, I figure that the RF is pretty good with a dual, that´s all.

3: Me to but I don´t think it´s possible to do the same with my Q9650 within reasonable volt settings, I´ll be happy if I can get it stable with bus @ 450 and ram @ 1200C5 tRD6 strong within reasonable volt settings.
08-05-2009, 11:42 AM
kuebk

Quote:

Originally Posted by A-Grey

My SB is cooled with a HR-05 IFX and a 80 mm fan. I've never seen the temperature over 40° Celsius with the fan running at low speed.

My room temperature is around 25° Celsius these days.

I'm most interested with mch + hr-05 cooling results.
08-05-2009, 12:52 PM
Alien Grey

Quote:

Originally Posted by Ghostleader

~1,33 under load to be stable with 4x2Gb ram in LinX, little less with 2x2Gb (less problem size -> less stress), ~0,10 less with P95 small FFT.

1: Didn´t know that, do you have a short explaination why it need more jucie?

2: Thank you, I figure that the RF is pretty good with a dual, that´s all.

3: Me to but I don´t think it´s possible to do the same with my Q9650 within reasonable volt settings, I´ll be happy if I can get it stable with bus @ 450 and ram @ 1200C5 tRD6 strong within reasonable volt settings.

I don't know why you need more DRAM Voltage with 4 X 2 Giga. It's just something that I've found in my early days with this board.

I could lower the voltage with 0.02V if I only used one stick of memory.
With every stick I added more I needed 0.02V higher voltage.

Some guys at the ASUS Rampage Formula forum that are using 4 X 2 Giga found that too.

A Q9650@4.05GHz and 4 X 2 Giga DDR2 1200MHz with tRD 6 would still be great. It doesn't matter if it isn't possible with Ai ClockTwister Strong.

The goal should be to have at least 9600Mb/s Read Speed. That would be awesome.
08-05-2009, 01:02 PM
Alien Grey

Quote:

Originally Posted by kuebk

I'm most interested with mch + hr-05 cooling results.

Some guy on a dutch forum says that the difference is 10° Celsius lower than the stock heatsink on the rampage Formula. I don't know if he was using a fan.
08-06-2009, 03:15 AM
Armored_Kech1

Breaking news or so it seems. Somebody got SLI working on his maximus formula. Read about it here.
08-06-2009, 03:50 AM
Ghostleader

Awesome, can´t with to try it.

Good find and thanks for sharing :up:
08-06-2009, 05:54 AM
kuebk

I hope that these drivers gonna be released soon. :)
08-08-2009, 04:08 AM
Fritz_the_germ

@Alien-Grey,
what´s about your Q9650? Did get any message from Intel yet?
I wouldn´t buy a i7 too because those chips are not very much better than a good LGA 775 Quad.

@Gostleader,
your system is maximum stable. Good work:clap:
08-08-2009, 06:03 AM
Alien Grey

Quote:

Originally Posted by Fritz_the_germ

@Alien-Grey,
what´s about your Q9650? Did get any message from Intel yet?
I wouldn´t buy a i7 too because those chips are not very much better than a good LGA 775 Quad.

No, I think that everyone is on vacation.:D

About i7. If your very lucky you'll have a low vid batch that can do 4GHz+ without problems on low CPU Voltage. If your not so lucky you'll have one that needs a lot CPU Voltage to stay stable at 4GHz+ and have trouble to keep that CPU cool.

And than you have my luck and you'll end up with a CPU than can't run at 4GHz+ and have to be satisfied with 3.6GHz+ on high CPU Voltage and wasted all your cash on something that gives you only headaches and sleepless nights.:yepp:
08-08-2009, 11:47 PM
traveler

Will the EK-NB/SB ASUS 1 coolingblock fit on the Rampage Formula SouthBridge? http://ekwaterblocks.com/shop/produc...roducts_id=156

EDIT: On closer inspection, it looks like not.

:cool:
__________________
Asus X48 Rampage Formula / Intel Q9450 2.66 @ 3.2 / 2 x 2GB OCZ Reaper HPC PC2 8500 /
Asus Radeon HD4870X2 Tri-Fan 2GB graphics card / 24" DeLL Ultrasharp 2408WFP monitor /
Western Digital Raptor X Hard Drive / Two Western Digital Caviar RE2 WD5001ABYS 500GB 7200 RPM SATA 3.0Gb/s Hard Drives /
Creative SoundBlaster X-Fi Elite Pro 7.1 / Creative Gigaworks S750 7.1 Speaker System /
Lian Li 343B cube case / PC Power & Cooling Silencer 750 Crossfire Edition /
IOGEAR GCS1782 GCS1782 Dual-Link DVI KVMP switch with 7.1 audio
08-09-2009, 04:35 AM
SirDaemon

Quote:

Originally Posted by traveler

Will the EK-NB/SB ASUS 1 coolingblock fit on the Rampage Formula SouthBridge? http://ekwaterblocks.com/shop/produc...roducts_id=156

EDIT: On closer inspection, it looks like not.

:cool:

I believe only EK-NB/SB ASUS 4 is compatible with Rampage Formula SB.
08-09-2009, 07:11 PM
traveler

Quote:

Originally Posted by SirDaemon

I believe only EK-NB/SB ASUS 4 is compatible with Rampage Formula SB.

Thanks.

