ASUS Maximus II Formula - new P45 king?

[elsupremo]

I upgraded from 8800GTX to 4870 X2 without any issues. First thing I did was uninstall the Nvidia drivers, power down, swap the cards and powered up into Windows. Installed the Catalyst Software Suite 8.8 (54.3MB) rebooted and that's that.

I was using bios 0901 at the time and have since flashed to 1301 and have still not run into any problems. My NB voltage is set to 1.28 in the bios.

Has your PSU got 8 pin PCIE connectors or are you using an adaptor? What's your PCIE Freq in bios set to? 100? Sorry for the obvious questions - Gotta cover all the angles.

Thank you for sharing.
I am using an adaptor that combines 2 6 pin PCIE plugs. My PSU came with 4 6 pin PCIe cables.
Not once have I messed with NB voltage. Perhaps that is an issue?

PCI Freq always set to 100.

Any help is appreciated!

[brenanh]

Won't hurt to try 1.3v NB, you never know! :-)

Gotta be either that or the power feed I'd say....

[Grnfinger]

@Grnfinger,
What is the stepping on your Q9550?

C1 but its still better than the E0 8600 I have, that was alot of hype for nothing other than to drain my wallet

Thank you for sharing.
I am using an adaptor that combines 2 6 pin PCIE plugs. My PSU came with 4 6 pin PCIe cables.
Not once have I messed with NB voltage. Perhaps that is an issue?

PCI Freq always set to 100.

Any help is appreciated!

Do you have another mobo to run the 4870X2 card on?
the fact that it wont run with stock bios settings is disturbing, and makes me lean towards the card. Could be the PSU but if it runs the nvidia card it makes me still lean towards the 4870
You could try upping NB volts but I doubt its the issue.
What brand of card is it?

[64dragon]

Thank you for sharing.
I am using an adaptor that combines 2 6 pin PCIE plugs. My PSU came with 4 6 pin PCIe cables.
Not once have I messed with NB voltage. Perhaps that is an issue?

PCI Freq always set to 100.

Any help is appreciated!

did you uninstall the nvidia drivers before putting in the 4870? the nvidia card may not care if the ati drivers are there but the ati card may not like it if the nvidia drivers are still installed. like the others that comented, just throwing out random and easily over looked stuff

[The Nemesis]

@elsupremo your issue is simple. Your psu doesn't have sufficient power to maintain your overclock with the 4870X2 in. I ran into the same problem when I upgraded from my BFG 8800GTX to BFG 280GTX. All I did was drop the card in & load drivers. Thought the system was stable it was not prime stable. I tried upping the voltages a bit with no success. Then I took a look @ my prime screenshots with the previous configuration and noticed the 12v rail idled below 12v with the 280GTX. I upgraded my 700w OCZ to a 1000w Antec(was on sale) and the problem went away. All my previous saved prime bios configurations worked. My advice would be to try a new psu perhaps a Corsair model.

[elsupremo]

Thank you for those of you that have offered advice so far. I really appreciate your thoughts. I went out and bought a Corsair 1000W, thinking that PSU problems were the only thing that made sense. The new PSU did not change a single result. I suppose at this point I have to look at the card, but why would the card be able to run 3dmarks (and get a very nice score, mind you - 22.4K) if it has a problem? I should note that when the OC fails, invariably, the LCD shows it is hung on "VGA BIOS". When this card is in, at stock speeds I can get into Windows and run some stuff (no artificating, looks correct/beautiful), but like I said, it still isn't stable. Perhaps this brand's BIOS on this card is incompatible here? Maybe the other 4870 X2 users with this board are using a different brand. I'm clutching at straws, but the options here are narrowing.

With the Corsair in, I did try some different settings (different FSB to achieve similar speed, increased NB voltage) and nothing helped. Let me know if you have any other ideas, or if there are any settings I should change that might be causing this phenomenon. Thank you! Otherwise, the Corsair and the X2 are going back to the store.

