ASUS Maximus Formula SE

[kup]

LS? liquid system?

you'll be using the Q6600?

Yeah, I'll use my 1.2375v VIS Q6600.

he means VapoChill LightSpeed not liquid system.

Bingo! Got a Rev 2 Vapo' LS so it can hold some nice temps!

oh. well that's upsetting. well, for me anyway. lol.

Why so, dude?

[TheSamurai]

Why so, dude?

i was hoping for those clocks with just regular water cooling.

[Amurtigress]

Hello,

I have had problems with the P5E and Maximus Formula that I couldn't solve so far. One thing right off, not meaning to show off but it shall answer some questions about wether I might have missed due to being an unexperienced noob...

I was building my own PCs for 17 years now and I have had an education in electronics development.

I bought my P5E board back in november, and had it replaced on warranty two weeks ago, I've got the Maximus Formula now.

What has me very confused is which NB voltage would be right for which NB/FSB clock. The reason is, the software issues I have make me believe that thise values that I would expect to be right, tell me something very different.


The issues:

Some games of the Need for Speed (especially Underground 2) series appear to crash frequently when raising the NB *OR* FSB Term. voltage above 1.32V . Whereas actually 1.45V would be more likely to be required for FSB 437 or 492 (E6600 B2 running at 8x437. 9/8*437=492 MHz)

What are your experiences on NB voltages, FSB term voltages and stability, and what are the most common effects of too much or too little voltage on this board?

What are the voltages that Intel specifies for the X38 northbridge at default clocks of 266 and 333? The AUTO setting in the BIOS gives the NB 1.35V at stock cklocks and jumps to 1.61 when overclocking beyond 333, which seems extreme to me.

Second Issue:

While running stable on Windows 2003 and XP, SecondLife running on any flavor of Vista, Vista 64 or Server 2008 freezes the whole OS sometime between 5 minutes and 6 1/2 hours after starting the secondlife viewer.
Symptoms are:

- System comes to an almost complete stop. Graphics output freezes, or the screen goes black. Keyboard input is completely dead.
- The system seems to be under full load yet not CRASHED for about 10-15 seconds. HD activity continues for a few seconds, network activity as well. Then the system comes to a complete halt. All I can do then is push the reset button. The event log never shows any entries except for "system did not shut down properly".

What I have tried so far to resolve the issue:

- change BIOS settings, including CPU voltage, NB volt, FSB term., RAM voltage...
- tried all available drivers for my 8800GTS-320 card.
- reversed all overclocks in the system to default.
- several fresh installs of Vista including "naked" setups with only the needed drivers, and all hotfixes. Also SP1.
- removed two RAM modules
- moved cards around to enforce a different IRQ assignment

Stability tests I ran:

-Orthos 2004, 6h stable in Vista and Windows 2003/XP
-Memtest 86
-OpenGL 2.0 test (fuzzy donut), 9 hours stable.
-ATiTool DirectX artefact scan, 9h without errors.


System Specs:

ASUS Maximus Formula 1.03G, E6600@437x8, 4x1 GB Corsair DDR2-800 CL5, Sparkle 8800GTS 320 MB, Creative X-Fi Fatal1ty, bottom PCI slot. Intel Pro/1000PT Network card, top PCIe slot (empty audio slot). Thermaltake Toughpower 750W PSU and a Enermax Liberty 400W (was swapped out for the Thermaltake, but the Enermax WAS sufficient!).

I guess the optical drives and HDs don't matter in this issue.

CPU Ratio Control: Manual
Ratio CMOS Setting: 8x
FSB Frequency: 437
FSB Strap to North Bridge: 400,333
PCI-E Frequency: 100,115
DRAM Frequency: 874
DRAM Command Rate : 2T
DRAM Timing Control: Manual
CAS# Latency: 5
RAS# to CAS# Delay: 5
RAS# Precharge: 5
RAS# ActivateTime: 15
Other Memory Timings: Auto
DRAM Static Read Control: DISABLED
Ai Clock Twister: Auto
Transaction Booster: Auto
CPU Voltage: 1,4625
CPU PLL Voltage: 1,50
North Bridge Voltage: various, 1.25-1.60
DRAM Voltage: 2,10
FSB Termination Voltage: various, 1.25-1.5V
South Bridge Voltage: 1,05
Loadline Calibration: Enabled
CPU GTL Reference: 0.63x
North Bridge GTL Reference: 0,67x
DDR2 Channel A REF Voltage: Auto
DDR2 Channel B REF Voltage: Auto
DDR2 Controller REF Voltage: Auto
SB 1.5V Voltage: 1,50
NB LED Selection: NB Volt
SB LED Selection: SB Volt
CPU LED Selection: CPU Volt
CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
C1E Suppport: Disabled
CPU TM Function: Disabled
Vanderpool Technology: Disabled
Execute Disable Bit: Enabled
Max CPUID Value Limit: Disabled