Too bad the ASUS 1 isn't. The coolant inflow and outflow openings on the EK-NB/SB ASUS 1 would have lined up perfectly between the fan shrouds of my two Asus 4870X2 Tri-Fan graphics cards in Crossfire. :( Water cooling the 4870X2 may be the real only solution.to getting a waterblock on the southbridge. Or leave it air cooled.

:cool:
__________________
Asus X48 Rampage Formula / Intel Q9450 2.66 @ 3.2 / 2 x 2GB OCZ Reaper HPC PC2 8500 /
Asus Radeon HD4870X2 Tri-Fan 2GB graphics card / 24" DeLL Ultrasharp 2408WFP monitor /
Western Digital Raptor X Hard Drive / Two Western Digital Caviar RE2 WD5001ABYS 500GB 7200 RPM SATA 3.0Gb/s Hard Drives /
Creative SoundBlaster X-Fi Elite Pro 7.1 / Creative Gigaworks S750 7.1 Speaker System /
Lian Li 343B cube case / PC Power & Cooling Silencer 750 Crossfire Edition /
IOGEAR GCS1782 GCS1782 Dual-Link DVI KVMP switch with 7.1 audio
08-09-2009, 11:45 PM
leo27

i've been reading this forum ever since i got my rampage formula, finally i can post after a few registration issues.

my rig is:
Antec P182
asus rampage formula x48 Bios 803
q6600 G0 @ 3.4Ghz
Gskill F2-6400CL4-2GBPI-B @850Mhz
TRUE with 2 noctua P12 Push/Pull

my settings (25 Linx 64-bit stable, Prime95 small and large fft's stable)

*Extreme Tweaker*
Ai Overclock Tuner [Manual]
OC From CPU Level Up [Auto]
CPU Ratio Setting [08.0]
FSB Strap to North Bridge [400]
FSB Frequency [425]
PCIE Frequency [100]

DRAM Frequency [DDR2-850MHz]
DRAM Command Rate [2N]
DRAM CMD Skew on Channal A [Greyed Out]
DRAM CMD Skew on Channal B [Greyed Out]
DRAM CLK Skew on Channal A [Auto]
DRAM CLK Skew on Channal B [Auto]

DRAM Timing Control [Manual]
1st Information
CAS# Latency [4]
RAS# to CAS# Delay [4]
RAS# PRE Time [3]
RAS# ACT Time [9]
RAS# to RAS# Delay [3]
REF Cycle Time [40]
WRITE Recovery Time [6]
READ to PRE Time [3]
2nd Information
READ to WRITE Delay (S/D) [7]
WRITE to READ Delay (S) [3]
WRITE to READ Delay (D) [4]
READ to READ Delay (S) [4]
READ to READ Delay (D) [5]
WRITE to WRITE Delay (S) [4]
WRITE to WRITE Delay (D) [5]
3rd Information
WRITE to PRE Delay [12]
READ to PRE Delay [3]
PRE to PRE Delay [1]
ALL PRE to ACT Delay [4]
ALL PRE to REF Delay [4]

DRAM Static Read Control [Enabled]
Ai Clock Twister [Stronger]
Ai Transaction Booster [08]
Pull-in of CHA PH1 [Enabled]
Pull-in of CHA PH2 [Enabled]
Pull-in of CHA PH3 [Enabled]
Pull-in of CHA PH4 [Enabled]
Pull-in of CHB PH1 [Enabled]
Pull-in of CHB PH2 [Enabled]
Pull-in of CHB PH3 [Enabled]
Pull-in of CHB PH4 [Enabled]

CPU Voltage [1.41250v] (1.4V in windows)
CPU PLL Voltage [1.50v]
North Bridge Voltage [1.37v]
DRAM Voltage [1.84v]
FSB Termination Voltage [1.32v]
South Bridge Voltage [1.05v]
SB 1,5V Voltage [1.50v]
Loadline Calibration [Enabled]
CPU GTL Voltage Reference [0.65x]
NB GTL Voltage Reference [0.67x]
DRAM Controller Voltage REF [AUTO]
DRAM Channel A Voltage REF [AUTO]
DRAM Channel B Voltage REF [AUTO]

CPU LED Selection [CPU Voltage]
NB LED Selection [North Bridge Voltage]
SB LED Selection [South Bridge Voltage]
Voltiminder LED [Enabled]

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

Now i have run into a problem in that i was unable to reduce the memory timings (4-4-4-12 at 1.92V in bios) with a PL7, however if i run PL8 with all phase pull-ins as above then i can achieve the timings and voltage shown in the settings above (4-4-3-9 at 1.84V).

I take it PL8 with all Pull-ins is equivalent to PL7?????

to A-GREY
I've read that you use UT2004 to determine the FSB termination voltage. What settings do you use in UT2004????
and do you run it with Prime95 small fft's or large fft's????
an explanation of the procedure would be appreciated.

Lastly is a CPU GTL of 0.65x recommended for a 65nm processor or is 0.63x advised.

