Testing / comparing : Intel D975XBX2 / Asus P5B DX ***56K WARNING***

[STaRGaZeR]

Just pointing out: minsc_tdp had
1-what appears was a watchdog stop
2-backleveled BIOS to pre watchdog
3- still had watchdog 'like' stop
4- went forward to latest BIOS, turned off watchdog and it works...

inferring that once you go to a BIOS that is watchdog, there's no going back....

Are you saying that if I go back from a BIOS that have Watchdog to the 2333 por example, Watchdog still will be there? My board cames two days ago with the 2333 and I want the fuking HDD led turned off. BTW nice and rock stable board

[slappynutz]

I never had this issue with an Asus board, now I am wondering if this BadAxe2 was my best choice.

After a decade of Abit, Asus and even Biostar boards, I'd come to accept that capacitor whine and cheap components were unavoidable in motherboards. This is the first board with absolutely no annoying electronic squeal ... that alone trumps any special features Asus brings to the table. I've gone through probably 6 Asus boards for excessive squeal, and I'm hardly a power user.

[SiGfever]

The temp issue has nothing to do with the board. It's solely a processor problem in that Intel provides a way to read the core temp sensors but not a way to read the tjunction temp on which they're based. You need both to determine the absolute temperature.

So the freezing up issue is a miss-read of the temps?

[minsc_tdp]

Just pointing out: minsc_tdp had
1-what appears was a watchdog stop
2-backleveled BIOS to pre watchdog
3- still had watchdog 'like' stop
4- went forward to latest BIOS, turned off watchdog and it works...

inferring that once you go to a BIOS that is watchdog, there's no going back....

Possibly, but if that really was the case, the remedy was to go back to the latest BIOS, load optimized defaults, set general stuff, go to maint mode and disable watchdog/c1e, shutdown, back to regular bios screen, change voltages, save and reboot (I wait for menu) and then shut down to properly set voltages, boot again, set memory settings, one at a time, reboot, confirm memory settings, (jeez!) and then start OC'ing the FSB.

[minsc_tdp]

I believe the e4300 has a 100c tjunction. There are a few threads in the info and support sub-forum that talk about this. You might take a look.

Good info, thanks. Sorry haven't read up, I'm somewhat new to Conroe and have spent a week of my life digesting this thread.

Threads on XS are like women - there's only time for a few of them but you try to do as many as you can

[gtj]

So the freezing up issue is a miss-read of the temps?

If you're referring to the freezing using older version of Everest then yeah, the freezing up was caused by Everest probing an address on the SMBus that it shouldn't be probing and it causes the voltage regulator to immediately drop the vcore back to default. If you needed the extra vcore to be stable then the result was an immediate hang.

[SiGfever]

If you're referring to the freezing using older version of Everest then yeah, the freezing up was caused by Everest probing an address on the SMBus that it shouldn't be probing and it causes the voltage regulator to immediately drop the vcore back to default. If you needed the extra vcore to be stable then the result was an immediate hang.

No, I am talking about when the system comes up to the login screen in XP pro it freezes the mouse and keyboard from a cold boot causing me to have to do a warm boot and then it will let me login.

[minsc_tdp]

I am having an issue with my new BadAxe2 board. When I shut down the system sometimes when I restart the computer the window will appear requesting you to "CTRL-ALT-DEL" to sign in and then the cursor and the keyboard freeze making me have to reboot. When it comes back up I can sign in fine, it only appears to happen from a cold boot.

To determine if the system is really freezing up here or maybe it's a mouse/kb/usb thing, do start > control userpasswords2 and uncheck "must enter a login name" and supply credentials. That'll auto login and dump you on the desktop, then you can have startup apps pop up to give system info to see if it's all kosher.

If it really is freezing on that screen the autologin won't help. At that step it could be any driver on your system that loads later in the boot process, could be any connected peripheral acting up, like sound.

Assuming you can reproduce this reliably from a cold boot, see if this helps: Cold boot but go into the bios. Sit on the hardware monitoring page for 2 minutes. Then exit without saving and it should continue booting (I think it doesn't even reboot in this case.) If you get into Windows reliably like that, it could be something needs to warm up who knows, but try it.

When I do this, I usually don't get the thermal alarm siren that happens just after post on many cold boots.

