ASUS P5W DH - Problems + Fixes Thread

[Trellame]

Why are people being a stinge with these damn BIOS's? put them up so i can host them for others...

Makes them feel "special"

[Falkentyne]

lawry, there are a dozen reasons why a file wouldn't be complete-- he could have attached a partial FTP download, or something. All I'm saying is that it's kind of ridiculous to say you have something, and then deliver something fake, unless you offered to send him money for it or something.

I can't make a judgement either way on the file, but a PM to Nemesis saying, "Hey, this file doesn't look complete/wrong" is better than calling him out publicly and saying don't ever buy or trust this guy.

It's not like he cheated you-- you didn't send him money did you? Anyway, getting way off topic here. Bottom line: It sounds like you got a partial file. Reasons why it could be a partial file are numerous, only one of which (highly unlikely) is that he deliberately misled you for intentions unknown.

Once is a random occcurance. Twice is a coincidence. Three times, maybe, and there's a pattern forming. This could go a long way with some of these posts on these forums too. "A lightning bolt exploded outside, and my FSB can now be overclocked an extra 1MHZ!!!!! EVERYONE, GO PUT UP LIGHTNING RODS FOR MAXIMUM OC!!" :)

Yes I agree.
larywild you were DEFINITELY out of line here, and you OWE Nemesis and everyone else an apology. People are NOT supposed to send out prereleases, doing so is a violation of trust from the company, yet EVERYONE and their momma seems to want them and does not care anyway. Yet he tried to do you a favor by doing something he really was not authorized to do, and things went wrong and you blasted him for it, which may have basically RUINED his chances of getting more test BIOSes. Because he was trying to help you and you screwed him.

Think about it.

Next time KEEP YOUR DISPUTES TO PRIVATE MESSAGES AND INSTANT MESSENGERS. You never know what negative effect something like this can have on people.

Imagine if someone had offered to give you an Engineering sample and something went wrong at the post office and you blasted him on a public forum. Who knows what legal problems you can cause by that crap?

[lopri]

Assuming the chipset clocks the same way P965 does (which it should), lower multies won't do much good without higher straps, which will results in lower performance clock-for-clock. And lower multies under same 1066 strap will only get you even lower FSB due to internal clock being higher under lower multies at a given FSB.

[BulldogPO]

Has anybody been in contact with Asus about those sub memory settings?
tRFC, and others would be nice in bios :)

Could somebody up those 1404 and 1405 bioses somewhere?

[[G.N.U.]Fragman]

I have the same problem with my machine, and it's driving me crazy. Natalia has the same issue with her P5WDG2 as well. I am trying to track down the common factor between these boards (besides that they are from Asus).

So far, one of the possible culprits could be the Raptors in RAID0. She has 2 Raptor 150s in RAID0. I also have 2 Raptor 150s in RAID0. You have 2 Raptor 74s in Raid0.

Which RAID controller are you using? The Intel one, or the JMicron one?

i sometimes have the same problem...not running any raid on my system (yet) booting without gfx and then putting back gfx normally solves the issue, but it shouldnt be like that. It doesnt happen everytime but just random.

[taemun]

How could you use the JMicron controller for raid anyways? None of the four SATA cables that came with my mobo fit in the rear slot?? :confused:

t

[topboy]

Yes I agree.
larywild you were DEFINITELY out of line here, and you OWE Nemesis and everyone else an apology. People are NOT supposed to send out prereleases, doing so is a violation of trust from the company, yet EVERYONE and their momma seems to want them and does not care anyway. Yet he tried to do you a favor by doing something he really was not authorized to do, and things went wrong and you blasted him for it, which may have basically RUINED his chances of getting more test BIOSes. Because he was trying to help you and you screwed him.

Think about it.

Next time KEEP YOUR DISPUTES TO PRIVATE MESSAGES AND INSTANT MESSENGERS. You never know what negative effect something like this can have on people.

Imagine if someone had offered to give you an Engineering sample and something went wrong at the post office and you blasted him on a public forum. Who knows what legal problems you can cause by that crap?


While i agree with your comments and i think this should be left to the 2 people involved to discuss offline, shouting and swearing about this is also a no no.

[Roch]

I'm getting an issue where the BIOS keeps freezing. It's weird, the blue bar along the top keeps shimmering (as if it's not crashed), but I have to use the reset switch to reboot as the keyboard doesn't respond. This is with 1305. After a couple of resets it's then ok. Once it's at desktop it's fine, never skips a beat during SP2004, Super Pi and extended gaming sessions.

[Falkentyne]

Fair enough. I suspect they probably already resolved it (hopefully), and with no lasting ill effects for Nemesis....

[Capo Crimini]

I'm getting an issue where the BIOS keeps freezing. It's weird, the blue bar along the top keeps shimmering (as if it's not crashed), but I have to use the reset switch to reboot as the keyboard doesn't respond. This is with 1305. After a couple of resets it's then ok. Once it's at desktop it's fine, never skips a beat during SP2004, Super Pi and extended gaming sessions.

Mine is doing that too, but it has something to do with detecting whatever it's trying to detect (fans/voltages etc). Just let it use its time (shouldn't take more than 30-40 seconds though).

[Roch]

Thanks for that, I'll try to be a bit more patient. ;)

[Capo Crimini]

Give me feedback if simply waiting works or not. It's always possible you have a different problem than me. :)

[politenessman]

i found a link on tom's hardware site. seems someone else had the same problem. stupid asus didn't include the correct intel drivers on my disc. my driver disc had intel raid ich8r instead of ICH7-R. i got ten blue screens trying to install windows over and over again thinking I was doing something wrong. I gave up and installed everything on two drives non-raid. now i'll have to start over and install from scratch. thanks asus! for over $200 you could at least include the right drivers. jerks.

if you make the floppy from dos, it makes it with the correct ich7r drivers.



very truly yours,
politenessman

[The Nemesis]

Fair enough. I suspect they probably already resolved it (hopefully), and with no lasting ill effects for Nemesis....
I apologize to all that this situation found its way to the forums & this thread. That said, I've moved on and learned a valuable lesson:)
http://img.techpowerup.org/060919/423 FSB Benchmark565.jpg

[ThorinMan]

What are your core temps like at 1.65 vCore?

[The Nemesis]

What are your core temps like at 1.65 vCore?
57' to 62'C during the 3DMark06 test. Freaking 76'C during orthos. I don't run over 1.5vcore on this cpu other than benching. It has proven to be not orthos stable over 3.7ghz @ any vcore, but it had the highest overclock @ auto vcore of any 6600 I've owned.