Thanks for your help and keeping up with this thread. It is a invaluable resource for overclocking this board.
08-10-2009, 05:13 AM
Alien Grey

Quote:

Originally Posted by leo27

i've been reading this forum ever since i got my rampage formula, finally i can post after a few registration issues.

my rig is:
Antec P182
asus rampage formula x48 Bios 803
q6600 G0 @ 3.4Ghz
Gskill F2-6400CL4-2GBPI-B @850Mhz
TRUE with 2 noctua P12 Push/Pull

my settings (25 Linx 64-bit stable, Prime95 small and large fft's stable)

*Extreme Tweaker*
Ai Overclock Tuner [Manual]
OC From CPU Level Up [Auto]
CPU Ratio Setting [08.0]
FSB Strap to North Bridge [400]
FSB Frequency [425]
PCIE Frequency [100]

DRAM Frequency [DDR2-850MHz]
DRAM Command Rate [2N]
DRAM CMD Skew on Channal A [Greyed Out]
DRAM CMD Skew on Channal B [Greyed Out]
DRAM CLK Skew on Channal A [Auto]
DRAM CLK Skew on Channal B [Auto]

DRAM Timing Control [Manual]
1st Information
CAS# Latency [4]
RAS# to CAS# Delay [4]
RAS# PRE Time [3]
RAS# ACT Time [9]
RAS# to RAS# Delay [3]
REF Cycle Time [40]
WRITE Recovery Time [6]
READ to PRE Time [3]
2nd Information
READ to WRITE Delay (S/D) [7]
WRITE to READ Delay (S) [3]
WRITE to READ Delay (D) [4]
READ to READ Delay (S) [4]
READ to READ Delay (D) [5]
WRITE to WRITE Delay (S) [4]
WRITE to WRITE Delay (D) [5]
3rd Information
WRITE to PRE Delay [12]
READ to PRE Delay [3]
PRE to PRE Delay [1]
ALL PRE to ACT Delay [4]
ALL PRE to REF Delay [4]

DRAM Static Read Control [Enabled]
Ai Clock Twister [Stronger]
Ai Transaction Booster [08]
Pull-in of CHA PH1 [Enabled]
Pull-in of CHA PH2 [Enabled]
Pull-in of CHA PH3 [Enabled]
Pull-in of CHA PH4 [Enabled]
Pull-in of CHB PH1 [Enabled]
Pull-in of CHB PH2 [Enabled]
Pull-in of CHB PH3 [Enabled]
Pull-in of CHB PH4 [Enabled]

CPU Voltage [1.41250v] (1.4V in windows)
CPU PLL Voltage [1.50v]
North Bridge Voltage [1.37v]
DRAM Voltage [1.84v]
FSB Termination Voltage [1.32v]
South Bridge Voltage [1.05v]
SB 1,5V Voltage [1.50v]
Loadline Calibration [Enabled]
CPU GTL Voltage Reference [0.65x]
NB GTL Voltage Reference [0.67x]
DRAM Controller Voltage REF [AUTO]
DRAM Channel A Voltage REF [AUTO]
DRAM Channel B Voltage REF [AUTO]

CPU LED Selection [CPU Voltage]
NB LED Selection [North Bridge Voltage]
SB LED Selection [South Bridge Voltage]
Voltiminder LED [Enabled]

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

Now i have run into a problem in that i was unable to reduce the memory timings (4-4-4-12 at 1.92V in bios) with a PL7, however if i run PL8 with all phase pull-ins as above then i can achieve the timings and voltage shown in the settings above (4-4-3-9 at 1.84V).

I take it PL8 with all Pull-ins is equivalent to PL7?????

to A-GREY
I've read that you use UT2004 to determine the FSB termination voltage. What settings do you use in UT2004????
and do you run it with Prime95 small fft's or large fft's????
an explanation of the procedure would be appreciated.

Lastly is a CPU GTL of 0.65x recommended for a 65nm processor or is 0.63x advised.

Thanks for your help and keeping up with this thread. It is a invaluable resource for overclocking this board.

PL 8 and all Phase Pull-ins enabled should give you the same bandwidth as if you would use PL 7.

I think that if you look at Anandtech's calculation that X is larger than what they are saying that it should be. That is probably the reason why a PL 7 isn't possible.

My settings for UT2004 (64 bit edition) are resolution 1024 x 768 pixels with everything at maximum quality.

Prime95 is running in Blend mode.

According to Anandtech for 65nm CPU's you should use CPU GTL 0.65X. I have no idea if this is better than using CPU GTL 0.63X.

You could try it with CPU GTL 0.63X and run LinX or IBT and see if it stays stable and what the GFlops result is. When it's higher than it's better to use the CPU GTL 0.63X.
08-10-2009, 11:56 PM
leo27

thanks for the reply. I'll try out the 0.63x GTL and see if it makes a difference in GFlops, I've already run it at 0.63x before with the same voltages so i know its stable.
Also are you running the demo of UT2004 or the full game?
08-11-2009, 04:56 AM
Alien Grey

Quote:

Originally Posted by leo27

thanks for the reply. I'll try out the 0.63x GTL and see if it makes a difference in GFlops, I've already run it at 0.63x before with the same voltages so i know its stable.
Also are you running the demo of UT2004 or the full game?

I'm running the full game at 1024 X 768 pixels in a window with 30 bots and unlimited time and kills. All you have to do then is hit spectate.

If you don't have the full game you can always use the demo. The only problem is that you can't change the amount of bots in the demo.
08-12-2009, 08:15 AM
BlackArchon

I'm using a Q9550 E0 with my Rampage Formula, BIOS 0902. When I set FSB strap and FSB to 400 MHz, resulting in 3400 MHz for the CPU, I need 1.41 V MCH. Except in the last hot weeks, this setting runs stably, but NB temp was in the high 70s. But when I read your MCH temps, this seems to be quite high.

