ASUS P5W DH - Problems + Fixes Thread

[lawrywild]

FYI:
- whether P5WDH BIOS can override processor VID values is another story, assuming all manual
- tough to defeat/exceed PROCHOT# and Tj and override processor shutdown limit with BIOS, Vmods, in any case
- and QX6700 VID is higher than Q6600 B3 and appears to be even lower with Q6600 G0

From Intel Q6000 Series Datasheet, page 18:

3. Each processor is programmed with a maximum valid voltage identification value (VID), which is set at
manufacturing and can not be altered. Individual maximum VID values are calibrated during manufacturing such
that two processors at the same frequency may have different settings within the VID range. Note this differs
from the VID employed by the processor during a power management event (Thermal Monitor 2, Enhanced Intel
SpeedStep® Technology, or Extended HALT State).

REF LINK:
Intel® Core™2 Extreme Quad-Core Processor QX6000Δ Sequence and Intel® Core™2 Quad Processor Q6000Δ Sequence Datasheet
http://www.intel.com/design/processo...hts/315592.htm

Not sure what your point was, but P5WDH can not alter VID values and we are talking about AUTO setting in bios (EIST enabled).

I don't get what you mean by qx6700 vid is higher than q6600 b3. VIDs are nothing to do with steppings. VIDs are assigned individually to each cpu depending on it's stability at what voltage at stock clocks. I.e. Not all q6600 b3 have the same vid, or qx6700 or whatever cpu.

[lawrywild]

Tried but luck.



Tried but no luck.



I think this is where the problem lyes.... Mine is an earlier version 1.85v Rev1.02g. Dammit I don't feel getting another mobo right now.

Hmm, I have an even earlier 1.65v revision. No G0 yet but hopefully in a few weeks.

(I will be using Auto vcore anyway. EIST FTW! Vcore mod FTW! )

[Chlikaflok]

Hey everyone,

I haven't kept up with the 200+ pages of stuff there is here, but I've searched the internet High and low for a fix, and it seems I'm not the only one with the problem, but I have been having pretty heavy problems with the whole LAN portion of this board, from both the 2 Marvell Gb NICs and the Realtek Wireless Adapter. I have sudden and short disconnects at random intervals, be it in games, applications, or simply browsing the web. I have tried many different driver versions for both the WLAN NIC and the 2 Wired NICs, even though evidence points to something other than drivers since the problem applies to both network components. My own personal (uneducated, even though I'm an IT graduate, I'm not really into hardware) guess is that the Chipset falls out of sync, but that's most probably a wrong guess.

Here's hoping someone knows something about this and would like to share

Specs:
Core 2 Duo E6600
P5W DH Deluxe (obviously) with the most recent drivers from Intel for the 975X chipset
2x1024Mb Corsair RAM (I can't remember the whole product number and stuff, and unless it's really useful I can't be bothered to dig up the exact name)
ASUS GeForce 8800 GTS (up to date drivers, forceware from week of July 22th)
Running Windows XP SP2 somewhat up to date (mid-june)
Stock Fans (but no temp problems whatsoever)
Everything is at stock speeds/memory rates/etc. No OC at all, and no intention to do so.

Thanks for reading and don't hesitate to ask for more info, I'm soooo eager to rid myself and most probably a few of us of this problem.

Cheers,
Chlikaflok

[djstudio]

Max OC E6600
ASUS P5W DH Deluxe


CPU Watercooled
Hyperpatch 3 - Disabled;Memory Voltage 2.20V;Chipset 1.55V;PCI Clock 33.33MHz

[Chrysalis]

does anyone have answer as to why speedstep is causing sound and video stuttering problems for me? this also occurs at stock speeds so isnt a result of too low voltage (I think).

Is it possible that whilst the cpu is jumping between clock speeds it causes a stutter.

[MacClipper]

no, EIST shouldn't do that. Check the DMA status on your host drive instead.

[nickfd]

Hi all.. So last night I took out my e6600, cleaned off the AS5, and installed the g0 q6700 that arrived. Prior to doing this, I had installed lawry's g0 200(4/6?) modified bios on the mb, and was content with it working OK.

So when I booted the box, I went into the bios and started looking at the hardware monitor page. I think it was like 1.21 or 1.23vcore by default, and temperature was 24-25c at idle on my water. I started by mildly overclocking, and got it to post at 333mhz*10 at 1.45vcore. I didn't go into the OS at all, just sticking in the bios.