P.S. It is a Diamond 4870 X2, btw.

[warpete]

I've now had my Maximus II Formula board for a few weeks. At first I simply installed everything and then started my burn-in process. My NB temp went through the roof (above 80ºC). I should have known better. I removed the heatsink and there was the normal Pond Scum Compound imported from some swamp in Louisianna. I chipped that away that and applied MX2 compound--but I noticed something I didn't like AT ALL (Gigabyte has done the same thing). First, you have your naked NB and apply compound to it. Then, the heat-pipe apparatus is installed which places a copper block on the NB. Lastly--ANOTHER layer of thermal compound on the copper block--and then the finned heat-sink is installed. A very reliable review and testing site really slamed Gigabyte for doing this exact thing, but never mentioned Asus. What kind of cooling can we expect when the heat must travel through TWO layers of compound? Should I decide to watercool my NB, I would have to completely remove the entire heat-pipe assembly which covers the NB, SB and MOSFETS etc.. I left the heat-pipe as is, and I have two layers of MX2 instead of the rock-hard Pond Scum on the NB. I attached a small fan to blow directly in the NB Heatsink fins to help remove some of the heat. Now, as I type this, my NB temp is at 52ºC and my SB is 49ºC. My Intel E8600 CPU is reported at 21ºC if you can believe that, while my MB is at 34ºC. PC Probe absolutely sucks. No matter what I do I can not get the Asus AI Suite to load (Windows XP) and the Six Engine freezes when I try to "Calibrate". This is on a clean install of XP Pro. Asus software is pathetic, to say the least. So---I am going to uninstall all the Asus software and do any overclocking directly in the 0901 BIOS which is fairly easy. I wanted to let people know that if their intentions are to watercool anything other than their CPU and GPU, they better take a closer look at how to do it. I must say, though, that this board has been rock-solid stable regardless of temps (which are very acceptable now). Right now, on air--I can easily top 4GHz with this E8600 CPU (E0 Stepping). This is with NO voltage adjustments (except RAM)whatsoever. Of course, I'm using Cell Shock Memory (MemoryC.com) and I believe it's the best in the world at the moment. The RAM heatsinks are BOLTED together--none of this thermal cement crap. Anyway, the Asus Maximus II Formula is quite possibly the best MB Asus has manufactured in a long, long time. I can easily deal with the NB Temp problem. PC Pitstop full tests reports my 'Puter to be in the top 2%, and I just have an ATI HD4870 (not HD4870x2). This board can keep up with the X38's and x48's. Even with Crossfire not being a TRUE 16x2 on this board (it reverts to 8x2), the benchmarking of Crossfire on this board and on the X48 boards is still very, very close. I am very, very happy with this Asus Maximus II Formula board!

[chiman2]

My temps are fine. I idle at 40 on both northbridge and southbridge.

[The Nemesis]

@elsupremo

Are you comfortable flashing the bios of your video card? I wonder if flashing to an asus bios would help?

**Edit**

Aren't any @ techpower up but there are ATI and others

[warpete]

Seems stock NB & Mosfets heatsinks are removable without removing heatpipes!
Sweat, but do you think MCW30 & MIPS P5E Mosfets waterblocks will fit on top of heatpipe?