So maybe I should remove the chipset cooler and check for good contact?
08-12-2009, 08:56 AM
Ghostleader

@BlackArchon

Very hot indeed for 1,41 MCHv, replace the TIM between the heatsink and IHS should be good idea to start with and also add a fan or replace the heatsink with some 3rd part cooler.
08-12-2009, 09:24 AM
mach82

Quote:

Originally Posted by BlackArchon

I'm using a Q9550 E0 with my Rampage Formula, BIOS 0902. When I set FSB strap and FSB to 400 MHz, resulting in 3400 MHz for the CPU, I need 1.41 V MCH. Except in the last hot weeks, this setting runs stably, but NB temp was in the high 70s. But when I read your MCH temps, this seems to be quite high.

So maybe I should remove the chipset cooler and check for good contact?

One more think to consider - FSB 400 MHz is a standard supported FSB frequency for your MB. This means, that for FSB 400 MHz you don't need any voltage increase for MCH! Lower it t the lowest voltage available in bios and then your temps will drop by much.
08-12-2009, 09:42 AM
Ghostleader

@mach82

He could probably do that if he for example ran a dual and some 2x1 PC-6400 @ CAS5 with a high tRD but he runs a quad and some 4x1 PC-8000 and probably as low tRD as possible and therefore he need 1,41 as he wrote in the post.
08-12-2009, 03:36 PM
leo27

If you want to reduce the vNB further you could reduce Ai clock twister to moderate, this would have less of an impact than lowering the tRD.
Also how are people finding bios 902, any reason to move from 803??
08-13-2009, 01:06 AM
BlackArchon

The board won't run stably with a VMCH of less than 1.41 V. I am using a PL of 6 and AI Twister to 'Moderate'. Some minutes ago I cleaned the heatpipe and I hope this will help. ;)

Are you able to run at 400 MHz FSB with stock MCH voltage?
08-13-2009, 09:47 AM
kuebk

Quote:

Originally Posted by BlackArchon

The board won't run stably with a VMCH of less than 1.41 V. I am using a PL of 6 and AI Twister to 'Moderate'. Some minutes ago I cleaned the heatpipe and I hope this will help. ;)

Are you able to run at 400 MHz FSB with stock MCH voltage?

With e8500 I'm able to run 500fsb 1:1 ram, pl7, ai twister on moderate using stock voltage on mch.
08-13-2009, 11:17 AM
BlackArchon

Nevermind, it seems that I destroyed my board... after remounting the heatpipe, Vista won't start and Memtest shows thousands of errors with several memory modules. :(
08-17-2009, 02:23 AM
Alien Grey

My new Q9650 arrived today.:up: It's a new batch L847C159. I can't find anything about this batch on the net.

I don't know what the VID is because I don't have the time to put it on my board.
08-17-2009, 02:27 AM
KRich

Quote:

Originally Posted by A-Grey

My new Q9650 arrived today.:up: It's a new batch L847C159. I can't find anything about this batch on the net.

I don't know what the VID is because I don't have the time to put it on my board.

Best of luck buddy :up: Can't wait to hear the good news :D
08-17-2009, 02:39 AM
Alien Grey

Quote:

Originally Posted by KRich

Best of luck buddy :up: Can't wait to hear the good news :D

Thanks mate. I'll try it this evening when I'm home. I've got to go to work.
08-17-2009, 03:18 AM
KRich

Quote:

Originally Posted by A-Grey

Thanks mate. I'll try it this evening when I'm home. I've got to go to work.

Have fun buddy :up: I hope you've got one of those low VID wonders with you. Will look forward to seeing your results.
08-17-2009, 01:36 PM
Alien Grey

I've found the golden batch.:banana:

Batch L847C159 VID 1.1875V.:up:
08-17-2009, 02:06 PM
KRich

Quote:

Originally Posted by A-Grey

I've found the golden batch.:banana:

Batch L847C159 VID 1.1875V.:up:

Woot!! Woot!! :up: Well done buddy, I am made up for you - Well done :D
08-17-2009, 03:10 PM
kuebk

Low VID doesn't necessary mean great oc.
08-17-2009, 03:14 PM
Alien Grey

I've been running some LinX tests at 9 X 450MHz and DDR 1199 with tRD 6.

I don't think that it's possible to run it at 4.05GHz with Vcore 1.20V- under load.

I can run 4 passes stable with 1.20V under load but than it fails.

It's something that I expected that wouldn't be possible because of the high GFlops. LinX gives me GFlops around 56.700 and I find that very impressive for 4.05GHz.

An other interesting thing is that I don't need high Vtt. I was running with 1.14V real. I did try higher, up to 1.26V, but it fails with a system crash with higher Vtt. I guess I don't need it higher for 9 X 450MHz.

What they are saying about low VID and higher temps is really true. The CPU temperature reached 57° Celsius under LinX load. The room temperature was 27° Celsius and the coolant temperature was 29° Celsius.

It's nice to have a low VID Q9650 under water with these high temperatures. I don't think that it would be possible to run it at 4GHz+ with air cooling. The CPU temperature would reach to high and result in an unstable CPU.

So far my first experiences with this Q9650.

I'm going to try to get it LinX stable for 25 passes and post some screenshots with my results. I'll do this when I've got some time to play a little more with this nice chip.
08-17-2009, 03:50 PM
Armored_Kech1

Quote:

Originally Posted by A-Grey

I've been running some LinX tests at 9 X 450MHz and DDR 1199 with tRD 6.

I don't think that it's possible to run it at 4.05GHz with Vcore 1.20V- under load.

I can run 4 passes stable with 1.20V under load but than it fails.

