ASUS Maximus Formula SE

[jonny_ftm]

Hi, no Asus motherboards with heatpipes have ever had anything other than air in them. I guess that's the way Asus designed them..
Chris

Another confirmation that the whole heatpipe thing motherboard manufacturs are making is a marketting only thingy. Serious overclockers has no other way than going with third party real heatsinks

you can crossship with asus for $25 and they'll do overnight shipping.

I'm interested to know how I could do that, as my motherboard is also RMA. Should I contact ASUS directly?

This will be the last best DDR2 Motherboard around and I hope ASUS continues to support it's base of loyal customers. Any company that sticks with bleeding edge only is on the dying edge IMHO.

What an affirmation when the X48 with its native 400MHz FSB support for quadcores is soon coming, with DDR2 support despite what it was claimed (remember that also X38 is not offically supporting DDR2)

[safan80]

I'm interested to know how I could do that, as my motherboard is also RMA. Should I contact ASUS directly?

you need to setup an RMA and tell them you want to an Advanced Exchange and they will email you the paperwork and then you send it back to them.

[Nuckin_Futs]

I'm interested to know how I could do that, as my motherboard is also RMA. Should I contact ASUS directly?

What an affirmation when the X48 with its native 400MHz FSB support for quadcores is soon coming, with DDR2 support despite what it was claimed (remember that also X38 is not offically supporting DDR2)

I did the cross shipment with Abit and they required a deposit on credit card so just ask to make sure but it doesn't go through if they get the mobo in a reasonable time.

Also, Unless you are talking bout moving to DDR3 on X38 or the very few X48's the current quad core issues are likely to be the same since it has more to do with the CPU internal BUS and the way it ties in with DDR2 controller for deviders. And just to note, this mobo does FSB of 500 just as nice as 400. It was built with FSB 400 in mind just cause Intel didnt give it the stamp at realese. So if you plan to stick with DDR2, then I wouldn't stress over X48, since it is likely to be replaced by the time most of us here move to DDR3, wich for today is the only improvemens it is offering (NB RAM and devider, it is still limited to the 2 slot config for over DDR1600 for stability and with memory profiling)

[pphx459]

Hey Nuckin, what are your settings for 500x8? For the life of me I cannot get 475x8 to work. I just upgraded to 0903 bios and that didn't help. I can only get the system to post by using the white memory slots.

Just found out I can't even do 475x7...

[JH_man]

Can someone who has tried it, tell me how much colder the NB and SB will get if I use some third party cooler on them? Also will that mean both a heatsink and some noizy small fan on top of it?

In short; what are my options and how much colder will the bridges get?

Edit:
Dont also the original heatsink thingy also cool some other components? Not just the bridge chips?..

JH_man

[TouGe]

400x9? I had the QX6700 and it was just barely OK to 490x8. 4GHz was a snap as 400x10 for that one bu I wanted higher FSB. Although my P5WDG2 then was cool with multi of 4:5 for DDR1125 as 4.1 as 450x9 in 4;5 devider. I dont have it anymore but wonder if it would do a tad better on the x38 li my Q6600 that was bearely able to get to the same 480x on the 975X.

I knwo you guys dot really like 11 and high FSB, but can you try it on a 1:1 to see how much FSB it can get w/ a 1.7v + PLL and FSBT of 1.5v? I am curious if they can go voer 500FSB like most of mine did so far. My only that didnt stable out at 500 was my older E6600 (495x was it on 975x). Oh ya, this , or my mobo is good t0 540x (2160FSB) with the right CPU (E6300). So it is clear, that it is the quads holing us back on FSB. My E6750 is sable onlt so far to 515x.

Nope, my dual-core does not like high FSB at all. Trust me I've tried using 1:1 a while ago but unless I jack the CPU voltage beyond 1.5V I can't clear 415MHz on a 9 multi. This is why I'm happy to have hit 410MHz on a 9 multi with a vcore under 1.5V. If I were to use 1.55-1.6V w/ ~1.7V PLL & ~1.5V FSBT I could probably hit 450MHz but since I'm looking for a 24/7 set-up I'd like to stay below 1.5V vcore.
Thanks for your suggestion though.