[minsc_tdp]

quick question, my core TAT temps are always a few C apart. As much as 3 to 4C difference, the second core being hotter. Is this normal?

I think my AC5 job was slighly lacking on the bottom half of the chip. Which half is each core on, anyone know? I'm willing to bet it's core #2

One more question, I notice in RMClock that under Automatic Thermal Protection: TM1, TM2, Sync TM1, and Extended Throttling are all disabled. Is this OK? Will it affect overclocking at all? Did these get disabled due to something I did in the BIOS, like disabling EIST or Fan Control or similar? I manually turned them on and then forced both Thermal and Normal Throttling to 50% to see if it would crash while under Small FFT load and it throttled OK, so I think I'll turn it on to protect my chip.

[gtj]

quick question, my core TAT temps are always a few C apart. As much as 3 to 4C difference, the second core being hotter. Is this normal?

I think my AC5 job was slighly lacking on the bottom half of the chip. Which half is each core on, anyone know? I'm willing to bet it's core #2

One more question, I notice in RMClock that under Automatic Thermal Protection: TM1, TM2, Sync TM1, and Extended Throttling are all disabled. Is this OK? Will it affect overclocking at all? Did these get disabled due to something I did in the BIOS, like disabling EIST or Fan Control or similar? I manually turned them on and then forced both Thermal and Normal Throttling to 50% to see if it would crash while under Small FFT load and it throttled OK, so I think I'll turn it on to protect my chip.

Perfectly normal for 1 core to be a little hotter than the other at idle because of how work is scheduled. If you load them both, are they still a few degrees apart? They shouldn't be. The actual cores are separated by no more than 5mm and they are drectly in the center of the chip...

E6700 with the heat spreader removed.


All of the TM stuff is disabled by default. Turning it on is a good idea.

[stmckin]

Are you saying that if I go back from a BIOS that have Watchdog to the 2333 por example, Watchdog still will be there? My board cames two days ago with the 2333 and I want the fuking HDD led turned off. BTW nice and rock stable board

I'm not familiar with how the watchdog might have been implemented, but minsc_tdp's experience suggests that once the option has been enabled by a late BIOS, going to a pre watchdog BIOS just takes away the ability to toggle it.... may be true, may not be...

For me it's not an issue anyway, circumventing the watchdog on late BIOS works, I'm able to get where I want to by going to 3.6 aircooled on the E6600 and have it rock solid, so.....

minsc_tdp's issues overall are a little mystifying though... on this board with an E6600, anybody should be able to
- flash newest bios and load defaults
- maintenance mode-> watchdog->off
- fsb->375
- enhanced power slope -> on
- vCore 1.475
- mem -> 533-5-5-5-15-2.1
- boot at 3.3 rock solid

even though the Intel Desktop board quality is so high (I've never had a dead board in 6 years), maybe his is 'funny'? I dunno

[minsc_tdp]

I doubt that having watchdog On in a newer bios, then flashing to 2333 leaves watchdog on. I would imagine that's a software feature of the bios and since it simply doesn't exist in 2333 it shouldn't function and the previous settings aren't carried over.

HOWEVER, maybe it was in 2333 all along, disabled and hidden, and restoring the 2333 bios did in fact turn it on by bringing the setting over from the newer bios. Feasible I guess. At least there's a known workaround now for a problem that may not even be real.

New issue:

The CPU fan header suddenly started overriding my Silverstone FM121 fan controller. It was comfortable at 1600 RPM but now the onboard connector pumps it back and forth betwen 1600 and 2000. Turning the Silverstone all the way down has no effect, which is a shock given that it's right on the power line and the mobo header is just a single yellow line.

Temps are good (35 C idle @ 2 GHz C1E) and <56C under full load (3 GHz, TAT)

I think I might have connected that CPU header wrong - in the manual it shows 3 pin connectors on pins 1-2-3 and a 4 pin connector on 1-2-3-4 where the 4th pin is yellow and on PIN 4. Should my single yellow pin connector be put on Pin 4? Right now it's on Pin 3.

Even better, can I unplug it entirely or will the board complain? I thought I saw somewhere that you MUST have a CPU fan on the right header or "bad things happen" but I don't remember where I saw that, and can't see how as long as I monitor temps. I don't need the RPM monitoring (my ears work fine.)