It's something that I expected that wouldn't be possible because of the high GFlops. LinX gives me GFlops around 56.700 and I find that very impressive for 4.05GHz.

An other interesting thing is that I don't need high Vtt. I was running with 1.14V real. I did try higher, up to 1.26V, but it fails with a system crash with higher Vtt. I guess I don't need it higher for 9 X 450MHz.

What they are saying about low VID and higher temps is really true. The CPU temperature reached 57° Celsius under LinX load. The room temperature was 27° Celsius and the coolant temperature was 29° Celsius.

It's nice to have a low VID Q9650 under water with these high temperatures. I don't think that it would be possible to run it at 4GHz+ with air cooling. The CPU temperature would reach to high and result in an unstable CPU.

So far my first experiences with this Q9650.

I'm going to try to get it LinX stable for 25 passes and post some screenshots with my results. I'll do this when I've got some time to play a little more with this nice chip.

What's the highest stable frequency you could get with 1.2v vcore under load?
08-17-2009, 07:46 PM
Ghostleader

@A-Grey

Nice m8, looking forward to see those results.
08-17-2009, 10:23 PM
OLDcomer

Quote:

Originally Posted by A-Grey

I've been running some LinX tests at 9 X 450MHz and DDR 1199 with tRD 6.

I don't think that it's possible to run it at 4.05GHz with Vcore 1.20V- under load.

I can run 4 passes stable with 1.20V under load but than it fails.

It's something that I expected that wouldn't be possible because of the high GFlops. LinX gives me GFlops around 56.700 and I find that very impressive for 4.05GHz.

An other interesting thing is that I don't need high Vtt. I was running with 1.14V real. I did try higher, up to 1.26V, but it fails with a system crash with higher Vtt. I guess I don't need it higher for 9 X 450MHz.

What they are saying about low VID and higher temps is really true. The CPU temperature reached 57° Celsius under LinX load. The room temperature was 27° Celsius and the coolant temperature was 29° Celsius.

It's nice to have a low VID Q9650 under water with these high temperatures. I don't think that it would be possible to run it at 4GHz+ with air cooling. The CPU temperature would reach to high and result in an unstable CPU.

So far my first experiences with this Q9650.

I'm going to try to get it LinX stable for 25 passes and post some screenshots with my results. I'll do this when I've got some time to play a little more with this nice chip.

Congratulations! Don't worry about the temperature. My q9650 reaches between 80-90C but runs flawlessly LinX.
08-17-2009, 10:39 PM
alexlyc

Quote:

Originally Posted by A-Grey

It's nice to have a low VID Q9650 under water with these high temperatures. I don't think that it would be possible to run it at 4GHz+ with air cooling. The CPU temperature would reach to high and result in an unstable CPU.

I've been running my Q9650 (1.15 VID) @ 4.02GHz w/ 1.22v + LLC enable for the past 3 months with TRUE + S-Flex F @ 1500rpm without overheat or BSOD. Air is possible for 4GHz, but I am on the edge in terms of max temp. That's why I just bought an Ultra Kaze for it... still testing but I think this will at least save me from going water for awhile...
08-18-2009, 12:14 AM
Alien Grey

Q9650@4.05GHz - DDR 1199MHz - tRD 6 - BIOS 0902

http://i431.photobucket.com/albums/q...D6-BIOS090.jpg

Code:

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/B [Auto] DRAM TimingControl [Auto] DRAM Static Read Control [Enabled] Ai Clock Twister [Moderate] 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 [1.33V] DRAM Voltage [1.80V] FSB Termination Voltage [1.24V] South Bridge Voltage [1.05V] SB 1.5V Voltage [1.50V] Loadline Calibration [Disabled] CPU GTL Voltage Reference [0.63X] 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] CPU Ratio Setting [9.0] C1E Support [Disabled] CPU TM Function [Enabled] Vanderpool Technology [Disabled] Execute Disable Bit [Enabled] Max CPUID Value Limit [Disabled]

OPT1 - Coolant temperature
OPT2 - Memory temperature

This is my first attempt to run it at 4.05GHz. I don't know if it's still stable with lower CPU Voltage. I'll try that later.

My mistake was to delay the DRAM CLK Skew on Channel A and B with 50ps. That made it fail after a few passes in LinX.
With the DRAM CLK Skew on Channel A and B on AUTO and my memory in slots B1 and B2 it stays perfectly stable.

This makes me start thinking that it isn't the CPU Clock Skew that made it fail with my Q9450@3.72GHz, DDR 1199MHz and tRD 6 but the DRAM CLK Skew on Channel A and B. It's almost impossible to keep it stable with the DRAM CLK Skew on Channel A and B on AUTO and Delaying both Channels with 50ps is probably to far.

Perhaps the DDR 1200MHz+ and tRD 6 fix from BIOS 0803 to 0902 didn't fix the problem completely.:shrug:

I'll see that later when I'm trying to run at 9 X 465MHz, DDR 1239MHz and tRD 6.

I'm very happy with the results so far.:D
08-18-2009, 02:43 AM
Ghostleader

Great work mate, looking good :up:
08-18-2009, 12:49 PM
Alien Grey

Quote:

Originally Posted by Armored_Kech1

What's the highest stable frequency you could get with 1.2v vcore under load?

I don't know. I just wanted it to be stable at 4.05GHz with 1.20V. I'm going to try to lower the CPU Voltage and see if I can still keep it stable in LinX.
08-18-2009, 12:54 PM
Alien Grey

Quote:

Originally Posted by OLDcomer

Congratulations! Don't worry about the temperature. My q9650 reaches between 80-90C but runs flawlessly LinX.