[lostsoul]

Hello every one. I bought this board as well, with a E6600 and 2 corsair sticks, 1 gig each (800) , and I cant get this thing to post. I tried with one stick, and even bought a cheap stick at bestbuy and that didnt even work.
What it does is, everything turns on, fans, lights and such, it sounds like the raptor tries to run a little, but thats it. The monitor always shows nothing at all, as if its not connected. It did work fine for a few days, I rana few games at max to see what it could do, and just last sat it went died and the enermax psu fan went full speed.
Im not sure if it has to do with the fact that I forgot to set anything up in bios, like the rams volts, to 2.1. I think the default setting for that is like 1.8. Could that be it? And if so, how would i get it to post so i can get to bios and set everything up?

[AeroSquid]

have you tried clearing the cmos lostsoul?

[imperiallord]

Hey Nemesis no need to make apologies for someone else mistake.

[lostsoul]

have you tried clearing the cmos lostsoul?
Yes, Ill tell you how I did it, just to make sure I did it right. Im semi new at building systems. I unplugged it, took out the batt, placed the jumper on clear, waited 30 sec, placed the jumper back, placed the batt back, and plugged it back in and turned it back on, and the same thing happens, it all goes on, all lights, all fans work, but no posting. Oh, one thing I forgot, I am using the 7950 as the video card.

[phile]

Yes, Ill tell you how I did it, just to make sure I did it right. Im semi new at building systems. I unplugged it, took out the batt, placed the jumper on clear, waited 30 sec, placed the jumper back, placed the batt back, and plugged it back in and turned it back on, and the same thing happens, it all goes on, all lights, all fans work, but no posting. Oh, one thing I forgot, I am using the 7950 as the video card.

Sorry if you already covered these items, but which BIOS do you have, and did you make sure to connect both the 20+4-pin and 4-pin PSU connectors?

-phil

[tommyshango]

I was wondering.. I am getting my heatsink and fan tomorrow and a bunch of other stuff. I have the p5w-dh deluxe and the x6800. I was wondering if someone can post a screenie of bios settings and various other settings for a max air overclock or at least so I can ballpark or compare what I am getting to others. This thread got so big after taking several hours to read it... well.. its big. Any specific help for me or shall i start a new thread for people with the x6800?

Thanks.

[tommyshango]

oh i forgot to mention i have the ocz gamestream 700w and the ocz ddr28000 voltage extreme 2 gb all in the thermaltake mozart tower. i bought the scythe infinity for cooling along with 10 120 mm fans.

[The Nemesis]

These settings yeilded me a bit over 3.3 ghz default voltages on my L628 E6600. There's a link to a screenie of it in my sig. It should be no troube for X6800 just keep in mind my multiplier is 9.

ASUS P5WDH Dlx Stock Overclock Settings

CPU : E6600
DDR2 : G.SKILL 2GB (2 x 1GB) F2-6400PHU2-2GBHZ
Bios : 1305
VGA : 2 eVGA 7900GT's SLI modified drivers
HDD : Two WD 320GB Serial ata raid
PSU : OCZ GameXStream 700w
OS : WinXp Pro SP2
OVERCLOCK (MHz) = 3306Mhz

BIOS SETTINGS

JumperFree Configuration

AI Overclocking set to: MANUAL

CPU Frequency : 367MhZ
DRAMM Frequency :735Mhz

Performance Mode :Auto
PCI Express Frequency : 100Mhz
PCI Clock Sync Mode : 33.33Mhz

Memory Voltage : 2.1v
CPU Vcore Voltage : Auto
FSB Termination V : Auto
MCH Chipset V : Auto
ICH Chipset V : Auto
--------------------------------------------------------

CPU Configuration

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

Advanced Chipset Settings (DDR2)

Configure DRAM Timing By SPD : MANUAL
DRAM CAS Latency : 4
DRAM RAS Precharge : 4
DRAM RAS To CAS Delay : 4
DRAM RAS Activate to Precharge Delay : 12
DRAM Write Recovery Time : 4

Hyper Path 3 : Disabled
DRAM Throttling : Disabled
__________________

[Super Nade]

Guys, what the hell is Microcode update? I could not find any info about this setting?

[tommyshango]

I played around with it a bit and got it to over 3.5 ghz with the stock cooler. I dont want to run it faster until I got osmething blowin on the memory sticks. My coretemp was 50 so far nothing higher. I think and hope to hit well over 4 ghz.

[lostsoul]

Sorry if you already covered these items, but which BIOS do you have, and did you make sure to connect both the 20+4-pin and 4-pin PSU connectors?

-phil

Yes, I did that, but I also connected the axtra 4 pins with the PSU connection, but that worked fine for a few days. Not sure if that was the right thing to do tho. I tried connecting just the 4 pin and leaving the other 4 pins out, and nothing. I never upgraded the bios, so its the original bios that came with the board, not sure any more, cant get in hehe.

[UZ7]

Hmm, maybe it's just me but I've noticed a slight speed boost in updating my ethernet drivers (September 13) v8.57.2.3. Just for those who didn't notice the update... :P. My internet has been lagging so maybe it's just me.

[Hicksetto]

Guys, what the hell is Microcode update? I could not find any info about this setting?

Hi,
I've found an information but it is too old dated...

"What is 'micro code update'? Well, all of Intel's 6th-generation processors have a little area that can store some software ('micro code') to deal with processor bugs right inside the CPU. Usually the motherboard BIOS loads this software into the processor right after boot-up. Each time Intel finds a new bug they try to create a new little software patch that is called 'micro code update'. These 'micro code updates' make the processor more reliable, but they always cost a bit of performance, because they basically turn off some internal features."

Reference: http://www.tomshardware.com/2000/07/31/intel/index.html

Thanks,

Hicks

[Grendel66]

i bought the scythe infinity for cooling along with 10 120 mm fans.
Aka "The Hurricane Tunnel" .. :p:

I never upgraded the bios, so its the original bios that came with the board, not sure any more, cant get in hehe.
You can find the BIOS version number written on a sticker that's on a socketed chip at the end of the middle PCI socket. Look for something like 0502, 0701, 1101 or similar.

[sniggle]

In many people's experiences that sticker number has been inaccurate. Mine says 0701 but was shipped with 0801.

[jagt]

What bios is considered to be the best right now? 1402? I should be getting this board in a few days, so it would be nice to know.