[TouGe]

Can someone who has tried it, tell me how much colder the NB and SB will get if I use some third party cooler on them? Also will that mean both a heatsink and some noizy small fan on top of it?

In short; what are my options and how much colder will the bridges get?

JH_man

There are a few posts covering this already somewhere in this thread.

NB --> HR-05 SLI/IFX or just a regular IFX depending on your CPU HS configuration.

SB--> Same as NB unless you are running CF then you'll need something low profile like this.

I use the SLI/IFX on my NB w/ fan and I sit at 40C while running OCCT w/ NB @ 1.57V and 34-35C @ idle.
I have the Enzo I linked to on my SB w/ 1.075V and two HD 2900XT's sitting on top of it and it sits at 37C under load and 34 @ idle.

Edit:
Dont also the original heatsink thingy also cool some other components? Not just the bridge chips?..

The heatpipe assembly actually comes apart in two pieces so if you remove the NB/SB heatpipes you will still have a separate set covering the mosfets to the rear and top. You can also remove this part and use two HR-09's type 2 to cool this area which is what I did.

[Ozan Baseski]

Is it bad to insert 4 pins instead of 8 pins on the board and let it work this way? I knew some boards won't even post with 4 pins but this one does. Is it also possible that this issue is the cause for my strange ram behavior? (Working in single channel mode)

[jonny_ftm]

you need to setup an RMA and tell them you want to an Advanced Exchange and they will email you the paperwork and then you send it back to them.

I did the cross shipment with Abit and they required a deposit on credit card so just ask to make sure but it doesn't go through if they get the mobo in a reasonable time.

Thank you for the info, will be very useful

Also, Unless you are talking bout moving to DDR3 on X38 or the very few X48's the current quad core issues are likely to be the same since it has more to do with the CPU internal BUS and the way it ties in with DDR2 controller for deviders. And just to note, this mobo does FSB of 500 just as nice as 400. It was built with FSB 400 in mind just cause Intel didnt give it the stamp at realese. So if you plan to stick with DDR2, then I wouldn't stress over X48, since it is likely to be replaced by the time most of us here move to DDR3, wich for today is the only improvemens it is offering (NB RAM and devider, it is still limited to the 2 slot config for over DDR1600 for stability and with memory profiling)

I don't fully agree. Most G0 Q6600 are FSB limited around 460-480 MHz. A 445 FSB should be easily supported by the CPU only (vcore needed will depend on the CPU of course).
Currently, all X38 motherboards, including the DDR3 series (P5E3, Maximus Extreme, Gigabyte...) are very hard to stabilize a Quadcore above 400MHz FSB. If it was only a DDR3 and memory divider issue, like you advance, we could see on the P5E3 threads some different results than the mess we see here on the Maximus.

You should understand that the contarints on the NB, PWM, and FSBT are so much more important with a quadcore than a C2D.

You keep talking about a 500Mhz that the X38 is capable of. Yes, you're true, but for C2D. As I told you, still waiting to see any OCCT RAM stable OC of a quadcore CPU with a +450MHz FSB. Personally, I didn't see any even at 445MHz. Be it a DDR3 or DDR2 setup. The 1:1 multi won't change it.

So, anyone claiming a 445 MHz stable OC on his quadcore, just show us some benches, and not only some CPU priming in small FFT, as we talk about motherboard NB-RAM stabilty (OCCT RAM, a long F@H...)

The X48 will support the 400MHz FSB on quad cores, at default voltage, that is on stock, and there lies the difference with the X38

In few weeks, we shall see it. Meanwhile, take fun to stabilize this X38 above 400MHz on a quadcore (not only screens, useless for every day use, but for world records)

[JH_man]

There are a few posts covering this already somewhere in this thread.

NB --> HR-05 SLI/IFX or just a regular IFX depending on your CPU HS configuration.

SB--> Same as NB unless you are running CF then you'll need something low profile like this.

I use the SLI/IFX on my NB w/ fan and I sit at 40C while running OCCT w/ NB @ 1.57V and 34-35C @ idle.
I have the Enzo I linked to on my SB w/ 1.075V and two HD 2900XT's sitting on top of it and it sits at 37C under load and 34 @ idle.



