ASUS P5W DH - Problems + Fixes Thread

Quote:

---

Originally Posted by J33pG33k

I tried 450x8 -> no go. So I loosened the memory timings a bit and bumped up the Vmch, Vcore, and Vdimm and still couldn't get the thing to POST. Yes, IIRC the Vdimm is still at 2.4V max. Any other suggestions for getting it beyond 400?

---

Is there a way you can get a hold of some D9GKX mem sticks? I've come to the conclusion that for some odd reason when it comes to higher 400+ FSB this mobo does not play nice with D9GMH regardless of the high binning. I have some high binned D9GMH and D9GKX and so far this has been my conclusion, of course I'm talking about 450x8 set in bios. Someone out there can prove me wrong. Or maybe it has something to do with the SPD timings that set in the memory itself.

Quote:

---

Originally Posted by nicepun

Is there a way you can get a hold of some D9GKX mem sticks? I've come to the conclusion that for some odd reason when it comes to higher 400+ FSB this mobo does not play nice with D9GMH regardless of the high binning. I have some high binned D9GMH and D9GKX and so far this has been my conclusion, of course I'm talking about 450x8 set in bios. Someone out there can prove me wrong. Or maybe it has something to do with the SPD timings that set in the memory itself.

---

When you say D9GMH you are speaking of the actual chip on the memory module, correct? How, may I ask, do you find this information?

Unfortunately I won't be able to get and other memory for a while, as I just don't have the funding right now (beer and books come first :D ).

J33pG33k,

since we have similar configuration, can you tell me what are your voltages (MB-NB/SB, CPU, RAM) and temperature?


Thanks.

Quote:

---

Originally Posted by nicepun

My Rev1.03G was fine for a month or so, after that it started to downgrade to the point that it could no longer do 390FSB.

---

Its funny, I think I´ve got the same problem, also with 1.03g. First week I was benching at 450, now, 1 month later, I can´t even boot at 440.
Do you change the thermal paste in the northbridge??
Do you use active cooling in the northbridge?
Its kind of weird a downgrade in such a short period of time.

Quote:

---

Originally Posted by monza1412

Its funny, I think I´ve got the same problem, also with 1.03g. First week I was benching at 450, now, 1 month later, I can´t even boot at 440.
Do you change the thermal paste in the northbridge??
Do you use active cooling in the northbridge?
Its kind of weird a downgrade in such a short period of time.

---

I also noticed something similar except mine was not at high FSB. I think it was my e6600 which dropped in clocks

Quote:

---

Originally Posted by monza1412

Its funny, I think I´ve got the same problem, also with 1.03g. First week I was benching at 450, now, 1 month later, I can´t even boot at 440.
Do you change the thermal paste in the northbridge??
Do you use active cooling in the northbridge?
Its kind of weird a downgrade in such a short period of time.

---

Yep, that sound about right, my kept degrading. I had the NB cooled and never saw temps above 34c.

Quote:

---

Originally Posted by lawrywild

I also noticed something similar except mine was not at high FSB. I think it was my e6600 which dropped in clocks

---

That's what I was thinking but recently changed mobos to a 1.02g w/ 1.85vmch and now I'm back to normal. The 1.04G had some issues so I sent it back for RMA.

hi all;)
i have this beautiful board, but i can´t overclock him over 3ghz.
i have an e6600 l624
p5wdh 1707 bios sn:66..
2 gig gskill ddr2-800
vcore i tested with 1,4v now it is 1,3875v
why i can´t overclock higher??

regards from austria;)

Quote:

---

Originally Posted by lawrywild

I also noticed something similar except mine was not at high FSB. I think it was my e6600 which dropped in clocks

---

Ugh. This is starting to happen to me as well. I used to be completely fine and stable at 385x9 1.60vcore 2.30vdimm 1.65vmch with 5-5-5-15 or even slightly tighter timings, and then I changed my hard drive configuration around to go to the ICH7 SB, now mirroring my two raptors as my OS/program files, and a couple of standalone drives as JBOD.

I also added the creative sound blaster x-fi platinum, and since I changed the HDD configs, I had to reload. However, ever since then, I've probably reloaded at least 10 times over the past several days, and even at 333x9 with relaxed mem 5-6-6-18 (my SPD setting), moving my tRFC and tRD to 42-8(or 8-42?), I'm barely stable.

Half the time, soon as I boot into windows, I'll get a bluescreen immediately and the box'll reboot (havent had a chance to turn off auto-restart to see what it is), while other times, I'll get registry complaints, other times certain pieces of my hardware won't work (sound blaster, geforce 8800, even my floppy drive), where usually just a reinstall of the drivers will fix it. I've increased my vdimm to 2.4 and my vcore to 1.61 and I'm still on good temps, but I'm getting *SO* tired of reloading. Seems like soon as I reload my box, it's okay. However, when I start adding drivers (logitech bluetooth mx5000, latest intel chipset, lan, wireless, geforce), then apps (symantec corp antivirus, systool/speedfan, java, nero, azureus), somewhere along the way, things get real unstable and the box will start rebooting, messing up the hardware along the way; at least until I can reinstall the driver and set my tRFC? to 8.