Then I upped the vcore and went up to 366x10. Still posted fine. Then did 400x10, and did 1.50vcore. Black screen, no post. After several hours of clearing the CMOS many, many times, and doing so without the battery in the mb, I finally got my bios back. I decided to go back to 333x10 and see if my os worked, testing the OS for the first time.

Bad news, nope. It would start booting, give the windows xp splash screen, then bluescreen immediately, and reboot. I couldn't tell what the msg was, since it rebooted so fast. Safe mode did the same thing. Then I tried Vista Ultimate 64bit, which I had on the same hd in a dualboot scenario. Same thing.. It would die shortly after splashscreen.

I was thinking worst case scenario.. something happened in the hd, and it corrupted the OS. K, so I formatted (reluctantly, since there was some stuff I didn't have backed up), and went to install XP SP2. It would finish installing the console/text portion of the OS, and when you reboot to have it go into the GUI, it would reboot constantly. Broke.

Testing the memory with memtest86 showed that there was no errors after an hour of crunching, back at normal 2.66ghz, with auto vcore. I tried upping the voltage to the cpu and it still made no difference. I also upgraded the mb bios from lawry custom to 2206, which had no impact or change on functionality, or lack thereof.

So I'm kinda stumped. I'm going to take out the q6700 tonite and pop my e6600 back in to see if it works. I don't know if the problem is the mb or the cpu may be a lemon, or got messed up while I was testing for higher oc? Either way, i'm frustrated.

I'm happy that other people posted pics with their g0's working on our mb, so at least I know mine SHOULD work. But I don't understand why it's not.

[GFORCE100]

Hi all.. So last night I took out my e6600, cleaned off the AS5, and installed the g0 q6700 that arrived. Prior to doing this, I had installed lawry's g0 200(4/6?) modified bios on the mb, and was content with it working OK.

So when I booted the box, I went into the bios and started looking at the hardware monitor page. I think it was like 1.21 or 1.23vcore by default, and temperature was 24-25c at idle on my water. I started by mildly overclocking, and got it to post at 333mhz*10 at 1.45vcore. I didn't go into the OS at all, just sticking in the bios.

Then I upped the vcore and went up to 366x10. Still posted fine. Then did 400x10, and did 1.50vcore. Black screen, no post. After several hours of clearing the CMOS many, many times, and doing so without the battery in the mb, I finally got my bios back. I decided to go back to 333x10 and see if my os worked, testing the OS for the first time.

Bad news, nope. It would start booting, give the windows xp splash screen, then bluescreen immediately, and reboot. I couldn't tell what the msg was, since it rebooted so fast. Safe mode did the same thing. Then I tried Vista Ultimate 64bit, which I had on the same hd in a dualboot scenario. Same thing.. It would die shortly after splashscreen.

I was thinking worst case scenario.. something happened in the hd, and it corrupted the OS. K, so I formatted (reluctantly, since there was some stuff I didn't have backed up), and went to install XP SP2. It would finish installing the console/text portion of the OS, and when you reboot to have it go into the GUI, it would reboot constantly. Broke.

Testing the memory with memtest86 showed that there was no errors after an hour of crunching, back at normal 2.66ghz, with auto vcore. I tried upping the voltage to the cpu and it still made no difference. I also upgraded the mb bios from lawry custom to 2206, which had no impact or change on functionality, or lack thereof.

So I'm kinda stumped. I'm going to take out the q6700 tonite and pop my e6600 back in to see if it works. I don't know if the problem is the mb or the cpu may be a lemon, or got messed up while I was testing for higher oc? Either way, i'm frustrated.

I'm happy that other people posted pics with their g0's working on our mb, so at least I know mine SHOULD work. But I don't understand why it's not.

Yeah I had the scenario you're having with my QX6700 at 10x333. It would work at 11x300 but not 10x333. The same situation with Windows XP not loading and rebooting at one point during start-up (though I'm sure the latter is because I had the restart after BSOD option enabled in device manager).

Finally I figured out one had to disable EIST and C1E for it to work, this was with the 1506 BIOS. Since then I just disable both regardless what BIOS I'm using when overclocking the QX6700.