I could really use advice on this issue. The system has fooled me into believing I had the issue solved twice. The Secondlife problem has been gone twice for a week onto to reappear out of the blue. It has taken me 6 weeks by now, and the problem was the same on both P5E and Maximus boards.

I'm clueless.

[Nuckin_Futs]

For me, NB of 1.5 was nice for 475 ~ 505FSB with an FSBT of 1.47~1.50v , a PLL of 1.632v, SB @ 1.10v, SB 1.5 as 1.58v, and CPU/NB GTL Refs of 63/67 respectively for my E6750. (all voltages as reported and displayed in CMOS Power Management)

My temps were usually @ 41c for NB , 32c SB, and 41c on CPU when average of 500x8 @ 4GHz, They came down a notch if I lessened it to 500x7 as 3.50GHz so I can use 1.47v NB, 1.47v FSBT, 63x/63x GTL Ref, and 1.612 PLL, with temps slightly less under load but only 1~2c less at idle. NB alwqays remained under 51c during OCCT or Orthos and Prime95. Never say SB higher then 38c at 1.10v

Quad cores and RAM dividers required much more volts and created more heat neede. But for the newer 1333FSB duals, this board is very easy on them.

[Raver]

Thanx Nuckin Futs

[Nuckin_Futs]

Go figure, CMOS gets kicked in the brain by a bad flash, and then weeks later it get's zapped again and now it works, I don't know, go figure. Lol!

Just moments ago (im using it now), I wanted to read my CMOS options for CPU Strap To NB for DRAM spd options and guess what, I noticed that after returning form previous messed up 907, NB and FSBT appeared too low to post but RAM actually fine for stock, so as soon as I manually fixed the volts as before, it posted for the 1st time in 3 weeks on BIOS v0602. And good thing I left it here, as this was the only BIOS version that actually allowed me to fix the 400 strap.

So for future reference, if this looks like it is happening to you, simply adjust the NB, FSBT, and maybe RAM (as mentioned by some), before setting anything else. If it POST again, great, continue your profile settings.

I might also note that I did leave it completely out of case with no BAT in and as often as I counld on the hour clear the PWR by tapping PWR button on mobo. This helped once before on my trusty P4C800-E and an older P3V-4X. It seams to discharge any leftover charges in the caps previously stored so you get a true fresh CMOS clearing to start over.

[Amurtigress]

Nuckin futs:

Then I seriously wonder what makes NFS:U2 crash. And what's even more paradox, why don't I get any negative effects on system stability with NB 1.30/FSBt 1.30? One should expect instabilities, like bluescreens and calculation errors in Orthos. Tho I never get to see any in Vista nor Win2003/XP.

Not to talk about those stupid SecondLife issues in Vista (see Post #5591). I really need SL, and I don't see why this happens. Especially since I know a whole bunch of people running SL and Vista on a P5E, Maximus or P5B series boards.

Storm your brains peeps, maybe someone has a seemingly dumb idea that could work.

Nuckin: I have had that P5C800-E myself. I adore that board...

[tommithy]

Nuckin futs:

Then I seriously wonder what makes NFS:U2 crash. And what's even more paradox, why don't I get any negative effects on system stability with NB 1.30/FSBt 1.30? One should expect instabilities, like bluescreens and calculation errors in Orthos. Tho I never get to see any in Vista nor Win2003/XP.

Not to talk about those stupid SecondLife issues in Vista (see Post #5591). I really need SL, and I don't see why this happens. Especially since I know a whole bunch of people running SL and Vista on a P5E, Maximus or P5B series boards.

Storm your brains peeps, maybe someone has a seemingly dumb idea that could work.

Nuckin: I have had that P5C800-E myself. I adore that board...

What are your temps while loaded? cpu, nb, sb, etc..