What do you guys think could be my problem? Half the times, programs will crash or complain during installation that a .cab or .msi is corrupted, but I lower my tRFC? to 9, it works. Did I burn out my CPU? MB? Ram? Is there an easy way to test? And what can I do since I'm past my 30 days with all my equipment.

Oh, lastly, I think I remember reading that people had issues with x-fi cards, but that they got resolved in 1707.. I'm on 1707, but I'm curious what issues people had. Maybe tonight I'll try to reload again for the third time today (did two reloads this morning before work) without the sound card in place to see if that's my culprit. I'd hate to get rid of that card, cause when the box is actually working, what you hear out of it is absolutely amazing, and blows away any onboard sound solution on the market today.

Thanks

Quote:

---

Originally Posted by xymon

hi all;)
i have this beautiful board, but i can´t overclock him over 3ghz.
i have an e6600 l624
p5wdh 1707 bios sn:66..
2 gig gskill ddr2-800
vcore i tested with 1,4v now it is 1,3875v
why i can´t overclock higher??

regards from austria;)

---

Disable nearly everything in your CPU Configuration, all the stuff below your memory timings in the chipset page, verify that your mem timings are manually set to proper spec that they came out of the box, and manually set your "jumperless config" page per my post a page or so back. Many of us have to push 1.60vcore to our chips to hit reasonable levels, especially those of us with earlier week cpu's, like you will probably have to, since you have an earlier week (24).

Best of luck.

Quote:

---

Originally Posted by monza1412

Its funny, I think I´ve got the same problem, also with 1.03g. First week I was benching at 450, now, 1 month later, I can´t even boot at 440.

---

Same here... There is no way it should be degrading like this...

I've noticed that there is a point with these Intel chips that they will get quite hot from a very small increase in voltage...like say, witha "b" chip, 1.45 is 40c under water, but 1.475 leads to 56c...a big jump for a minor increase.

At the same time this happens, if you run for extended periods, the cpu degrades. For "A" chips this value seems to be around 1.575v or 1.6v, however, for "B" chips it's around 1.45v to 1.5v.

You can always increase votlage a bit more, and get more clocks, but the gains you get will fall off rapidly...at this point you are close to detroying your chip, with the combination of heat and current.

Seems to me the resistor packs on the bottom of the "b" chips is what makes the difference in those...probably smoothing out the power less and delivering more current ot the chip than an "A" would get with it's different resistor packets.


Anyway, watch temps as you start to increase voltage...see that big jump in temps? You are in "TOO MUCH VOLTS FOR THIS CPU" land.


I am unsure is this is sometihng to do with the sodlering of the IHS onto the cpu, and the point at which the solder becomes in-effective @ dealing with the heatload, but it really seems to me that heat is not a major factor in these chips...it's current. Too much, and your cpu is dead, or degrading.

This also seems to apply regardless if the cpu is under water, or under air or phase...as soon as you see that big jump in temps, your cpu is in trouble.

So now with my new cpu cooler i idle at 30celsius and now iam more worried about the MB. Probe and speedfan shows 44celsius idle. SB, NB and mofset heat sinks is pretty hot when i touch em.. ive removed the covers on the NB and SB and put some AS5 instead. no AC5 on the mofset, still original there.

Thats strange im running my Core2Duo @ 333*9 and get a temp of 23oc

but a tip for the motherboard as ive got mine down to 30oc i just got a Powerfan that i plugged into a Molex and postioned it facing the SB and that basically killed off all the heat

This is my rig atm.. i think when i replaced the stock cooler the northbridge dont get any air at all on the heat spreads.

http://img470.imageshack.us/img470/8...0047jp3.th.jpg

Quote:

---

This is my rig atm.. i think when i replaced the stock cooler the northbridge dont get any air at all on the heat spreads.

---

A couple of 40mm fans screwed into the North & South sinks does the job nicely! ;) .....makes a hell of a difference :)

That North of yours looks tight with that 'beast' over part of it but if there is 8-10mm clearance you should be OK :)

Quote:

---

Originally Posted by scaramonga

A couple of 40mm fans screwed into the North & South sinks does the job nicely! ;) .....makes a hell of a difference :)

That North of yours looks tight with that 'beast' over part of it but if there is 8-10mm clearance you should be OK :)

---

hehe.. well theres like 3-5mm clearance =P the beast is an Noctua NH-U12.

thinking if i will be able to squeeze a 40mm onto the SB.

i just installed the mofset fan that cames with the MB i thought if the NB gets hot and send the heat to the mofset i will be cooled there at least.. since this cpu cooler blows sideways it wont be interfered as the stock cooler would have.. even read some warnings about it in the manual i think.

but if it wont cut it and i still get to high temps i was wondering about if i really should cut the heatpipe between the NB and mofset just to try installing this one. And then i will have to face the cpu cooler to blow up and out through the power supply.