You can do what you are asking, but I would not suggest it. First--the NB. The heat-pipe assembly is one large piece. The NB gets compound (get rid of the Pond scum and apply your own compound) and a a large copper block encased in the heat pipe assembly sits on the NB. Then, you have ANOTHER layer of compound to apply before installing the finned heatsink on top of the copper block, (which is installed on top of the NB). That's TWO layers of compound that heat must try to infiltrate. Gigabyte has the same setup. The MOSFETT finned heatsinks sit on TOP of the HEATPIPE and mine had little contact (also Pond scum). This means there is also compound UNDER the heatsink to make contact with the MOSFETTS. Another TWO layers of compound. So, the only way I can see to efficiently install your own MCW30 etc., is to completely remove the heatpipe assembly. Let me add that after chipping away the patented Asus Pond Scum imported from a secret Pond in Louisanna, and applying Arctic Cooling MX2 compound (to ALL areas), my temps drastically improved. I mean DRASTICALLY! Please don't misunderstand--the Asus Maximus II Formula is one of the best boards Asus has ever made and it can easily keep up with the X38/X48 boards. The heat-pipe assembly is very, very pretty, but really is not as effective as Asus would like you to believe. For instance, with my board--the solid copper block embedded in the heat-pipe assembly (for the NB), was installed at a slight angle, meaning it wasn't flat. It didn't make complete contact with the NB chip and in order to have the NB finned heatsink make complete contact with the top of the copper block, I had to put thermal compound on very think on the "lower" side. I eventually was able to force the copper block to lay flat, so---just beware. I think most buyers of this board would probably be just fine the way it is straight out of the box, but I know from experience of the "Pond Scum" issue, and I want the best thermal solution I can get. I have no problems with the current heat-pipe assembly, getting my Intel E8600 CPU past 4GHz on air with no voltage adjustments other than RAM. That is really SWEET. Soon, I will cool everything with water and go for the 5GHz mark.

[barong]

@Grnfinger

i already tried this setting, but it wont pass dual super pi 32mb , any other setting? to reach 500x8.5 or 525x8 for 24/7 im using 1307 bios

CPU Ratio Setting: 8
FSB Strap to NB: 333 (keep the 333 strap)
FSB Freq: 500
PCIE Freq: 100
DRAM Freq: 1000
DRAM Command Rate : 2T
DRAM Timing Control: Manual
CAS# Latency : 5
RAS# to CAS# Delay : 5
RAS# Precharge : 5
RAS# ActivateTime : 15
RAS# to RAS# Delay : 3
Row Refresh Cycle Time : 55
Write Recovery Time : 6
Read to Precharge Time : 3

Read to Write Delay (S/D) : 8
Write to Read Delay (S) : 3
Write to Read Delay (D) : 5
Read to Read Delay (S) : 4
Read to Read Delay (D) : 6
Write to Write Delay (S) : 4
Write to Write Delay (D) : 6

Write to PRE Delay : 14
Read to PRE Delay : 5
PRE to PRE Delay : 1
ALL PRE to ACT Delay : 5
ALL PRE to REF Delay : 5

DRAM Static Read Control: Auto
Dram Read Training : Auto
MEM OC Charger : Auto
Ai Clock Twister : Stronger
Transaction Booster : Manual

Common Performance Level [10] once stable reduce this value to 9 or 8

Pull-In of CHA PH1 Disabled
Pull-In of CHA PH2 Disabled
Pull-In of CHA PH3 Disabled
Pull-In of CHA PH4 Disabled
Pull-In of CHA PH5 Disabled
Pull-In of CHB PH1 Disabled
Pull-In of CHB PH2 Disabled
Pull-In of CHB PH3 Disabled
Pull-In of CHB PH4 Disabled
Pull-In of CHB PH5 Disabled


CPU volt:1.35 and work up, you may need 1.40 if your chip scales poorly
CPU PLL Voltage: 1.53975
NB Voltage: 1.39150
DRAM Voltage: 2.17100
FSB Termination Voltage: 1.33850
SB Voltage: 1.1
SB 1.5 Voltage: 1.5

CPU GTL Reference 0 : +40mv
CPU GTL Reference 1 : +30mv
CPU GTL Reference 2 : +30mv
CPU GTL Reference 3 : +40mv
North Bridge GTL Reference : AUTO
DDR2 Channel A REF Voltage : AUTO
DDR2 Channel B REF Voltage : AUTO
North Bridge DDR Reference : AUTO

Load Line Calabration : Enabled
CPU Sread Spectrum : Disabled
PCIE Spread Spectrum : Disabled

[kwalker]

C1 but its still better than the E0 8600 I have, that was alot of hype for nothing other than to drain my wallet

Same experience here with my C1 stepping 9550
a lot of CPU for the money

[The Nemesis]

I wish they'd just make a motherboard that could hit 500FSB 12-24hr Prime stable with 45nm quad. Our systems are fsb limited for hitting 4GHz. If a cpu can do 3.6GHz with 1.25v @load typically it would do 4GHz @ 1.4v-1.45v if it didn't hit a wall.