The heatpipe assembly actually comes apart in two pieces so if you remove the NB/SB heatpipes you will still have a separate set covering the mosfets to the rear and top. You can also remove this part and use two HR-09's type 2 to cool this area which is what I did.

I use a Ultra Extreme 120 facing backwards with a fan blowing on it, so I guess I can just use a normal HR-05 on the NB?.. The CPU fan with then blow on it also, and the Antec 182 fan on the back will also pull air out just behind it (if there is room for the HR-05 between the case fan and the Ultra 120 Extreme).

I have 1 8800GT gfx card, (I am at work now so I am not sure how the spacing is), so do I need a SLI for the SB?..

JH_man

[thelawns]

I use a Ultra Extreme 120 facing backwards with a fan blowing on it, so I guess I can just use a normal HR-05 on the NB?.. The CPU fan with then blow on it also, and the Antec 182 fan on the back will also pull air out just behind it (if there is room for the HR-05 between the case fan and the Ultra 120 Extreme).

I have 1 8800GT gfx card, (I am at work now so I am not sure how the spacing is), so do I need a SLI for the SB?..

JH_man

I have my Ultra 120 horizontally, facing upwards with a 120mm fan underneath pushing up. The NB & SB HR-05 IFX are also Horizontal, waiting for fans to be fitted underneath, again pushing up. I posted pics of this a few pages back...

[eSp!s0]

If you're using the E6750, you dont need 40 strap to get 5:6. You can get 5:6 in Auto strap or 333 maunal (under v0903, manual strap under 901/902).

Yes I can, but that`s always a few mhz less. And yesterday sb. in the thread from forumdeluxx postet his settings with the same ram for 1200 at 450x8 (400strap) and they aren`t stable for my system. Then I tried moderate instead of strong and transaction booster enabled, disabled ect. but somethin is ing ty with my config

I also changed to white slots yesterday with absolutly no changes.

Also 333strap 1200mhz 500x7 is working with NB 1,29V, VFSB 1,38V and ram 2,14V and 1200mhz 450x8 400strap isnt`working with NB1,49V VFSB 1,40V and ram 2,14V or 2,20V, too. BSOD at boot or errors in memtest or OCCT after a few minutes.
Strap to auto tested without any changes. CPU tested with ram to 900mhz strap auto and the cpu runs OCCT 30min without an error with NB @ 1,35V. I don`t understand, why I need that much more voltage on my NB for same ram frequency and even slower strap!?
That`s crazy.

[DMight]

I experienced the same in both 901 & 903. I rarely turn off my PC completely. Usually S3 or S4. But I got the same issue during regular restarts for OC profile test or installing apps requiring restart. It has happend twice for me too so far. I was able to reload from my CMOS profile saves.

Well, when I woke up this morning the system lights were on, the fans were blowing, but there was no response from my PC. Mouse LED was out, no signal sent to the monitor. The LCD Poster read DET DRAM. Is this some sort of screw-up with an ACPI function? Can I simply turn ACPI Functions off for the time being?

1- I'm concerned about my ability to return to an earlier BIOS (I'm running a 64-bit OS and it seems 32-bit is required to flash back).
2- 903 has proven the most stable for me thus far and I don't want to flash back and lose my OC.

Nuckin_Futs, have you yet flashed back to 802 as you'd intended? If so, have your same voltage settings worked as with the 9xx BIOS?

[DMight]

OK, this is really disturbing. I was restoring my settings in the BIOS and the system shut itself down...while I was in the BIOS! What is going on? Currently I can't get into the BIOS, the system shut itself down when the LCD Poster read INIT ROM. Please tell me this can eventually be remedied if I return to an earlier BIOS revision. Arrrrgggg, I'm so frustrated, it seemed I'd finally gotten this system stable and now this. *sigh*

[JH_man]

I have my Ultra 120 horizontally, facing upwards with a 120mm fan underneath pushing up. The NB & SB HR-05 IFX are also Horizontal, waiting for fans to be fitted underneath, again pushing up. I posted pics of this a few pages back...

Tnx, found it. Thinking back I have obviously missunderstood where the NB chip is My 120 is fitted vertically, so not sure how that will work out, and its a good fit, so I dont want to redo it, I think.