[Mordor]

Bios 1305 seems to be the most stable bios right now, personally I am using 1401 and it is working well. From some of the comments here it seems 1402/1403/1404/1405 are very much in development, and I am going to wait until they settle on a more final version before changing to a newer version.

[jagt]

Adjustable multipliers were implementet from bios 1402 and up, correct?

[catscit]

Adjustable multipliers were implementet from bios 1402 and up, correct?

correct, works on mine with 1402

[jagt]

Righ okay. What's wrong with bioses from 1402 and up then? I see alot of people avoid using them. Are they very beta-ish? If so, what are the "symptoms"?

[Timmaah]

1403 bios seems to oc better than the 1402 bios, or at least for me. Also reducing the multiplier, as said before tends to require more voltage.

[catscit]

1403 bios seems to oc better than the 1402 bios, or at least for me. Also reducing the multiplier, as said before tends to require more voltage.

more voltage?

I can run 380x8 at 1.65 but can't even boot 370x10 at 1.65!
I'm running 1402

[Dani]

Weird, with my Vmch modded P5WDH (BIOS 1402) MCH voltage goes in hand with FSB voltage :eek:

FSB= 1.40V + MCH= 1.50V --> 1.567V (measured)
FSB= 1.50V + MCH= 1.50V --> 1.652V (measured)

is that normal or what:confused:

[eva2000]

Weird, with my Vmch modded P5WDH (BIOS 1402) MCH voltage goes in hand with FSB voltage :eek:

FSB= 1.40V + MCH= 1.50V --> 1.567V (measured)
FSB= 1.50V + MCH= 1.50V --> 1.652V (measured)

is that normal or what:confused:
weird sure you didn't do the mod incorrectly ?

[Dani]

The mod is ok, nothing wrong there..
Mobo is "1.65Vmch version"

[deadlock7]

I also have problems with vmch.I've done all the mods(except vfsb) but i cannot raise the vmch above 1.75v.If I set 1.76v the pc boots but there is no display.This is a problem with my 6600gt card,because if i use a pci one i can raise it more than 2v!
Another serious problem i have is the fsb on boot.I cannot raise it above 410 in bios,but i can 450+ in windows.Has anyone found any solution to this?
My board is one of the first that were made with 401 bios.I've used many of the new bioses and now i have 1403.

[Icer]

<-------Having no new bios posted withdrawl symptoms.:yawn:

[berk]

^lol...
ive tried the latest bioses and all i get is a XP bar loading screen that freezes.
The only way back to my desktop is by setting F5 (save all defaults).
It's a real crapper. :(

[Timmaah]

more voltage?

I can run 380x8 at 1.65 but can't even boot 370x10 at 1.65!
I'm running 1402

probably shouldve specified which voltage i meant, im on about vcore voltage, with the multiplier set to x9 i can get 3.0ghz on stock voltages, with the multiplier set to x8 and fsb of 375 instead of 333x9 the pc wont even boot up.

[Joe Camel]

I also have problems with vmch.I've done all the mods(except vfsb) but i cannot raise the vmch above 1.75v.If I set 1.76v the pc boots but there is no display.This is a problem with my 6600gt card,because if i use a pci one i can raise it more than 2v!
Another serious problem i have is the fsb on boot.I cannot raise it above 410 in bios,but i can 450+ in windows.Has anyone found any solution to this?
My board is one of the first that were made with 401 bios.I've used many of the new bioses and now i have 1403.
ive found similar problems...

i cant boot with my MCH mod reading higher than the BIOS setting when i have a single video card in but i can boot @ 2.0V + (MCH) when i have both cards in (see sig)

IMO once you mod (any mobo) all bets are off on whats "normal".


PS just like your FSB boot limit, you can raise the MCH voltage higher once the rig has booted (but i have noticed a "video card related" limit on that also)



Dani,
once i get the Camel Pipe back from MikeG, ill see if my V-fsb effects my V-MCH

[Mxtra]

i'm thinking of running my conroe 6600 @ stock speed cuz it runs too hot oc'ed IMO. @3.2ghz it's almost 60 degrees on full load. @3.6 it idles in the mid 50's. too hot. my question is should I set vcore to auto @stock or should I bump it up an increment or two cuz of vdroop? i have it @stock now with vcore 1 step up from default. idles @36-37 degrees. wouldn't mind getting it down another degree or two if you guys think it's ok to drop it down. please let me know cuz I'm a novice at this stuff. thanks.

[Brahmzy]

i'm thinking of running my conroe 6600 @ stock speed cuz it runs too hot oc'ed IMO. @3.2ghz it's almost 60 degrees on full load. @3.6 it idles in the mid 50's. too hot. my question is should I set vcore to auto @stock or should I bump it up an increment or two cuz of vdroop? i have it @stock now with vcore 1 step up from default. idles @36-37 degrees. wouldn't mind getting it down another degree or two if you guys think it's ok to drop it down. please let me know cuz I'm a novice at this stuff. thanks.
What cooler/paste etc. are you using?

[Mxtra]

arctic silver 5 and the big typhoon in an antec p180 case with the front fan disconnected. @3.2ghz and the side panel off it maxes out 57 degrees on full load. seems to hot for me. must sound silly in this forum.

[Trellame]

arctic silver 5 and the big typhoon in an antec p180 case with the front fan disconnected. @3.2ghz and the side panel off it maxes out 57 degrees on full load. seems to hot for me. must sound silly in this forum.

I have that same case, if you disconnect the front fan, you don't really have anything bringing cool air into it...

[Mxtra]

i used to have it on but it's pointed the wrong way. air blowing out and plus it makes my pc louder. so at stock should i set it to auto or compensate for big vdroop? i don't know anything about this. does vdroop get worse the more a pc is oc'ed? is it not so bad at stock speed? i'm expecting my very first good video card (1950xtx)in the mail thursday so i'd like to get this settled before I play f.e.a.r. and oblivion. I've never owned a video card that cost more than $70 before.

[Trellame]

i used to have it on but it's pointed the wrong way. air blowing out and plus it makes my pc louder. so at stock should i set it to auto or compensate for big vdroop? i don't know anything about this. does vdroop get worse the more a pc is oc'ed or is it not so bad at stock speed?

You can just flip it around then... All the fans that came with my case also had a little white control attached which allowed 3 different speed settings.

[Mxtra]

the low setting is the only one my ears find acceptable. i doubt it'll make a difference on low. I have another issue that made me revert to stock. if I overclock my pc for more than a day and I run a multithread program like nero recode or svcd2dvd, the second core does nothing. i reboot and try it again and suddenly both cores are being used. i'm now running at stock to find out if the same thing will happen. i also convinced myself to be responsible and try to make this pc last for many years instead of killing it quickly so i'll have an excuse for quad core.