[nicepun]

So am I the only one with a P5W DH and a Q6600 G0?? Anyone else having problem with vcore value staying "Auto" no matter what BIOS version and settings you use? Thanks, it looks like I'll have to go with a P5K Premium.

[nickfd]

Yeah I had the scenario you're having with my QX6700 at 10x333. It would work at 11x300 but not 10x333. The same situation with Windows XP not loading and rebooting at one point during start-up (though I'm sure the latter is because I had the restart after BSOD option enabled in device manager).

Finally I figured out one had to disable EIST and C1E for it to work, this was with the 1506 BIOS. Since then I just disable both regardless what BIOS I'm using when overclocking the QX6700.

Well, I can't get into any OS to disable the autoreboot, and now after format and start of xp install, the gui part of the install doesn't work.

My CPU settings are:

Microcode Update : Enabled
Max CPUID Value Limit : Disabled
Execute Disable Function : Disabled
CPU Internal Thermal Control : Disabled
Virtualization Technology : Enabled
Intel SpeedStep Technology : Disabled

^- all the same stuff I had on my e6600 that worked an hour before I took it out and popped in the q6700.

And also remember that even when not overclocking, I can't get an OS either. So like I said, I think it must be the CPU that's messed up, or my frankenstien p5w dh deluxe that finally threw in the towel. If that's the case, Im gonna get a p35. I'll update after I put the e6600 back in tonight after work.

[RangerXLT8]

So the question still stands can* the P5W DH work with Intel 45nm?
I know the badaxe 2 does,,,,,,,,

[MarcusTaz]

So am I the only one with a P5W DH and a Q6600 Q0?? Anyone else having problem with vcore value staying "Auto" no matter what BIOS version and settings you use? Thanks, it looks like I'll have to go with a P5K Premium.

I assume you mean G0? I am waiting on www.frys.com to sell out their old stock before they can ship me a G0, then I will be able to let you know if I experiance the same issue. Did you try Larry's modded bios?

[lawrywild]

Hey everyone,

I haven't kept up with the 200+ pages of stuff there is here, but I've searched the internet High and low for a fix, and it seems I'm not the only one with the problem, but I have been having pretty heavy problems with the whole LAN portion of this board, from both the 2 Marvell Gb NICs and the Realtek Wireless Adapter. I have sudden and short disconnects at random intervals, be it in games, applications, or simply browsing the web. I have tried many different driver versions for both the WLAN NIC and the 2 Wired NICs, even though evidence points to something other than drivers since the problem applies to both network components. My own personal (uneducated, even though I'm an IT graduate, I'm not really into hardware) guess is that the Chipset falls out of sync, but that's most probably a wrong guess.

Here's hoping someone knows something about this and would like to share

Specs:
Core 2 Duo E6600
P5W DH Deluxe (obviously) with the most recent drivers from Intel for the 975X chipset
2x1024Mb Corsair RAM (I can't remember the whole product number and stuff, and unless it's really useful I can't be bothered to dig up the exact name)
ASUS GeForce 8800 GTS (up to date drivers, forceware from week of July 22th)
Running Windows XP SP2 somewhat up to date (mid-june)
Stock Fans (but no temp problems whatsoever)
Everything is at stock speeds/memory rates/etc. No OC at all, and no intention to do so.

Thanks for reading and don't hesitate to ask for more info, I'm soooo eager to rid myself and most probably a few of us of this problem.

Cheers,
Chlikaflok

I know what you mean. I had that problem with the wireless. I couldn't fix it so I just bought an abit airpace pci-e x1 card and it's perfect. (also bought a huge aerial which gets 100% signal constantly)

[MacClipper]

So am I the only one with a P5W DH and a Q6600 G0?? Anyone else having problem with vcore value staying "Auto" no matter what BIOS version and settings you use? Thanks, it looks like I'll have to go with a P5K Premium.

No problem on my rig, I can change the Vcore once C1E/EIST is disabled - tried with both 2205 and 2206. However, I am using Auto cos I love C1E/EIST power/heat reduction. Have you tried clearing the CMOS by removing the battery for a few mins.? Have personally seen weird disappearing options with corrupted CMOS settings.

the gui part of the install doesn't work

@nickfd - have u tried the FAQ advice to disable all the unnecessary onboard devices eg. 1394, audio etc.. while installing yet?