Update: Now, lowering the SilverStone controller has no effect, even if I set the Proc Fan to Manual in IDCC and set it all the way min or max. Something somewhere thinks it's smarter than both of these manual fan controls and is maxing out the volts to the fan. WTH?

[gtj]

New issue:

The CPU fan header suddenly started overriding my Silverstone FM121 fan controller. It was comfortable at 1600 RPM but now the onboard connector pumps it back and forth betwen 1600 and 2000. Turning the Silverstone all the way down has no effect, which is a shock given that it's right on the power line and the mobo header is just a single yellow line.

Temps are good (35 C idle @ 2 GHz C1E) and <56C under full load (3 GHz, TAT)

I think I might have connected that CPU header wrong - in the manual it shows 3 pin connectors on pins 1-2-3 and a 4 pin connector on 1-2-3-4 where the 4th pin is yellow and on PIN 4. Should my single yellow pin connector be put on Pin 4? Right now it's on Pin 3.

Even better, can I unplug it entirely or will the board complain? I thought I saw somewhere that you MUST have a CPU fan on the right header or "bad things happen" but I don't remember where I saw that, and can't see how as long as I monitor temps. I don't need the RPM monitoring (my ears work fine.)

Update: Now, lowering the SilverStone controller has no effect, even if I set the Proc Fan to Manual in IDCC and set it all the way min or max. Something somewhere thinks it's smarter than both of these manual fan controls and is maxing out the volts to the fan. WTH?

If all you're connecting is the 1 yellow wire to pin 3 (3rd from the right) on the MB, then the MB is not the problem. That's just the tachometer signal to the sensor chip. There's no way for the MB to control that fan unless you're actually getting power from that or one of the other fan headers.

[minsc_tdp]

If all you're connecting is the 1 yellow wire to pin 3 (3rd from the right) on the MB, then the MB is not the problem. That's just the tachometer signal to the sensor chip. There's no way for the MB to control that fan unless you're actually getting power from that or one of the other fan headers.

You know, that's what I thought too. But, this will blow your mind:

Before it got stuck at max RPM, I SWEAR, using IDCC and setting CPU control to Manual and sliding it back would cut the RPMs.

I was using both that and my Silverstone controller to get it where I wanted. But IDCC's fan settings don't stick well so I hadn't looked at it in a while.

Nonetheless, the board is sending something to my fan on pin 3 of that header, through that yellow cable, and it is affecting my fan.

From the technical reference:

CAUTION
The processor fan must be connected to the processor fan header, not to a chassis fan
header. Connecting the processor fan to a chassis fan header may result in onboard
component damage that will halt fan operation.

Fan Connector Maximum Available Current
Processor fan 3.0 A
Front chassis fan 1.5 A
Rear chassis fan 1.5 A
Auxiliary rear fan 3.0 A
MCH fan (optional) 1.5 A

Table 20. Processor Fan and Auxiliary Rear Fan Header
Pin Signal Name
1 Ground
2 +12 V
3 FAN_TACH
4 FAN_CONTROL

How is this possible? To be clear, the FM121 has one connector with the single yellow on the mobo cpu header pin 3, and a 2 pin connector going to the controller knob, and the power connector is yet another 4-pin standard molex which I plugged into the extra one coming off the aux pci express power connector.

[gtj]

4 pin fans aren't backwards compatable with 3 pin headers but 3 pin fans ARE backwards compatable with 4 pin headers.

The motherboard doesn't care if there's something plugged into the processor fan header or not. I just unplugged mine with the system running with no ill effect.

Try it.

EDIT: Wait, are you saying that the fan is getting its power from the 4 pin molex connector on the motherboard?

[minsc_tdp]

4 pin fans aren't backwards compatable with 3 pin headers but 3 pin fans ARE backwards compatable with 4 pin headers.

The motherboard doesn't care if there's something plugged into the processor fan header or not. I just unplugged mine with the system running with no ill effect.

Try it.

EDIT: Wait, are you saying that the fan is getting its power from the 4 pin molex connector on the motherboard?

I believe you. After coming to my senses I realize the fan or controller must be broken.

And no the fan power is on it's own big 4-pin molex straight into the PSU. Only the yellow line (on a 3 pin connector with 2 pins nothing) is on the mobo.