[Kemp3]

ok I have seen other members post this same question , but have yet to see a difinitive answer. Here is the scenerio; try a higher oc , add a lil more volts and prime crashes in a few seconds. Ok add another bump on voltage and it last a few minutes, but the next bump on voltage makes the comp restart during prime. Any following voltage after that has the same results, Prime does not error my comp restarts. I have verified its not a ram issue with several runs of memtest, I also have tried adding voltage to MCH and FSB ( i have maxed them ) and still recieve the same results.

Any Ideas?

[ineedaname]

Hey guys what's the max fsb on the P5W DH??
asus website says 400 but it doesn't seem like that's the case.

[The Nemesis]

ok I have seen other members post this same question , but have yet to see a difinitive answer. Here is the scenerio; try a higher oc , add a lil more volts and prime crashes in a few seconds. Ok add another bump on voltage and it last a few minutes, but the next bump on voltage makes the comp restart during prime. Any following voltage after that has the same results, Prime does not error my comp restarts. I have verified its not a ram issue with several runs of memtest, I also have tried adding voltage to MCH and FSB ( i have maxed them ) and still recieve the same results.

Any Ideas?
When I get to this point, I accept I've reach the limit of the current configuration. That could mean the cpu has reached its limit with the cooling being used or it can also mean the chipset cannot handle the stress prime is placing on it @ such a high FSB. The only way to know if the board can't handle the high fsb is to try same FSB with a lower multiplier. For example, there was no way for me to know my board could do orthos stable with my E6600 @ 440 FSB because my cpu would only complete 32mb pi @ 3960Mhz. Lowering the multiplier to 8 made it 3520 which was passed orthos over 12 hours. Had this failed it would tell me it was my chipset since my system was already 12 hours orthos stable @ 400 X 9 = 3600mhz with less volts.

[EJTREM3]

why the south brige its soo damm hot, temp is 80-85c, can remove de plate whit name asus, and put one 40mm fan ?

[The Nemesis]

why the south brige its soo damm hot, temp is 80-85c, can remove de plate whit name asus, and put one 40mm fan ?
Alot of people have but I am unsure if removing the plate voids Asus's 3 yr warranty on the motherboard.

[phile]

why the south brige its soo damm hot, temp is 80-85c, can remove de plate whit name asus, and put one 40mm fan ?

80C is ridiculous. You may want to RMA the board. It could just be that the heatsink is not making good contact with the ICH7R chip, but you do risk voiding the warranty by removing and reseating it with a fresh application of thermal compound. Removing the heatsink caps will help, but not enough to bring a temp of 80C down to normal.

-phil

[revenant]

80C is ridiculous. You may want to RMA the board. It could just be that the heatsink is not making good contact with the ICH7R chip, but you do risk voiding the warranty by removing and reseating it with a fresh application of thermal compound. Removing the heatsink caps will help, but not enough to bring a temp of 80C down to normal.

-phil

I pulled my SB sink off to put AS5 between it and the chip and only the very edges of the factory thermal paste were making contact with the chip!!!! the whole entire chip middle was totally spanking clean.. crazy! I would take your sinks off and take a look to make sure they're making contact well.. I am really glad I did. whew.

[Capo Crimini]

Hey guys what's the max fsb on the P5W DH??
asus website says 400 but it doesn't seem like that's the case.

1305 goes to 550.

[BulldogPO]

Same thing here on my board.
Only edges of SB were touching heatsink.
SB sink is banana.

[SteveOCZ]

Here is the 1405 bios
http://www.spec3.net/asus/P5W-DH-ASUS-Deluxe-1405.rar

Before flashing to this bios I was limited to 405FSB stable. I've only tried upto 440FSB, but it's stable so far on a stock board.
:woot:

[taemun]

zomg I love you Steve :D

t

[deadlock7]

ive found similar problems...

i cant boot with my MCH mod reading higher than the BIOS setting when i have a single video card in but i can boot @ 2.0V + (MCH) when i have both cards in (see sig)

IMO once you mod (any mobo) all bets are off on whats "normal".


PS just like your FSB boot limit, you can raise the MCH voltage higher once the rig has booted (but i have noticed a "video card related" limit on that also)

Interesting info.I thought that the problem was that my card can't stand lot of volts and i was going to buy a new one.If you have the same problems with a single card and you can get 2v with cf,then i don't know what is going on.
A friend of mine doesn't have any problem getting vmch 1.9(he hasn't tried more yet) with a single 1900xtx,so i'm confused again.

The vmch doesn't help getting fsb more than 414 on boot.If i set 415 i get blank screen,even with a pci card(that it helps me get 2v mch).The same friend also had fsb boot problem around 405,but he flashed back to 801 bios and then he could boot fine 450+.Now he uses the 1403 and raises fsb just like doing it in windows.I tried it too,but no luck..

[SteveOCZ]

I'm at 460FSB now :)

[arne72]

ist this final or beta?

[SteveOCZ]

Beta

[arne72]

460 @ ? mch

[SteveOCZ]

1.85

[Mykou]

1.85
do you know if it support ES chips ?

[SteveOCZ]

I assume it's just like all the other 140x files we've seen. If they support it so does this one.

[BulldogPO]

Thanks! Gotta try that 1405 bios tomorrow morning.
Damn eveningshift :(

[SteveOCZ]

My board doesn't seem to want to run over 450FSB stable now, but 45 extra FSB from a bios flash aint bad. Now I wait for 1406 :)

[GLENBOY]

yes all the 140x bios work ok for es i have a 6700 and its fine, also downward multiplier works from bios

[Fr3ak]

I am wondering why most people want a new bios. I am running 701 and everything is working great. Can boot into windows 01 stable at 445Mhz, after that my cpu is at its end I suppose.
The only annoying thing is the nic loosing drivers and the silly restart behaviour which I doubt wont be disabled with just a simple bios update.
I am almost sure Asus doesnt try to make the board more overclocker friendly with bios updates as themajority of people who buy the board dont know what overclocking is.
Never change a running system, so unless you have major problems, dont flash t he bios in hope for better oc results. Maybe other components are the limiting factor.
Of course I am always praying for a new unbelievable oc behavior after a bios update, but experiance has shown that thats rarely the case :)

[ThorinMan]

I am wondering why most people want a new bios. I am running 701 and everything is working great. Can boot into windows 01 stable at 445Mhz, after that my cpu is at its end I suppose.
The only annoying thing is the nic loosing drivers and the silly restart behaviour which I doubt wont be disabled with just a simple bios update.
I am almost sure Asus doesnt try to make the board more overclocker friendly with bios updates as themajority of people who buy the board dont know what overclocking is.
Never change a running system, so unless you have major problems, dont flash t he bios in hope for better oc results. Maybe other components are the limiting factor.
Of course I am always praying for a new unbelievable oc behavior after a bios update, but experiance has shown that thats rarely the case :)

Lower mults?