Can you remember how many degrees (celcius) the SB and NB temps droped after fitting the HRs?.. Just wondering if its worth it... I have a Antec 182 case if it matters.

JH_man

[DMight]

Now the system is stuck again with the LCD Poster reading VGA BIOS.

At different time the LCD Poster has stopped at VGA BIOS, CHECK CPU, and once at INIT ROM. Usually hitting CLR CMOS a couple of times will get it to post to BIOS again...still trying.

[eSp!s0]

OK, I found settings, which allowed me to run MemTest 3.6 two times to test full free ram to pver 100% coverage without errors. Will test more now.
Settings are:
BIOS 903
Extreme Tweaker
Ai Overclock Tuner : Manual
OC From CPU Level Up : Auto
CPU Ratio Control : Manual
-Ratio CMOS Setting : Auto
FSB Frequency : 450
FSB Strap to North Bridge : 400
PCI-E Frequency: 100

DRAM Frequency: 1200
DRAM Command Rate : 2T
DRAM Timing Control: Manual
CAS# Latency : 5
RAS# to CAS# Delay : 5
RAS# Precharge : 5
RAS# ActivateTime : 5
RAS# to RAS# Delay : 15
Row Refresh Cycle Time : Auto
Write Recovery Time : Auto
Read to Precharge Time : Auto

Read to Write Delay (S/D) : Auto
Write to Read Delay (S) : Auto
Write to Read Delay (D) : Auto
Read to Read Delay (S) : Auto
Read to Read Delay (D) : Auto
Write to Write Delay (S) : Auto
Write to Write Delay (D) : Auto
DRAM Static Read Control: Disabled
Ai Clock Twister : Light
Transaction Booster : Enabled
-Boost Level 1

CPU Voltage : 1.34300
CPU PLL Voltage : Auto
North Bridge Voltage : 1.35
DRAM Voltage : 2.16
FSB Termination Voltage : 1.40
South Bridge Voltage : 1.050
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 : DDR2-REFF
SB 1.5V Voltage : Auto

NB LED Selection : NB Volt
SB LED Selection : SB Volt
CPU LED Selection : CPU Volt
Voltiminder LED : Enabled

CPU Spread Spectrum : Disabled
PCIE Spread Spectrum : Disabled

Hope I got the trick now

[DMight]

My system has never taken this long to reset before...and it still has not. No matter what I do the POST stops at VGA BIOS. The CLR CMOS button on the back does nothing but reset the POST, I still can't get into the BIOS. The on/off button on the motherboard itself has sometimes worked for me, now it merely resets the POST and I get the same error. The CLR CMOS button on the motherboard next to the on/off button does nothing at all. I've powered down completely and unplugged the system (what I'm currently doing), this does nothing so far.

Wow, so ASUS's brilliant OC recovery tools are brilliant so long as they work.

[kup]

If it's stopping at that same point, the problem is with the graphics card, not the motherboard.

[thelawns]

Tnx, found it. Thinking back I have obviously missunderstood where the NB chip is My 120 is fitted vertically, so not sure how that will work out, and its a good fit, so I dont want to redo it, I think.

Can you remember how many degrees (celcius) the SB and NB temps droped after fitting the HRs?.. Just wondering if its worth it... I have a Antec 182 case if it matters.

JH_man

The NB chip is just under the CPU. The SB is just to the right and lower of the Top PCI-Express slot...

Well I did report all this, even further back in the thread and I posted the same pics, so if you hunt for them you'll see.

Off the top of my head, on idle the temps dropped 3-4ºC without any fans on them. I haven't got around to properly testing them on Load yet, but thats where these bad boys come into their own. They really do keep things cooler on load. Plus I have 80mm fans waiting to be fitted. So yes, its worth it!

[Renegade5399]

This board + 0903 is just awesome on my dual core.

Using my previous settings + 2 more GB of RAM, I dropped the multi to 8 and upped the FSB to 506. Only thing I had to bump was the NB volts from 1.58 to 1.62. This is expected as there is more stress on the memory controller. The RAM is of 2 different brands, but they are the same timings and speeds. I think they are even the same chips, but I'd have to check the DDR2 chip chart to be sure. The RAM is at DDR2 1215, 5-5-5-15-2T with 2.25v. I haven't tried to tighten the timings, but I'm sure I can lower them slightly. Of course, there's always Memset. The performance of my rig is outstanding at this point and I am finally happy with my purchase. Sometime after the first of the year, I am going to try and get my hands on a new Quad and skip over the current Q6xxx series altogether.