[Nuckin_Futs]

Nuckin futs:

Then I seriously wonder what makes NFS:U2 crash. And what's even more paradox, why don't I get any negative effects on system stability with NB 1.30/FSBt 1.30? One should expect instabilities, like bluescreens and calculation errors in Orthos. Tho I never get to see any in Vista nor Win2003/XP. Nuckin: I have had that P5C800-E myself. I adore that board...

I believe it is the low volts that are crashing your system. Now I can't understand why mine would boot as (F2, load defauls) but not when set as such in CMOS. I did notice when as such low NB, FSTB in AUTO it crashed under stressfull apps or even demanding networked medial of Divx playback (quickly scanning through video, normal play was a snap).

But what's even more strange, is that this same exact set up had always loaded as default since new and up to last BIOS flash as I had always made sure that it was set to default and restarted and back to CMOS for my EZ-Flash. I will try it for test later to see if it does now.

[Amurtigress]

tommithy:
No unusual heat, around 45C for the NB. CPU up to 62C. I am using a Chieftec Mesh case and a Zalman CNPS 9700 A. So there is a good air circulation with a constant airstream from the Zalman through the 120mm fan at the back of the case.

FN: Yes, it seems plausable for the Vista/SL issue. Only why does NFSU2 get stable at 1.30NB/1.30FSBt...

I have to clarify something here: THis is not about NFSU2, which is certainly an OLD game. I just think this smells like an issue that can cause other problems behind the scenes. However, SL is important to me.

Lil typo: I meant P4C800-E

[CrackaBoi]

Hey guys.

I haven't flashed my BIOS since I got the board, which was back in November. I couldn't get my X3220/Q6600 stable at any higher speeds than 3.78Ghz back then, and I remember I blamed the BIOS divider settings mostly for that (even tho I'm happy with the speed I acheived).

So my question is, are there anyone who has experienced better OC potential with this board with the newer revision BIOSes that have been released since the board came out?

All info appreciated!

[TheSamurai]

So my question is, are there anyone who has experienced better OC potential with this board with the newer revision BIOSes that have been released since the board came out?

All info appreciated!

depends on when you got your board...

to answer your question... most likely everybody!

[CrackaBoi]

depends on when you got your board...

to answer your question... most likely everybody!

Like I said above, I was dealing with this back in November. I bought the board right away when it came out in Oct.

But thanks for the answer. So the divider settings are really improved? Any other significant improvements?

Thanks!

[Grnfinger]

Like I said above, I was dealing with this back in November. I bought the board right away when it came out in Oct.

But thanks for the answer. So the divider settings are really improved? Any other significant improvements?

Thanks!

Well I can say that divider issues are not resolved, or very little if any at all.
I personally have seen no major improvment since bios 602, basically only cpu support.
KUP has flashed back to 505 and has claimed better oc'ing and major stabilitiy improvement from 907, wich he was having issues with.

IMO unless you have a major issue and are hoping a bios update will fix, leave well enough alone

[cstkl1]

Well I can say that divider issues are not resolved, or very little if any at all.
I personally have seen no major improvment since bios 602, basically only cpu support.
KUP has flashed back to 505 and has claimed better oc'ing and major stabilitiy improvement from 907, wich he was having issues with.

IMO unless you have a major issue and are hoping a bios update will fix, leave well enough alone

hmm i still have issues with strap 400 on FSB higher than 460 on 4 sticks.

[kup]

What kinda problems with 4 sticks over 460MHz FSB?

And yes, as GrnFinger pointed out, I'm back on Bios 0505 because all my notes suggested it was the best bios thus far, and you know what, it is. Dividers work, FSB overclocks as far as newer bios if not further. I've had ONE, yes ONE, reboot since downgrading and I'm sure it was my fault for closing the window and letting the PC get too hot. This is compared to Bios 0907 with several reboots a day but can Prime/OCCT for 24 hours without crashing... :S

Oh, and this is all with lower voltage on CPU and Mobo than on Bios 0907...

Performance seems better too, Folding is as fast @ 3.6GHz on 0505 as it was @ 3.8GHz on 0907 for me!

[dude_500]

Can I downgrade to 505 from 907 within the bios upgrade system the same way I'd upgrade or is there some trick to it?

[kup]

There's a trick...

Either use the Asus Update in the first post of this thread, or...