[Fr3ak]

Why do I want to lower the multiplier? 4ghz is fast enough if you ask me ;) ven 3.6Ghz is enough for 24/7 usage.

But Allendale users might want to have that option. Didnt know it was included in latest Bios files. So I might give it a go to :)

[Super Nade]

1.85

So the early adopters are outta luck? Are we stuck with 1.65V?:(

[_JP_]

Here is the 1405 bios
http://www.spec3.net/asus/P5W-DH-ASUS-Deluxe-1405.rar

Before flashing to this bios I was limited to 405FSB stable. I've only tried upto 440FSB, but it's stable so far on a stock board.
:woot:

Link down :(

[phile]

http://www.xtremesystems.org/forums/showthread.php?t=116777 :slapass:

-phil

[Pjoeloe]

ASUS launches Intel Core duo 2 P5N32-SLI Premium/WiFi-AP Nvidia Nforce 590 Quad SLI with 3 x PCIe x16 for a third PCI Express x16 slot / Nvidia's upcoming Physics card

http://www.asus.com/news_show.aspx?id=4329

hope it clocks :D

[Mykou]

hmm just saw that , angry for some benches with that mobo :D

[eva2000]

Here is the 1405 bios
http://www.spec3.net/asus/P5W-DH-ASUS-Deluxe-1405.rar

Before flashing to this bios I was limited to 405FSB stable. I've only tried upto 440FSB, but it's stable so far on a stock board.
:woot:that's a huge difference there!

[Blacklash]

439 on an 8x multi was the best I could ever boot and run with my P5W DH D, then again I had one that only did 1.65v on the MCH :( I had great pi times with it though.

[Boglwe]

I just pulled that copper cover off the NB heat sink. When I did this, I pulled the heatsink from the chip. I dont really want to OC this rig with my e6600 that much, Is it ok if I just snug it back in, or do I need to replace the compound now?

[revenant]

you should really pull the sink off to remove that cover anyways.. You can just push it back on.. wiggle it side to side /very/ gently as you push down (gently) so the thermal compound makes contact with the whole chip top.. the Asus compound is kind of dry and puddy like..

[yumzlare]

hi bros out there,

Anyone of of u guys using the P5W DH DELUXE with sycthe ninja plus??? Is it compatible???? UIRgent

i've having this problem all my USB ports can only be used when only i remove the firewire port. Y is this so??? how am i gonna use my firewire port then.

if i were to install the firewire port only myI/O panel USB can only be used... any of u guys having the same problems,..

REGARDS

[Trellame]

Here's an excellent thread (though not in English) about the P5W-DH with a database of overclocks along with links to the latest everything, including 1405 bios:
http://www.forumdeluxx.de/forum/showthread.php?t=256507

[sherpa13]

I'm looking for help with Dual Channel RAM. My memory is currently running in Single Channel mode.

I. I have two 1 GB Dimms in Dimm A1 and A2 - it will post in this config

A. Whenever I try ro run them in Dimm A1 and B1 it won't post
B. Whenever I try to run them in Dimm A2 and B2 it won't post
C. Whenever I try to run them in Dimm B1 and B2 it won't post

Basically, whenever I insert a module into either Channel B Dimm, it will not post. This occurs all the time, i.e bench testing, no graphics card, no overclocking, overclocking...

I have two more 1 GB modules I would like to install but I seem to be missing something.

Any help?

[ppcli]

whats type of speeds should i be getting with HD tach running 2x 74 gig raptors,with the p5w dh?i was hopeing that they would be faster then the asus nf4 sli del.

cheers

[R.MnTnA]

Another good one here alot of people using P5W
http://www.hardforum.com/showthread.php?t=1075792

Thanks for everyone's input on this forum it has helped me get my E6660 right where I wanted it and stable. 3.25ghz Thanks again!

[sniggle]

So where is this mysterious 1405 official release?

[MacClipper]

Just tried out SLI (2 x el cheapo 7300GTs) for the 1st time on my P5W the past week and noticed certain strange quirks about RAM being reported differently by the mobo, Windows and CPU-Z.



Installed is 2 x 1 GB sticks = 2048MB - everything reported OK without SLI. With SLI, the mobo POST reports as 1920MB (OK, kind of expected), Windows as 1.6GB (WTH, @%%!) while CPU-Z maintained at 2048MB.



Is something being configured wrongly? Looked around but can't seem to find a good answer, TIA.


http://img115.imageshack.us/img115/3960/systempropertiesmk1.png (http://imageshack.us)http://img115.imageshack.us/img115/8917/cpuzramih2.png (http://imageshack.us)



PS: regarding the RAM timings - still have to run my RAM at SPD on my P5W mobo till now.

[phile]

I'm looking for help with Dual Channel RAM. My memory is currently running in Single Channel mode.

I. I have two 1 GB Dimms in Dimm A1 and A2 - it will post in this config

A. Whenever I try ro run them in Dimm A1 and B1 it won't post
B. Whenever I try to run them in Dimm A2 and B2 it won't post
C. Whenever I try to run them in Dimm B1 and B2 it won't post

Basically, whenever I insert a module into either Channel B Dimm, it will not post. This occurs all the time, i.e bench testing, no graphics card, no overclocking, overclocking...

I have two more 1 GB modules I would like to install but I seem to be missing something.

Any help?

What an odd problem. Do you have vdimm set to the rated voltage? Probably 2v-2.2v.

What happens if you run two dimms in the B channel, with no dimms in the A channel?

-phil

[taemun]

Just tried out SLI (2 x el cheapo 7300GTs) for the 1st time on my P5W the past week and noticed certain strange quirks about RAM being reported differently by the mobo, Windows and CPU-Z.

Installed is 2 x 1 GB sticks = 2048MB - everything reported OK without SLI. With SLI, the mobo POST reports as 1920MB (OK, kind of expected), Windows as 1.6GB (WTH, @%%!) while CPU-Z maintained at 2048MB.