[nickfd]

@nickfd - have u tried the FAQ advice to disable all the unnecessary onboard devices eg. 1394, audio etc.. while installing yet?

Yep. Ive been a user on this thread since page 145 or so, when I bought it and got my e6600. Audio, 1394, and bottom nics are disabled. jmicron is disabled as well. All the basic debugging is done, and all that's left is swapping the q6700 with the e6600 to cross out motherboard or cpu.

What I don't understand is why memtest86 works fine when I boot it, and also the text based installer, but not when the graphical part of the OS wants to start.

Hmm i really hope my 8800gtx isnt going.

[nicepun]

No problem on my rig, I can change the Vcore once C1E/EIST is disabled - tried with both 2205 and 2206. However, I am using Auto cos I love C1E/EIST power/heat reduction. Have you tried clearing the CMOS by removing the battery for a few mins.? Have personally seen weird disappearing options with corrupted CMOS settings.

What rev is your mobo? I cleared the CMOS but not by removing the battery, I used the jumper. I will give it a try by removing the battery.

Additional Comments: NO GO!! I'm still getting the same "Auto" value regardless. I haven't tried 2205 but if 2206 doesn't do it I doubt 2205 will make a difference.

[lawrywild]

nicepun, the manual has the correct way to completely clear cmos.

[nicepun]

nicepun, the manual has the correct way to completely clear cmos.

No go my friend.

[GFORCE100]

nickfd

So does the P5W DH work with your E6600 or is Windows XP doing the same thing?

Sounds very odd, as if some capacitor was broken off.

[icrunk]

I've got a P5W-DH, but my IDE pioneer burner gets detected on boot up, but then after some time it just disappears from the list in My Computer.

What could be the problem?


Also a few months back, i attempted to boot up my computer and i heard something pop and smoke was coming out of a chip. It is circled in red in the picture.



The things i have discovered from the pop, is that the 2 fan connectors next to the chip don't work...

What is that chip supposed to do?

[MarcusTaz]

IC Regulator or power output transistor. Call Asus and setup an RMA, if you want a crossship so you do not have any downtime you can arange that with them if the mobo is a current model... Always worked for me...

[lawrywild]

That's a mosfet; http://en.wikipedia.org/wiki/Power_MOSFET

[nickfd]

Well, I took off the cpu, and was shocked to see what happened. It looks like there are ~25 charred pads on the CPU, and the corresponding pads on the MB are charred as well.

I was able to clean it up a little bit, and some of the substance came off as well when I scraped at it with a knife. However, it leaves little marks where I scrape, so I don't want to do that.

I'm figuring when I was cleaning the board, some as5 cleaner made its way between the chip + mb, and oxidized the connections when the board supplied it with power. I don't know for sure, cause I've never seen anything like this before.

I took the chip slightly better than what you see in the pics, and put it on the board and put my waterblock on the cpu. I was able to actually see the board+cpu POST, but due to my absolutely WONDERFUL luck, the pump on my water reservoir decided to die on me.

Being too afraid to see if it worked, cause I went into the HW Monitor and saw the temperature just rising without stopping, I shut it down. As you guys with watercooling solutions know, the cpu block is very small, and isn't designed to run very well without any kind of flow.

So I guess my plan now is to toss the motherboard, get a p35 (any suggestions?), and go air cooling for a little while. I don't have the extra cash to spare now to get a whole new water solution, which i desperately need, cause the pump going is just the latest of problems that have been plaguing my crap boxset water solution.

If the CPU still causes me problems when I get the new mb, I'll cross my fingers and HOPE that intel takes the chip in for warranty. I just don't know how to even approach them about what I see on the cpu without having them void my warranty for misuse or something.

Sigh. Stuff like this always happens at the worst times. I leave on a cross continental vacation in less than a week, and I really need this time to research plans, options, and things to do. GRRRRRR. </rant>

pics:

^- look at the bottom pins, which lined up to the left pins on the image below (mb)


^- look at the left part of the board.

[mstp2009]

Judging by the INSANE amount of thermal grease I see on the underside of the socket mounting clip, I think we know what your problem is.

You put too much thermal paste on there, and when the CPU was clamped down it oozed into the socket.

This is user error, and both Intel and ASUS are going to say sorry, no warranty.

[Potato Salad]

A little dab will do ya.. literally.