[kwalker]

I wish they'd just make a motherboard that could hit 500FSB 12-24hr Prime stable with 45nm quad. Our systems are fsb limited for hitting 4GHz. If a cpu can do 3.6GHz with 1.25v @load typically it would do 4GHz @ 1.4v-1.45v if it didn't hit a wall.

They are doing better
The only wall I seem to hit is at 500 FSB.
I can post and play around in windows but Its short lived.
A least 500 is possible on a 45 nm quad.

[The Nemesis]

I can't boot @ 500 but I can use setfsb or aisuite to get there but no where near stable not even for Pi. It's just cheaper than buying a higher multiplier cpu. I'm not gonna purchase another 45nm Quad. I'll probably go through a few E0 8X00's though before & after nethlam. One thing I've noticed though. The P45 clocks the 45nm quad with alot less voltage than the X38 P35 or nividia offerings. It's definitely a good chipset if your looking for features efficiency and MHz. Not so good performance at same clocks vs the others though.

[kwalker]

The only thing I miss about my Maximus formula is the 16X Crossfire .
you loose a little at 8X + 8X but make up the difference in raw MHz

[CryptiK]

My maximus II formula just turned up. So did my 2 kits of D9 GKX equipped cellshock. Should be a good night

I hope this board is fully functional unlike my P5Q-Deluxe was.

[vaderat-at]

I suppose at this point I have to look at the card, but why would the card be able to run 3dmarks (and get a very nice score, mind you - 22.4K) if it has a problem? I should note that when the OC fails, invariably, the LCD shows it is hung on "VGA BIOS". When this card is in, at stock speeds I can get into Windows and run some stuff (no artificating, looks correct/beautiful), but like I said, it still isn't stable. Perhaps this brand's BIOS on this card is incompatible here? Maybe the other 4870 X2 users with this board are using a different brand. I'm clutching at straws, but the options here are narrowing.

Mine is a Powercolor brand. Maybe the best thing to do is try another 4870 X2 card if possible. Let us know how you get on.

I see some people setting the NB to 1.39. Isn't that quite high for P45 chipset?
I thought they could run high FSB with around 1.24 - 1.30.

[Grnfinger]

You can do what you are asking, but I would not suggest it. First--the NB. The heat-pipe assembly is one large piece. The NB gets compound (get rid of the Pond scum and apply your own compound) and a a large copper block encased in the heat pipe assembly sits on the NB. Then, you have ANOTHER layer of compound to apply before installing the finned heatsink on top of the copper block, (which is installed on top of the NB). That's TWO layers of compound that heat must try to infiltrate. Gigabyte has the same setup. The MOSFETT finned heatsinks sit on TOP of the HEATPIPE and mine had little contact (also Pond scum). This means there is also compound UNDER the heatsink to make contact with the MOSFETTS. Another TWO layers of compound. So, the only way I can see to efficiently install your own MCW30 etc., is to completely remove the heatpipe assembly. Let me add that after chipping away the patented Asus Pond Scum imported from a secret Pond in Louisanna, and applying Arctic Cooling MX2 compound (to ALL areas), my temps drastically improved. I mean DRASTICALLY! Please don't misunderstand--the Asus Maximus II Formula is one of the best boards Asus has ever made and it can easily keep up with the X38/X48 boards. The heat-pipe assembly is very, very pretty, but really is not as effective as Asus would like you to believe. For instance, with my board--the solid copper block embedded in the heat-pipe assembly (for the NB), was installed at a slight angle, meaning it wasn't flat. It didn't make complete contact with the NB chip and in order to have the NB finned heatsink make complete contact with the top of the copper block, I had to put thermal compound on very think on the "lower" side. I eventually was able to force the copper block to lay flat, so---just beware. I think most buyers of this board would probably be just fine the way it is straight out of the box, but I know from experience of the "Pond Scum" issue, and I want the best thermal solution I can get. I have no problems with the current heat-pipe assembly, getting my Intel E8600 CPU past 4GHz on air with no voltage adjustments other than RAM. That is really SWEET. Soon, I will cool everything with water and go for the 5GHz mark.