Is something being configured wrongly? Looked around but can't seem to find a good answer, TIA.
Do those cards steal RAM under the guise of making video ram larger? If so, thats obviously what you are seeing... if not... dunno...

t

[MacClipper]

AFAIK, the 7300GT cards do not utilise any TurboCache style RAM stealing.

[NO1B4ME]

whats type of speeds should i be getting with HD tach running 2x 74 gig raptors,with the p5w dh?i was hopeing that they would be faster then the asus nf4 sli del.

cheers

It would smoke your Nforce 4 setup IF you setup the Matrix raid.

[AndrewZorn]

Problem: DVD drive spins at 2x max on intel port
solution: buy a longer cable and use the jmicron port

its all i or asus could come up with besides RMA

[BulldogPO]

My NEC ND-3520 works fine in intel port.

[MacClipper]

Problem: DVD drive spins at 2x max on intel port
solution: buy a longer cable and use the jmicron port

its all i or asus could come up with besides RMA
imo, 2x max speeds is a giveaway that it is somehow operating in PIO mode.

[sherpa13]

What an odd problem. Do you have vdimm set to the rated voltage? Probably 2v-2.2v.

What happens if you run two dimms in the B channel, with no dimms in the A channel?

-phil


Nothing...the fans spin but no post. Under the current config, when you boot the VGA fan spins really fast then slows down. After the VGA fan slows, the systems posts and continues to load windows. Whenever there is a module in Channel B, the VGA card fan continues to spin really fast...but no post, no boot.

Maybe a conflict with the card?

I'm stumped.

[phile]

Nothing...the fans spin but no post. Under the current config, when you boot the VGA fan spins really fast then slows down. After the VGA fan slows, the systems posts and continues to load windows. Whenever there is a module in Channel B, the VGA card fan continues to spin really fast...but no post, no boot.

Maybe a conflict with the card?

I'm stumped.

Obviously, if you had spare parts you could swap in, in order to rule-out each component, that would be helpful. Instinctively, though, this sounds to me like either an issue with the mobo, or the PSU (assuming the memory is fine).

Have you tried booting with minimal hardware connected, with BIOS at factory settings, and all unneeded components disabled? Strip it down to the bare essentials.

-phil

[Blauhung]

it goes to standby just fine, but when I come back, the fans all start up, and i think i hear the hard drives go, but video doesnt seem to come back, and the keyboard doesnt initialize to the point where i can turn on and off the locks (num/caps/scroll)

Wondering if this has something to do with full power off on rebot or if i have some other setting that is screwing with me.

Core 2 x6800 (10*350 1.4vcore)
P5W DH bios 1305
ocz plat ddr2-800 rev 2 (1:1 @700 4-4-4-12 2.1vdimm)
2x 74gb raptor raid0 on ezraid ports
evga 7950gx2
antec TPII 550W
win xp home sp2

[sherpa13]

Obviously, if you had spare parts you could swap in, in order to rule-out each component, that would be helpful. Instinctively, though, this sounds to me like either an issue with the mobo, or the PSU (assuming the memory is fine).

Have you tried booting with minimal hardware connected, with BIOS at factory settings, and all unneeded components disabled? Strip it down to the bare essentials.

-phil

I have taken it all the way down...out of the chasis, multiple psu's, , another set of RAM (Cricial), no hard drives. The only thing I can't test is the graphics card (the x1900xt is the only PCEe card I have).

I am going to run it down to our local computer shop and see what they can find.

Thanks - I'll post any results ASAP

[Falkentyne]

Problem: DVD drive spins at 2x max on intel port
solution: buy a longer cable and use the jmicron port

its all i or asus could come up with besides RMA

I'm using two DVD drives hooked up to the Intel port without any problems. Master/slave.

Liteon 166S, and 1693S.

Didnt you say that the drives were being forced into PIO mode?
And that getting a different cable did not help ?

[Laughy]

Quick Tip:
If you're having trouble booting into Windows, or it takes a while for the screen to come up, increase your MCH voltage.

To those getting 450+FSB, do you think it's really worth it? Wouldn't pumping that much voltage into CPU/MEM/MCH greatly shorten the life of your components with not much performance gain?

I'm currently at 410FSB with:

CPU MULT: 8x
CPU VOLT: 1.5V
MEM VOLT: 2.0V
MCH: 1.65V

I think these voltages are low enough to get the most out of overcloking the mobo without drastically lowering the life of anything. Opinions??

[phile]

I have taken it all the way down...out of the chasis, multiple psu's, , another set of RAM (Cricial), no hard drives. The only thing I can't test is the graphics card (the x1900xt is the only PCEe card I have).

I am going to run it down to our local computer shop and see what they can find.

Thanks - I'll post any results ASAP

Well, it would appear you have narrowed it down to either a faulty vid-card or mobo. Hope you get that sorted-out quickly.

-phil

[revenant]

People running G.Skill 2gbhz on the p5w dh del.. what vdimm are you using? I have mine at 2.00v but thinking maybe 2.10v might be better? maybe that's why I was having issues at 3.6ghz.. been camped out at 3.5 with no real need for 100mhz more.. except for bragging rights. lol.

Anyways.. did the 1405 bios really help people with higher more stable OCs?

[The Nemesis]

I use 2.1v 4-4-4-12 800mhz. My sticker said 2.1-2.2v 800Mhz so I've always set it there @ 9 X 400. The 1405 bios gained me a few mhz 3DMark stable. It went from 420Mhz to 423Mhz. All other max clocks pretty much stayed the same. Pi scores got a tad better. From my understanding the 1405 final bios will be a tad better than the beta we've all used, maybe.

[Laughy]

Interestingly, I can run at x9 = 3.690ghz on my E6600, but my graphics are all screwed up when playing 3d games. I'm ONLY changing the multiplier too.

x8 = 3.28 ghz works fine. Raising vcore to something like 1.65V helps (but does not fix) in x9, but I don't want to raise it any higher just for 400 extra mhz!

[hulleyrob]

Originally Posted by kevpuk
I have read that a newer version of the app often solves problems....

Quote :

To get 1.00.29 to work, I copied over \Acpi\, \Acpi64\, \IO\ and \LANG\ from 1.00.27. Then replaced all the files under \IO\ with the same ones in 1.00.29 root folder. Do not replace files in ACPI because it will give you another error when you try to install.

I can confirm that this version of DH Remote (1.00.29) does work without any help with USB hub or anything. Simply plug in USB receiver to the ports under LAN2 and DH Remote application picked up the receiver right away.