Yeah, I just plugged the FM121 cpu fan into a different molex up near the DVD drive and the controller worked again momentarily, dropping to 1600, but seconds later jumped back to 2000. So I'll just have to leave the controller at max until I can try another controller or replace the fan.

BTW GTJ, you are a true hero for continuing to support the newbs like me coming late to the party long after getting your own board stable. Your legend will never die.

[neonflx]

i am having an issue, i was running my E6600 at 3.7 no problem, i updated to the latest bios from 2333 and the computer will give me BSOD's after reboot , note that the only thing i changed was the bios, i changed back to 2333 but i still getting the BSOD's, now how can that be if the computer was running fine all this time, and i reverted back to 2333 but it seems that whatever the new bios changed reverting back to 2333 is not fixing it so now the max i can run is 3.6

voltage is the same my cpu likes volts so i have it max at 1.60 (measured actual is 1.55)

thanks for any suggestions

[gtj]

i am having an issue, i was running my E6600 at 3.7 no problem, i updated to the latest bios from 2333 and the computer will give me BSOD's after reboot , note that the only thing i changed was the bios, i changed back to 2333 but i still getting the BSOD's, now how can that be if the computer was running fine all this time, and i reverted back to 2333 but it seems that whatever the new bios changed reverting back to 2333 is not fixing it so now the max i can run is 3.6

voltage is the same my cpu likes volts so i have it max at 1.60 (measured actual is 1.55)

thanks for any suggestions

Did any of your BIOS settings get reset during the update? Check your SATA controller mode setting I.E. IDE, AHCI, RAID.

[shimmishim]

Not that this is amazing news but I got SLi to work on this board.

Did a simple test of sli config vs. non-sli config with a 7900gt at stock.
clock speed was 3.6 ghz (not 3.7) on an untweaked windows xp install.

3dmark05

[minsc_tdp]

i am having an issue, i was running my E6600 at 3.7 no problem, i updated to the latest bios from 2333 and the computer will give me BSOD's after reboot , note that the only thing i changed was the bios, i changed back to 2333 but i still getting the BSOD's, now how can that be if the computer was running fine all this time, and i reverted back to 2333 but it seems that whatever the new bios changed reverting back to 2333 is not fixing it so now the max i can run is 3.6

voltage is the same my cpu likes volts so i have it max at 1.60 (measured actual is 1.55)

Try Load Optimized Defaults and start over with all the settings. Especially watch the RAM timings and voltage as it seems to reset to different values between boots. The Load Optimized Defaults seemed to help with some similar problems I had.

[azzido]

Last week I thought my board had killed one memory stick (sitting in the second blue slot ) I tried all kinds of different things (higher voltage, lower timings) but it would not boot, the other stick was working just fine. So I tried it on my friends DFI infinity 975 and the "dead" stick posted just fine. Came back home popped it back in and it was working, ran orthos for 10 hours at stock speed and everything was fine.
Yesterday I shut down my computer; this morning it refused to boot. I wasn't getting usual 3 beeps (no memory error) but the board was not posting either. It seemed like it was restarting itself every 2-3 seconds. Took one memory stick out now it's posting again.

Did anyone experienced anything like this before?

Thanks for your input

[minsc_tdp]

So now that I'm at 3 GHz and about 48C under Small FFT load, I'm trying to decide where to stop.

I'm at 1.45V core, 1.625 MCH, 1.325 FSB.

What temps under load should be my cutoff if I want this chip to last at least two years or so?

[shogo_ca]

Great! Do you think you loose any performance at all because youre using Hacked SLI drivers?

Id like to see R600s xfired on this board VS 8800GTX SLIed with hacked drivers eheh

Not that this is amazing news but I got SLi to work on this board.

Did a simple test of sli config vs. non-sli config with a 7900gt at stock.
clock speed was 3.6 ghz (not 3.7) on an untweaked windows xp install.

3dmark05

[heikis]

how far can you guys push you FSB on settings:
* 266 ref.freq.
* ~1,7V MCH
* 4:5 divider

I did 385FSB this morning and didnt try higher (however I know 400 cant be done):

[screwtech02]

Great! Do you think you loose any performance at all because youre using Hacked SLI drivers?

Id like to see R600s xfired on this board VS 8800GTX SLIed with hacked drivers eheh

Personally, i'd just like to see the ever touted "R600's" period....

Maybe next year.....