That's good to know. I've been reading the posts in the Q9650 thread and some of the guys seem to have problems to keep it stable when the CPU reaches 65° Celsius. They are using that other brand boards.:D
08-18-2009, 01:03 PM
Alien Grey

Quote:

Originally Posted by alexlyc

I've been running my Q9650 (1.15 VID) @ 4.02GHz w/ 1.22v + LLC enable for the past 3 months with TRUE + S-Flex F @ 1500rpm without overheat or BSOD. Air is possible for 4GHz, but I am on the edge in terms of max temp. That's why I just bought an Ultra Kaze for it... still testing but I think this will at least save me from going water for awhile...

I know these low VID chips needs good cooling to run them at 4GHz+. If you can't cool them enough you're better of with a higher VID chip.
08-18-2009, 01:09 PM
Alien Grey

Quote:

Originally Posted by Ghostleader

Great work mate, looking good :up:

Thanks mate.
08-18-2009, 09:44 PM
leo27

a-grey
good to hear you finally got a decent chip.

Just a quick question, i just got UT2004 to do a final stress test on my system to check the FSB termination voltage (stable for 25 pass linx a full run through of large ffts and a full run through of small fft's)
Just to clarify when i run Ut2004 + P95 in blend, what will i be looking for i.e will P95 fail or will something else happen???
08-18-2009, 11:01 PM
Alien Grey

Quote:

Originally Posted by leo27

a-grey
good to hear you finally got a decent chip.

Just a quick question, i just got UT2004 to do a final stress test on my system to check the FSB termination voltage (stable for 25 pass linx a full run through of large ffts and a full run through of small fft's)
Just to clarify when i run Ut2004 + P95 in blend, what will i be looking for i.e will P95 fail or will something else happen???

Prime95 can fail or UT2004 crashes and in the worst case a full system crash.
08-18-2009, 11:09 PM
Alien Grey

For 4.05GHz a CPU Voltage of 1.20V under load is the lowest I can go. The computer reboots or LinX fails immediately with lower CPU Voltage.
08-19-2009, 01:49 AM
leo27

thanks a-grey i'll give it a shot later in the week

Did you end up getting your ram to run higher than 1200mhz, and whats your final thoughts on bios 902?
08-19-2009, 02:02 AM
Alien Grey

Quote:

Originally Posted by leo27

thanks a-grey i'll give it a shot later in the week

Did you end up getting your ram to run higher than 1200mhz, and whats your final thoughts on bios 902?

Yes, I did. It runs fine at DDR 1239MHz with 1.9V. I'll try to run my Q9650 at 9 X 465MHz later to see if I can get it to run stable with DDR 1239MHz and tRD 6 or tRD 7 and all Phase pull-ins enabled.
08-19-2009, 03:17 AM
leo27

is there any benefit going to bios 902 from 803??
08-19-2009, 03:57 AM
kuebk

I don't think so.

EDIT: A-Grey what mch voltage do you need to stabilize 400fsb, 600mem 5-5-5-15 trd6? Could you please check it for me? Looks like I need 1.45v what looks pretty much for me.
08-20-2009, 07:59 AM
Armored_Kech1

Has anyone tried removing the ROG logo from the stock SB heatsink? Is it even possible? It looks like it's severely hindering the SB's heat dissipation.
08-20-2009, 08:26 AM
ShArKo

Quote:

Originally Posted by Armored_Kech1

Has anyone tried removing the ROG logo from the stock SB heatsink? Is it even possible? It looks like it's severely hindering the SB's heat dissipation.

I do, Is easy to remove, but you must do it with a hear dryer (be carefull) and pull from the heatsink.

This you must do with hold headsink out from the motherboard.
08-20-2009, 12:39 PM
Alien Grey

Quote:

Originally Posted by leo27

is there any benefit going to bios 902 from 803??

Only if you need DDR 1200MHz+ and tRD 6.

I've been running for moths now with BIOS 0902 and everything runs perfectly stable.
08-20-2009, 12:54 PM
Alien Grey

Quote:

Originally Posted by kuebk

EDIT: A-Grey what mch voltage do you need to stabilize 400fsb, 600mem 5-5-5-15 trd6? Could you please check it for me? Looks like I need 1.45v what looks pretty much for me.

I've been running a few tests and I didn't need more voltage on the MCH to run LinX for a few passes stable.

I did a test with CPU Clock Skew Delayed 200ps and NB Clock Skew Delayed 100ps. It stayed stable but there was no need to dial in a delay.

I tryed with CPU Clock Skew and NB Clock Skew on Auto and everything was running smooth.

I have to say that I only did 3 passes in LinX but I think it should be enough to see if it's really possible with lower voltage.

The 266MHz strap isn't the problem. It must be something else. Perhaps your MCH temp is to high (my MCH temp was 34° Celsius during the test) or you need that high voltage to keep your memory stable at 1200MHz.

You could try with your memory running at default speed and see if that would help to keep it stable with lower voltage. When that doesn't work try to find a way to lower the MCH temp to see if that's the problem.
08-20-2009, 11:34 PM
Armored_Kech1

Quote:

Originally Posted by ShArKo

I do, Is easy to remove, but you must do it with a hear dryer (be carefull) and pull from the heatsink.

This you must do with hold headsink out from the motherboard.

