ASUS Maximus Formula SE

[nowakjan2000]

NB chip - 44 degrees C
SB Chip - 43 degrees C
I havent tried remounting the cpu block 'cos I thing it's placed correctly with Arctic Silver 5. I was thinking more of replacing it with D-Tek Fusion.

[Durkadurka]

Nowakjan, even tho your watercooling kit (except NB/SB blocks) is poor, it shouldn't be that bad at idle.

You probably have a bad mount, try remounting the cpu-waterblock.

[nowakjan2000]

Ok, As you say, I'll give it a try and remount the cpu block and report back tomorrow. Thanks anyways guys.

[ElEctric_EyE]

came home today and tried bioshock with same settings I posted earlier from achieving 4+GHz last night, gave me BSOD, tweaking more V's...

[Grnfinger]

NB chip - 44 degrees C
SB Chip - 43 degrees C
I havent tried remounting the cpu block 'cos I thing it's placed correctly with Arctic Silver 5. I was thinking more of replacing it with D-Tek Fusion.

I'm using the Fuzion block that came with the board ( remounted with AS5 )
and my temps are

NB 39 celcius

SB 37 celcius

you might want to take a really close look at your install and make sure everything is a-o-k.

[nowakjan2000]

Well, I have to correct myself, at idle I get:
NB - 39 C
SB - 38 C
CPU - 58 C
MB - 34 C

[Grnfinger]

cpu is high for idle

I run 450x8 at 1.55v q6600 cpu.
I have a swiftech apogee gt cpu block and my cpu idles at a rather nice 35 celcius. Full load I see 58-62 celcius depending the heat in the room

[DMight]

I've tried it on Vista 64-bit and XP 32-bit SP3

In Task Manager it uses 100%, but It only reads 75% on OCCT.

Before, Like when a first got these board, It would do 100% in OCCT.

So do we think this is a problem of OCCT misreporting it's percentages, or does it appear that OCCT is not really using 100%. If Task Manager is reporting 100%, is that a more accurate reading?

[nowakjan2000]

The decision has been made - I'm getting D-Tek Fuzion CPU Block. As soon as it arrives I'll reassamble the whole loop, double check it and than get back to you guys with the results.

[ElEctric_EyE]

had to up SBv to 1.175. system froze after doing a save to HDD was a clue... I know all my volts are high and electromigration is/may become a factor. I would never run this system 24/7. I consider this box a dedicated computer for games/video encoding and is only on for those purposes for no longer than 2hrs at a time. I'm gonna encode and burn an episode of six million dollar man in <1/2hr.. w00t!

[Grnfinger]

The decision has been made - I'm getting D-Tek Fuzion CPU Block. As soon as it arrives I'll reassamble the whole loop, double check it and than get back to you guys with the results.

I think that is a wise choice

you might want to check out the lph rate on that Thermaltake pump
if it's rather low altho I think it's around 500 lph you might want to think about replacing that aswell

[nowakjan2000]

It' 500L/h acording to the specs.

[safan80]

Yea, I think so, Also I didn't crash as quickly as the Beta.

So 1.1.1 is good, I'm about to try 1.1.1b. But I think the 1.2.0 Beta is worthless to us right now. Maybe the official release will be better.

1.2 beta works fine here.

[Grnfinger]

It' 500L/h accordingto the specs.

pump should be ok

[Warboy]

1.2 beta works fine here.

do you get 99.00%-100% when stressing your CPU?

[kup]

The ram voltage dropping is happening to me, just not as frequently...

I've been getting random reboots without any idea of a cause, can OCCT/Prime for 10hrs+ but will crash @ idle?! So I upped the vDimm to 2.26v bios (2.37v real) and not a single hiccup so far. Now I know this ram is good for the clocks with a much lower voltage but it seems when it dips too low it crashes/reboots, but with the vDimm set higher, even with the sudden and momentary droop, it's not falling low enough to crash!

System is once again stable!

[Eastcoasthandle]

Benefit? YES!!!

I popped mine second board as soon as I got it..came off easily...first board would run 50c or so under 1.45v, now I can run 1.8v, and barely top 41c, using ceramique.

It actually seems now that the southbridge connection to teh NB heatsink dictates temps...but also because I am using thermal probes attached to various places I am aware that both SB and NB temps are not reported accurately in ANY instance....both heatsinks are 2c higher by thermal probe than reported temps by on-board sensors.

the heatsink assembly , after changeing paste, seem quite good...it's the stock TIM that fouls things up!

I left the MB in the frig for about 3 hours and tried again. Just as you said, the NB and SB HS popped right off with no real force.

[ElEctric_EyE]

Jeez, I tried encoding an avi to dvd compatible mpg using TMPGEnc and it's using close to 90% CPU....guessed settings, this is first time stable, not tweaking yet. This will be my new real world stress test.

[Grnfinger]

any ideas as to whats going on here?

[kup]

This is exactly what WarBoy was saying, the massive ram flucuations are causing instability and more than likely responsible for killing all the ram...

EDIT: And from what I can see, yours peaked @ ~12-13v according that that graph before it errored!!!

[Eastcoasthandle]

One word comes to mind, ripple. I suggest using a different wall socket from the other side of the room (not the socket next to the one you are currently using). See if that helps, also make sure you are using an 8-pin 12V connector for your MB as well.

[iadstudio]

Anybody doing an auto-shutdown for their pump with this mobo? I know I could hook up the 3 pin rpm sensor to a mobo header and have speedfan shut my pump down, but I wondered what kind of rmp sensing and make even shut down features this mobo has.

[Grnfinger]

thats what I thought
not looking to buy new ram anytime soon, guess I'll dial it back to safe 3.4ghz
I can be happy with that for now

[Grnfinger]

One word comes to mind, ripple. I suggest using a different wall socket from the other side of the room (not the socket next to the one you are currently using). See if that helps, also make sure you are using an 8-pin 12V connector as well.

I have the 8 pin connected

I will try a different outlet, what about power bars? can they be a factor?
The reason I ask is I'm using a good one but its rather old, maybe 6 years now

[kup]

One word comes to mind, ripple. I suggest using a different wall socket from the other side of the room (not the socket next to the one you are currently using). See if that helps, also make sure you are using an 8-pin 12V connector for your MB as well.

My average ripple for my vDimm is only 2.16%, which barely moves on the graph for 2.3v. But I'm in a really new building with good wiring and top notch electrics!

Are there any mods to use a Molex line as the 2nd 12v plug if your PSU only has a 4pin 12v?