Rename Bios 0505 to "FORMULA.ROM" (without the quotes) and load it onto a USB stick. Now flash your bios with any bios (after 0901 you'll need to do this with a 090x series bios) via the bios and turn it off as it's flashing. It'll corrupt it and fail to boot.

Plug the USB Stick containing the renamed 0505 in and power on, it'll load a Bios Flash util and reflash to 0505 for you.

This is how I did mine, saves installing crappy Asus flashers and was faster.

[Nuckin_Futs]

BIOS downgrading notes; I had no difference naming older BIOS ROM of any below of what was current (v9x or not). It always reported "BIOS is older" and could not downgrade via EZ-Flash, only through ASUS Live Update. Then it didn't matter what vesion or name, as long as you knew it was the right one you wanted and check box under "options" was selected for "old BIOS"

The "crash/corrupt" method will work, but if you have a stable install of Windows, it is just as safe and if not, well, you get the crash free recovery option after.

If using crash free BIOS recovery, it doesn't only have to be on USB, it can be the origional from ASUS CD (you flash upto what you want in EZ-Flash after) or on any CD/DVD, USB card or HDD, just so long as the ROM file is in main directory or if in folder it is in folders with all caps and no spaces (underscore is OK. It can be several folders in as long as they same as above. In this case, it (the ROM file) does need to be named as FORUMLA.ROM. It doesn't matter what version cause it will only use the 1st one it finds so try to have only one to search, or you might flash the wrong one.

As for Live update flash back, I do know for sure v6.10.22 is fully XP and S2k3 x64 compatable. I cannot speak for VISTA yet.

[kup]

In this case, it (the ROM file) does need to be named as MAXIMUS.ROM.

Nope, FORMULA.ROM. I used "formula" because that is what my driver disc uses.

[Nuckin_Futs]

Nope, FORMULA.ROM. I used "formula" because that is what my driver disc uses.

yse, that's, sorry, my miswording.

[Amurtigress]

Hi,

those of us who used a P965 board before know about the following equation:

NB clock = (max CPU multi / current CPU multi)*FSB.

Does this also apply to our X38? I am a tad confused currently, since I am running my E6600 at 437x8. this would mean the NB actually runs at close to 500 MHz and would require roughly 1.55V.

In case that FSB clock= NB core clock, I would need less voltage. Around 1.46-1.48 V...

So, any information on wether the old P965 eqation applies to the X38?

Any tools available that display the correct NB clock?

[Newms118]

just wondering if i should stay with bios 0907 or go up to 1003, i would update but that one guy had all those issues with it. 0907 worked fine on my last board thanks

also is it strange to have parts not perfectly aligned on the board, for instance i got my new one in a few days ago and the two "squares" near the ram slots, i think they are the voltage regulators, but maybe not im a noob lol, anyway the two squares are kind of off center, is this just bad manufacturing and i should just ignore it if the board runs well, or should i complain yet again

thanks

[Nuckin_Futs]

Depends on the CPU. I had plenty issues with v0907 myself with E6750. Poor and or lack of suspend mode function for Stand By and or Hibernate (S3/S4), plus random drop of USB mouse detection and failure of some USB 2.0 devices to be detected if resume from S4 actually worked.

Furthermore, I experienced nothing but weirdness since v0907 and attempting to flash went fine, but BIOS blocks were corupted ever since even when trying to move to any other BIOS version from that time on.

Now maybe my case is isolated since nobody else responded with anything to comment on, so I just RMA'd this week.

[*TLC*]

Plz guys do NOT try the 1003 bios, it's a total waste of time... luckily I was able to reflash to 0907 via EZ flash...

Issues :

1)Couldn't touch the multi ( up nor down ) resulted in a no post... had to pop battery to bring it back to life, Cmos reset didn't work

2)400FSB doesn't post with my Qx9650, 450 does but not being able to get it stable with even higher voltages then my 0907 settings...

3) Vcore didn't change for me once above 1.4volts( bios setting)-- stayed put here at 1.375 (only tried up to 1.45 in bios)...

4) Messed with the TRFC options too but that didn't help in letting her post and be stable in windows...

PLZ REMOVE IT ASAP !!

I cant confirm this. With my E8400 the Bios runs like a charm...

And the TRFc settings are godlike @55 i can run PerfLev 5 @600 Mhz 5-5-5-5-15 with 4x1GB...stable


Greetz