So if I understand correctly the only way to safely remove the ROG logo is with the heatsink removed from the motherboard? My UD3P's NB plate came off with just a screw driver because it only uses double sided tape.
08-21-2009, 02:51 AM
kuebk

Quote:

Originally Posted by Armored_Kech1

So if I understand correctly the only way to safely remove the ROG logo is with the heatsink removed from the motherboard? My UD3P's NB plate came off with just a screw driver because it only uses double sided tape.

This should too, haven't tried myself but afair someone already done this that way.
08-21-2009, 03:33 AM
ShArKo

Quote:

Originally Posted by Armored_Kech1

So if I understand correctly the only way to safely remove the ROG logo is with the heatsink removed from the motherboard? My UD3P's NB plate came off with just a screw driver because it only uses double sided tape.

You can do it with the heatsink on.........but do it at your own risk...
The think is, that logo is very "glue" to the rest of the heatsink, so if you force the logo to take it off you can break something (like the SB of NB)...

Good Luck!!!:up:
08-21-2009, 04:24 AM
Armored_Kech1

Quote:

Originally Posted by kuebk

This should too, haven't tried myself but afair someone already done this that way.

Quote:

Originally Posted by ShArKo

You can do it with the heatsink on.........but do it at your own risk...
The think is, that logo is very "glue" to the rest of the heatsink, so if you force the logo to take it off you can break something (like the SB of NB)...

Good Luck!!!:up:

Alright, thanks. :)
08-21-2009, 08:53 AM
hutkikz

I've removed 2 of them with a screwdriver no problem at all.:up:
08-21-2009, 10:56 PM
leo27

Quote:

Originally Posted by leo27

i've been reading this forum ever since i got my rampage formula, finally i can post after a few registration issues.

my rig is:
Antec P182
asus rampage formula x48 Bios 803
q6600 G0 @ 3.4Ghz
Gskill F2-6400CL4-2GBPI-B @850Mhz
TRUE with 2 noctua P12 Push/Pull

my settings (25 Linx 64-bit stable, Prime95 small and large fft's stable)

*Extreme Tweaker*
Ai Overclock Tuner [Manual]
OC From CPU Level Up [Auto]
CPU Ratio Setting [08.0]
FSB Strap to North Bridge [400]
FSB Frequency [425]
PCIE Frequency [100]

DRAM Frequency [DDR2-850MHz]
DRAM Command Rate [2N]
DRAM CMD Skew on Channal A [Greyed Out]
DRAM CMD Skew on Channal B [Greyed Out]
DRAM CLK Skew on Channal A [Auto]
DRAM CLK Skew on Channal B [Auto]

DRAM Timing Control [Manual]
1st Information
CAS# Latency [4]
RAS# to CAS# Delay [4]
RAS# PRE Time [3]
RAS# ACT Time [9]
RAS# to RAS# Delay [3]
REF Cycle Time [40]
WRITE Recovery Time [6]
READ to PRE Time [3]
2nd Information
READ to WRITE Delay (S/D) [7]
WRITE to READ Delay (S) [3]
WRITE to READ Delay (D) [4]
READ to READ Delay (S) [4]
READ to READ Delay (D) [5]
WRITE to WRITE Delay (S) [4]
WRITE to WRITE Delay (D) [5]
3rd Information
WRITE to PRE Delay [12]
READ to PRE Delay [3]
PRE to PRE Delay [1]
ALL PRE to ACT Delay [4]
ALL PRE to REF Delay [4]

DRAM Static Read Control [Enabled]
Ai Clock Twister [Stronger]
Ai Transaction Booster [08]
Pull-in of CHA PH1 [Enabled]
Pull-in of CHA PH2 [Enabled]
Pull-in of CHA PH3 [Enabled]
Pull-in of CHA PH4 [Enabled]
Pull-in of CHB PH1 [Enabled]
Pull-in of CHB PH2 [Enabled]
Pull-in of CHB PH3 [Enabled]
Pull-in of CHB PH4 [Enabled]

CPU Voltage [1.41250v] (1.4V in windows)
CPU PLL Voltage [1.50v]
North Bridge Voltage [1.37v]
DRAM Voltage [1.84v]
FSB Termination Voltage [1.32v]
South Bridge Voltage [1.05v]
SB 1,5V Voltage [1.50v]
Loadline Calibration [Enabled]
CPU GTL Voltage Reference [0.65x]
NB GTL Voltage Reference [0.67x]
DRAM Controller Voltage REF [AUTO]
DRAM Channel A Voltage REF [AUTO]
DRAM Channel B Voltage REF [AUTO]

CPU LED Selection [CPU Voltage]
NB LED Selection [North Bridge Voltage]
SB LED Selection [South Bridge Voltage]
Voltiminder LED [Enabled]

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

Now i have run into a problem in that i was unable to reduce the memory timings (4-4-4-12 at 1.92V in bios) with a PL7, however if i run PL8 with all phase pull-ins as above then i can achieve the timings and voltage shown in the settings above (4-4-3-9 at 1.84V).

I take it PL8 with all Pull-ins is equivalent to PL7?????

to A-GREY
I've read that you use UT2004 to determine the FSB termination voltage. What settings do you use in UT2004????
and do you run it with Prime95 small fft's or large fft's????
an explanation of the procedure would be appreciated.

Lastly is a CPU GTL of 0.65x recommended for a 65nm processor or is 0.63x advised.

Thanks for your help and keeping up with this thread. It is a invaluable resource for overclocking this board.