Temps wise, I have added another Bonnie core to the loop due to plumbing in my 8800GT. This has dropped my NB a couple more degrees and at 1.62v the NB doesn't get over 34°C during OCCT testing. CPU cores are about 56°C each loaded unless small FFTs are used, then they get up around 60°C.

I have stopped using PC Probe as the latest version of Everest reads all the boards sensors correctly now. Everest gives you the actual core temps, not to mention all the info you want to see at a glance from CPUz and GPUz plus more. All in one neat little package. I highly suggest it. It has Vista sidebar support as well as the OSD (which I prefer to use).

As far as ASUS stopping the run of this board, this was to be expected. The SE was listed as a limited quantity from the get go anyways. Hopefully they give us a couple more BIOS releases before they totally call it quits on it.

[ninogui]

Why would you think that?

This is getting confusing, and close to nonsense. They should be called water pipes and thats it.... these are no heat pipes at all, and there´s no gas or ever been inside them.


voi la lololol

[thelawns]

Should I be turning anything back on, in the CPU Bios screen or just leave them all as disabled?

[ninogui]

Yeah Baby, that`s good stuff and I have this Kit too, perhaps you can help me with the 400strap after you tested them a few days

Ye sure i will!

I´ve putten 2x40mm fans on top of them just to be on the safe side when messing with rams.

I suspect that somehow these pc1200 TX1200QLJ-2GK axerams are the plain and previous pc1066 with a new cooler packing, and thats it.

With the transcend axeram 1200 presently i´m really ram stable only on my 425x8 (which will be my starting point to go up) with auto divider, that means 1133 3:4 and the timmings at 5-5-5-15, 2.08 ram volt on 0903 bios (everest reports 2.16volt / vtt1.07) ccct and prime blend stable.

the next step whatever the divider is on this setup will be 1278 for the ram, tried that with 5-5-5-15 and 5-5-5-18 with no success, in fact i stopped when reaching 2.3 ram volt on bios (everest 2.4 / vtt 1.20), XP booting ok, apps running too but occt crashed after 3 minutes top.

And thats quite deep into the yellow area bios ram volt setting. Tried with different clocktwister and auto also, tried with trans. booster either auto or set to 2, or 0, no go.

I think i should use extra care when messing with ram voltage, because of electromigration, will cause your rams not to boot at stock settings anymore and progressively shorten their life.

just take a look in

http://www.xtremesystems.org/forums/...d.php?t=140870

in these very forums.

Unfortunately theres not much web info or testings with our particular rams, just found a chinese written one at

http://babelfish.altavista.com/babel...occ2%2f66.html

already with babelfish inserted into, but from what i saw there the guy can reach 1279 at 5-5-5-15 with 2.2V (i would guess in the bios right?) which somehow i cant (can you?)

Maybe bios 0903 is not so good on memory side of equation... i still wonder if asus took it out, must be for some reason on something unstable they found out. I bet they read these forums throughly, so they could also post the reasons why... to help us out getting the best of their hardware so in fact helping them!

Tell me what you´ve reached so far and at what bios voltings, i will keep on testing for stability and will post on results.

Has someone else spotted a transcend axeram TX1200QLJ-2GK OC test.? if so please let us know.





greets

[ninogui]

anybody got some suggestions to try

Running bios 802
425x8 stap set to 400 ...stable
450x8 strap set to 333 ...stable
400x9 strap set to 400 ...stable

450x8 strap set to 400 it fails everytime, usally in the first 15 mins OCCt or Prime
I used to run this until I flashed to 902, after I flashed back to 802 my settings are still unstable.
I'm thinking it might have to do with transaction booster and clock twister
Twister is set to Strong
Booster is set to Auto
Any ideas would be great

Hi grnfinger,

have you tried with 5-5-5-16 or 5-5-5-18 or 5-6-5-16.? and a notch up on the ram voltage.?

the straps have been failing me too, at least on bios 0903.