Appearantly this version installed an "ATK0110 ACPI UTILITY" under System Device which was missing in all the previous versions of DH Remote app.

But have yet to track down said version......living in hope here, guys & gals....

From

http://forums.overclockers.co.uk/showthread.php?t=17620351&highlight=p5w+remote

Rob

[i||uSi0n^]

there is newer version for DH Remote on the asus ftp "DHRemote10036.zip" i never gave it a try as mine worked without any problem since day one...anyhow here is the link:

ftp.asus.com.tw/pub/ASUS/misc/utils/DHRemote/

i||uSi0n^

[sniggle]

there is newer version for DH Remote on the asus ftp "DHRemote10036.zip" i never gave it a try as mine worked without any problem since day one...anyhow here is the link:

ftp.asus.com.tw/pub/ASUS/misc/utils/DHRemote/

i||uSi0n^

I cannot seem to ever get the remote to be responsive, no matter which version of the software I use. The only program I ever try to control with it is Winamp, and very consistently I have to keep pressing a button for about 10 seconds before it will react, and then as long as I keep pressing buttons within 3-4 seconds of each other it is responsive. As soon as I don't press anything for at least 5 seconds though, it takes another 10 seconds to get it to react. Wish I could figure out why that is.

[miochza]

So my situation is that it requires alot more vcore for lower multipliers. Is it worth the extra voltage?

[Falkentyne]

So my situation is that it requires alot more vcore for lower multipliers. Is it worth the extra voltage?

Just change the multiplier in windows with crystalCPUID (I think that's what its called). It requires that much voltage because the chip is initially detected as the multiplier it was designed at, so it has to start an initial power up at that multiplier, and then the motherboard changes the multiplier.

Are there any boards that on fully unlocked CPU's, are able to change to a much lower multiplier without requiring an initial voltage to just power the chip at that FSB+default multiplier first?

(example: Trying to use 6x450 (2750 mhz) on an X6800 would require the same vcore as it would take to have the chip to at least pass the power on cycle at 4950 mhz! (450x11).

[revenant]

The Nemisis - thanks for the info! :)

[revenant]

doh... dbl post.. gah!

Interestingly, I can run at x9 = 3.690ghz on my E6600, but my graphics are all screwed up when playing 3d games. I'm ONLY changing the multiplier too.

x8 = 3.28 ghz works fine. Raising vcore to something like 1.65V helps (but does not fix) in x9, but I don't want to raise it any higher just for 400 extra mhz!

Take your pci-e frequency setting off of auto. I think that moves up when on auto as you raise the fsb.. I found out the hard way also... I hard set mine at 100 just to take it easy on my cards... 110 I got all kinds of corrupted graphics when there was a lot of movement on the screen.

[Laughy]

doh... dbl post.. gah!



Take your pci-e frequency setting off of auto. I think that moves up when on auto as you raise the fsb.. I found out the hard way also... I hard set mine at 100 just to take it easy on my cards... 110 I got all kinds of corrupted graphics when there was a lot of movement on the screen.

That's not it...it only happens when I change CPU multiplier, not FSB. My FSB at 410 x 8 = stable graphics, but 410 x 9 = corrupted graphics. Also my pci-e freq. is already at a set value. :(

[jagt]

What controller do you guys recommend running RAID on? I understand there's this EZ RAID thingy, in addition to the chipset RAID? What's the fastest? I have 2x 200GB WD SATA that I will run in RAID 0, and 2x 320GB WD SATA2 that I'm going to use for storage.

[mikepaul]

Take your pci-e frequency setting off of auto. I think that moves up when on auto as you raise the fsb.. I found out the hard way also... I hard set mine at 100 just to take it easy on my cards... 110 I got all kinds of corrupted graphics when there was a lot of movement on the screen.
I've been wondering what might be causing the flashes I've been seeing. My 'cheap' 7300GS didn't seem to have any problems before I used the 'Overclock 30%' profile. I'll set that frequency by hand and see if the flashes stop...

OK, small problem: using the OC profile, I can't play with individual settings for my E6300. Is there a program that can spit out what I've been using successfully so I can set things manually and then adjust? So far, every time I try settings found in OC postings, like http://www.techspot.com/article/13-intel-core2duo-e6300-e6700-overclocking/page2.html I just get to watch my computer choke and restart...

[Tamerlane]

During post when the Jmicron SATA/IDE controller identifies attached hardware, does everyone else get the "Press any key to continue ..." prompt with a short pause, even when you have set the IDEO timeout setting or whatever in BIOS to 0 seconds?

Has any fellow impatient out there figured out a way to bypass or avoid this pause/prompt?

[Durzel]

Just change the multiplier in windows with crystalCPUID (I think that's what its called). It requires that much voltage because the chip is initially detected as the multiplier it was designed at, so it has to start an initial power up at that multiplier, and then the motherboard changes the multiplier.

Are there any boards that on fully unlocked CPU's, are able to change to a much lower multiplier without requiring an initial voltage to just power the chip at that FSB+default multiplier first?

(example: Trying to use 6x450 (2750 mhz) on an X6800 would require the same vcore as it would take to have the chip to at least pass the power on cycle at 4950 mhz! (450x11).Actually that's slightly wrong. XE chips are the exception to the rule.

XE chips have something internal to them that mean their multipliers actually are truly unlocked. XE chips were able to have their multipliers changed in the BIOS long before the latest BIOS came out which made it possible for ES and Retail (yes, Step 5 ES is no different to retail in terms of not being able to change multipliers in BIOS).

So it's likely when you set the mulitplier to 6x on an XE, it will POST at 6xFSB, not its default multiplier of 11x. ES and Retail on the other hand appear to boot (or try to) at whatever their factory multiplier is before the BIOS clocks it down ala CrystalCPUID.

That's why they need so much vCore for lower multipliers even when the actual clockspeed (once its actually using the multiplier) is much lower.

[MacClipper]

Just tried out SLI (2 x el cheapo 7300GTs) for the 1st time on my P5W the past week and noticed certain strange quirks about RAM being reported differently by the mobo, Windows and CPU-Z.



Installed is 2 x 1 GB sticks = 2048MB - everything reported OK without SLI. With SLI, the mobo POST reports as 1920MB (OK, kind of expected), Windows as 1.6GB (WTH, @%%!) while CPU-Z maintained at 2048MB.



Is something being configured wrongly? Looked around but can't seem to find a good answer, TIA.


http://img115.imageshack.us/img115/3960/systempropertiesmk1.png (http://imageshack.us)http://img115.imageshack.us/img115/8917/cpuzramih2.png (http://imageshack.us)



PS: regarding the RAM timings - still have to run my RAM at SPD on my P5W mobo till now.
Got a different revision of the P5W mobo, swapped it in and... RAM is still reported as 1.62GB so that probably rules out the mobo too.