WZ, lawry, and all that have seen substantial decreases in clock performances with your equipment:

Do you guys recommend buying another CPU as replacement, or going a different motherboard route, or both?

I'm thinking to first start with CPU replacement if my tests this evening are still reacting the same way. If that won't solve the problem, I'm thinking to do motherboard.. Even though we get better performance for our mhz vs the rd600's, I think that with a few bios updates and when ppl figure out tweaks, the rd600's have a good possibility to beat the 975's clock for clock. So probably the DFI RD600. :(

I was so happy at 385x9 and everything was running so well, that it literally breaks my heart that these problems have occurred. I'm eagerly anticipating the launch of Vanguard, and was hoping to eat up the last week and a half of the beta trying out new class/race combinations and exploring as much of the world as possible, but it's hard when you have an unstable, dead box. Hrmpf.

Quote:

---

Originally Posted by semiskilled

hehe.. well theres like 3-5mm clearance =P the beast is an Noctua NH-U12.

thinking if i will be able to squeeze a 40mm onto the SB.

i just installed the mofset fan that cames with the MB i thought if the NB gets hot and send the heat to the mofset i will be cooled there at least.. since this cpu cooler blows sideways it wont be interfered as the stock cooler would have.. even read some warnings about it in the manual i think.

but if it wont cut it and i still get to high temps i was wondering about if i really should cut the heatpipe between the NB and mofset just to try installing this one. And then i will have to face the cpu cooler to blow up and out through the power supply.

---

BTW- I looked at the pic of that bottom nb cooler you posted. Since I replaced mine with the Thermaltake Spirit II, and since one of my hooks holding it broke, I have a bit of experience in mounting the thing. IMHO, I don't think the arms of that cooler in your url look like they're going to fit well with the hooks of the nb on this mb. You're going to want to have adjustability on how much the arms go out from the center of the cooler, and from the pictures on the website, it looks like it's all stationary.

Your best bet is to go with one of the coolers listed on the first post of this article thread, since we've all used one of those and know they fit without much work.

Quote:

---

Originally Posted by nickfd

Do you guys recommend buying another CPU as replacement, or going a different motherboard route, or both?

I'm thinking to first start with CPU replacement if my tests this evening are still reacting the same way. If that won't solve the problem, I'm thinking to do motherboard.. Even though we get better performance for our mhz vs the rd600's, I think that with a few bios updates and when ppl figure out tweaks, the rd600's have a good possibility to beat the 975's clock for clock. So probably the DFI RD600. :(

---

I tried playing the stupid enthusiast when calling the intel support line, not mentioning any overclocking, just saying that I tried putting more voltage, and getting more favourable results. They basically said, "if you can post, your cpu is fine," and won't consider replacing it until you try a new motherboard, ram, and psu combination, or at least take them out of the troubleshooting loop.

On the bright side, we can get oem e6600's for just under 300$, but it blows that getting any kind of warranty replacements for free is out of the question soon as you do some mods. :)

Guess what?! Memtest showed me that one of my 1gb pc8500 dimms was bad. I tried it in different slots to rule out the mb, and it was bad in all, while the other was oki.

Gonna look up corsair's warranty and cross my fingers. I spent a lot of coin on these dimms. Glad to see I'm still able to hit my normal 385x9 and everything's functioning fine. Prime95s are running as I type. Woohoo!

My MB temp is 45c according to probe2. i know it shows less then the actual temps but can anyone tell me if this temp is acceptable? iam running only 333fsb with all volt settings set to auto.

Attachment 54929Booted under 390fsb clockgen up to 430fsb memory at 1.1 dimms needed 2.4 to be stable here. vcore 4.75.

What do you think of this..?
Maybe if i lower Vcore the temp will drop a bit?
I'm still in doubt what 24/7 config i will use. This one or 333x9.
At 333x9 i've got more bandwith so..

http://i121.photobucket.com/albums/o...check400_8.jpg

looks good redrock. My temps are low because im chilled liquid cooling. Have u manualy set your mem timings?" try 4.4.4.12. That mem should be able to do it.

Quote:

---

My MB temp is 45c according to probe2. i know it shows less then the actual temps but can anyone tell me if this temp is acceptable? iam running only 333fsb with all volt settings set to auto.

---

Kinda high for my liking buddy........mines sits at 29c and I run the same FSB as yourself........;)