I would have to disagree with you, my MCW30 sits rather nicely on my NB thank you,
32c idle and 38c load temps tell me my waterblock is doing the job rather well.
I'm not sure what exactly your refurring to about the thermal paste Asus applied. It wiped right off without any effort. No chipping required

[Nunzi]

Lanport 1 just died on me, should I be concern.............. enabled port 2 all is working fine

[CryptiK]

I have my board set up now and I'm stable at 4GHz / DDR 1113 MHz, however I have an issue.

When I restart the pc from windows, it fails to post, I just get a black screen. It also happens when I go into the bios for a second time and try to alter a setting. After the cmos has been cleared, I can set the 4GHz overclock, and it reboots fine, just anytime after that it wont. Also when I power off the pc & go to turn it on again it wont post. I also noticed when it doesn't post it doesnt cycle the voltage lights like it does while posting normally.

I have tried bioses 0504, 0604, 1307 and they all do it.

Any advice?

[FinnEire]

I have my board set up now and I'm stable at 4GHz / DDR 1113 MHz, however I have an issue.

When I restart the pc from windows, it fails to post, I just get a black screen. It also happens when I go into the bios for a second time and try to alter a setting. After the cmos has been cleared, I can set the 4GHz overclock, and it reboots fine, just anytime after that it wont. Also when I power off the pc & go to turn it on again it wont post. I also noticed when it doesn't post it doesnt cycle the voltage lights like it does while posting normally.

I have tried bioses 0504, 0604, 1307 and they all do it.

Any advice?

I know it is not the same board but my P5Q-E did the exact same when I was either too aggressive with my RAM timings @ 4.2 GHz (PL set too low or Clock twist set at anything more than "moderate"), or not enough volts to either NB or memory.

[vaderat-at]

When I restart the pc from windows, it fails to post, I just get a black screen. It also happens when I go into the bios for a second time and try to alter a setting. After the cmos has been cleared, I can set the 4GHz overclock, and it reboots fine, just anytime after that it wont. Also when I power off the pc & go to turn it on again it wont post. I also noticed when it doesn't post it doesnt cycle the voltage lights like it does while posting normally.

What FSB strap and CPU multi are you using?
I have noticed with my setup if I use 1:1 FSB and the default multi I have had the same problem as you. I fixed that by running 266MHz strap or Auto. Seems to have done the trick.

[CryptiK]

The 24/7 setting I want to run is 445 FSB, 9x multi and 266 strap giving 1116MHz ram speed.

I have tried all multi's from 6 to 9, but I have not set the strap manually, but I have tried 1:1, 4:5 etc memory dividers by selecting various ram speeds with strap set to AUTO (which should be changing the strap depending on what ram speed I have selected) and it still does it all the time - no matter what I set it refuses to reboot.

I could try selecting the strap manually but I doubt that will change anything.

[vaderat-at]

I have tried all multi's from 6 to 9, but I have not set the strap manually, but I have tried 1:1, 4:5 etc memory dividers by selecting various ram speeds with strap set to AUTO (which should be changing the strap depending on what ram speed I have selected) and it still does it all the time - no matter what I set it refuses to reboot.

I still think it could be a multi / strap issue. What's NB voltage set to in the bios?
Try disabling DRAM static read control if it isn't already.