I've just been testing out my system with UT2004+P95 in blend and the problem i have is that P95 keeps failing.

As you can see above i am stable for 5hrs on large ffts and 5hrs on small fft's as well as 25 passes of linx. I should also say that the memory has passed 8 tests of Memtest86+ test no5 as well as 400% of HCI memtest.

i have tried increasing the FSB termination.
1.30V lasted 5hrs
1.32V lasted about 2hrs
1.34V lasted about 2hrs

after seeing this i decided to try lowering FSB termination to see if higher wasn't better

1.26V lasted 1hr

I would appreciate any advice as to what is causing the instability with P95 and UT2004 or do i just need to play with FSBterm some more????
08-22-2009, 01:10 AM
Alien Grey

Quote:

Originally Posted by leo27

I've just been testing out my system with UT2004+P95 in blend and the problem i have is that P95 keeps failing.

As you can see above i am stable for 5hrs on large ffts and 5hrs on small fft's as well as 25 passes of linx. I should also say that the memory has passed 8 tests of Memtest86+ test no5 as well as 400% of HCI memtest.

i have tried increasing the FSB termination.
1.30V lasted 5hrs
1.32V lasted about 2hrs
1.34V lasted about 2hrs

after seeing this i decided to try lowering FSB termination to see if higher wasn't better

1.26V lasted 1hr

I would appreciate any advice as to what is causing the instability with P95 and UT2004 or do i just need to play with FSBterm some more????

If you lower the FSB termination Voltage you should check it first with LinX to see if that stays stable.

Where does it fail in Blend Mode? Is it Small or Large FFT's?
08-22-2009, 01:41 AM
leo27

this is the log for the test (P95 blend + UT2004 in 64-bit) with FSBterm at 1.34V (known stable in Linx)

[Sat Aug 22 13:36:16 2009]
Self-test 1024K passed!
Self-test 8K passed!
Self-test 8K passed!
Self-test 8K passed!
[Sat Aug 22 13:53:18 2009]
Self-test 10K passed!
Self-test 10K passed!
Self-test 10K passed!
[Sat Aug 22 14:08:34 2009]
Self-test 896K passed!
Self-test 896K passed!
Self-test 896K passed!
Self-test 8K passed!
[Sat Aug 22 14:23:32 2009]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
Self-test 768K passed!
Self-test 768K passed!

I assume that means it failed in LargeFFt's
08-22-2009, 01:55 AM
Alien Grey

Quote:

Originally Posted by leo27

this is the log for the test (P95 blend + UT2004 in 64-bit) with FSBterm at 1.34V (known stable in Linx)

[Sat Aug 22 13:36:16 2009]
Self-test 1024K passed!
Self-test 8K passed!
Self-test 8K passed!
Self-test 8K passed!
[Sat Aug 22 13:53:18 2009]
Self-test 10K passed!
Self-test 10K passed!
Self-test 10K passed!
[Sat Aug 22 14:08:34 2009]
Self-test 896K passed!
Self-test 896K passed!
Self-test 896K passed!
Self-test 8K passed!
[Sat Aug 22 14:23:32 2009]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
Self-test 768K passed!
Self-test 768K passed!

I assume that means it failed in LargeFFt's

Try with 1 step higher NB Voltage or Memory Voltage or both.

Edit: If all goes well I finish mine at 4.05GHz in a few hours.:D
08-22-2009, 02:35 AM
leo27

will do thanks A-grey, and thats awesome news hope it goes well
08-23-2009, 02:18 AM
leo27

Increasing the NB voltage by 1 step to 1.39V worked a treat. I'm stable now for 10hrs with FSB termination at 1.30V. When i have time i'll try to lower the FSB termination.

http://img8.imageshack.us/img8/5484/p95ut2004jpg.th.jpg
08-23-2009, 04:00 AM
Alien Grey

Quote:

Originally Posted by leo27

Increasing the NB voltage by 1 step to 1.39V worked a treat. I'm stable now for 10hrs with FSB termination at 1.30V. When i have time i'll try to lower the FSB termination.

http://img8.imageshack.us/img8/5484/p95ut2004jpg.th.jpg

Nice, good work.:up:
08-23-2009, 08:53 AM
kuebk

Can anyone confirm hr-09 + rampage formula compatybility list, link: http://www.thermalright.com/new_a_pa.../05-09-mcl.htm?
08-23-2009, 10:17 AM
Alien Grey

Quote:

Originally Posted by kuebk

Can anyone confirm hr-09 + rampage formula compatybility list, link: http://www.thermalright.com/new_a_pa.../05-09-mcl.htm?

The Thermalright Mosfet Coolers fit on the ASUS Rampage Formula. I have them on my board. You need a HR-09 U Type 2 and a HR-09 S Type 2.
08-23-2009, 10:29 AM
kuebk

A-Grey can you make a photo of how does it look mounted on mobo?
08-23-2009, 11:25 AM
Alien Grey

Quote:

Originally Posted by kuebk

A-Grey can you make a photo of how does it look mounted on mobo?

There are a few pictures of my system on the ASUS Rampage Formula forum. They aren't the best quality pictures but I don't have a good digital camera.

You can find them here: http://vip.asus.com/forum/view.aspx?...e=en-us&page=2
08-25-2009, 02:37 PM
kuebk

Is there any possibility to mod 802 bios so by default it would be setting the highest possible tREF (16k) instead of 3120?

Show 100 post(s) from this thread on one page

All times are GMT -8. The time now is 10:09 PM.

XtremeSystems