The only thing is certain is,

P5W + (2 x 1GB RAM) + (2 x Leadtek 7300GT) = missing RAM in Win XP/SP2.

P5W = tried 2 different revisions
RAM = tried 2 different brand/models
7300GT - tried different cards
Win XP = tried totally clean install without drivers

I'm totally stumped.

[mikepaul]

I'm totally stumped.My 3DFuzion 7300GS is supposed to swipe 256MB of system RAM to become a 512MB card, so maybe that's what's happening to you, too?...

[Falkentyne]

Actually that's slightly wrong. XE chips are the exception to the rule.

XE chips have something internal to them that mean their multipliers actually are truly unlocked. XE chips were able to have their multipliers changed in the BIOS long before the latest BIOS came out which made it possible for ES and Retail (yes, Step 5 ES is no different to retail in terms of not being able to change multipliers in BIOS).

So it's likely when you set the mulitplier to 6x on an XE, it will POST at 6xFSB, not its default multiplier of 11x. ES and Retail on the other hand appear to boot (or try to) at whatever their factory multiplier is before the BIOS clocks it down ala CrystalCPUID.

That's why they need so much vCore for lower multipliers even when the actual clockspeed (once its actually using the multiplier) is much lower.

Durzel, I think you are wrong here.
This doesn't work on my retail x6800.
It isn't about the x6800's being truly unlocked (they are), but its about how the P5W handles the multipliers. But I think I was wrong by talking about "POST", as the x6800 never even makes it to that point--it doesn't even initialize.

336x11 (3700) POSTS just fine @ 1.52v on my p5wdh and loads windows (requires more than 1.6v on air for stability and temps go through the roof so forget it), however,

370*10 *does NOT* POST at all at 1.52v.
I have to raise the vcore to 1.575v for it to POST.
So the chip is initializing at 4070 mhz first, with the 11 multiplier, before the BIOS is able to take over and set it to 370*10. You see? How can the multiplier be lowered if the chip can't even powerup enough to allow the BIOS code to run?

Note, after failure to POST (at 1.52v), if I hold down the power button to power off and then power on again, I get the expected "overclock failed" and the system is reset to 2266*11 powerup.

I am not the only X6800 user who has seen this. Check in the Intel support forum. This problem has been discussed there too.

In order for your method to have worked, the last requested multiplier would have to have been saved on the CPU somehow internally. I remember some years ago there was a forum post explaining exactly what happens on power on, before BIOS code can take over.

So very low multipliers (like 6 or 7) with high fsb's will never work unless the cpu could power on with its default multiplier at that FSB and mhz.

I shouldn't have said 'the chip is detected as...", rather "the chip powers on at the multiplier it is designed for", and therefore, the BIOS never has a chance to load, because it is the BIOS that has to change the multiplier. If I'm wrong about this, tell me. The CPU does not have "flash memory", so the new multi can't be "flashed" to it (to be retrieved at cold PWR good/on). It can be requested, but to get to that point requires, at power up, that the chip at least be able to power up and execute enough basic code so the BIOS can load.

[virtualrain]

Durzel, I think you are wrong here.
This doesn't work on my retail x6800.
It isn't about the x6800's being truly unlocked (they are), but its about how the P5W handles the multipliers. But I think I was wrong by talking about "POST", as the x6800 never even makes it to that point--it doesn't even initialize.

336x11 (3700) POSTS just fine @ 1.52v on my p5wdh and loads windows (requires more than 1.6v on air for stability and temps go through the roof so forget it), however,

370*10 *does NOT* POST at all at 1.52v.
I have to raise the vcore to 1.575v for it to POST.
So the chip is initializing at 4070 mhz first, with the 11 multiplier, before the BIOS is able to take over and set it to 370*10. You see? How can the multiplier be lowered if the chip can't even powerup enough to allow the BIOS code to run?



As you probably know, the MCH could also be a factor in whether you can boot with an FSB of 370.


So very low multipliers (like 6 or 7) with high fsb's will never work unless the cpu could power on with its default multiplier at that FSB and mhz.


I come from the AMD world where people have been running lower multi's with higher FSB without issues. How does AMD do it? Can it not be the case that the clock generator reads a flash register to clock the CPU at startup therefore not requiring any BIOS to execute in order to run at the right speed? Alternatively, it might be a register that is kept alive during power off much like the system clock. There are really a ton of different ways the clock multipler can be set and read without he Bios loading. I have no idea how it actually works, but I throw this out as a possibility. I mean, if you are right and Intel designed it the way you are describing, they are retarded.

[StoneRhino]

Need advice!!

Check SIG, System is stable up to 425 fsb with 32 hour dual prime95.

430 435 give me random restarts up to 2 hours with dual prime95.

440 450 give me random restarts within 15 mins with dual prime95.

Never has dual prime95 ever failed with my current settings.

Just restarts.

So what I did was flashed to 1405 from 1305 bios. Lower the multi to 8 from 10.

Run Dual Prime95 with memory 1:1 and system fails Prime95 with in the same time constrants as the 10 multi.

But system does not restart.

I played with memory timmings nothing is stable with memory over 425 fsb with the multi lowered.

What should I try next do you guys think it's the board or the memory?

[rpr]

What happened to the "review of this board with the 1405 BIOS on a major site" the moderator mentioned in a post about a week ago???

[Falkentyne]

Virtualrain:
Its not the MCH. I have the newer board with 1.85 vmch.
Nor the FSB.

It's the actual CPU initializing at 370x11 and failing to even PWRGOOD to be set at 370x9.

Now what I have NOT tried yet is setting the multiplier to 9 first, and THEN rebooting and then changing the FSB up. I'll try that later, and post back if that works. If this does work, then you are indeed right.

[The Nemesis]

What happened to the "review of this board with the 1405 BIOS on a major site" the moderator mentioned in a post about a week ago???
The review was supposed to be done @ anandtech. I'd like to see their results as well since the reviewer has both the 1.65v & 1.85v mch versions of the board. My assumption is that since the bios has yet be made offical, the review won't come until it is. Of course if the bios changes from 1405 to 1408 then the review will probably be on it. Hang in there guys, there's a bios that allows changes to tref, multiplier, etc. coming in the near future:)

[taemun]

The Nemesis: 1405 (and a few before) already can change multipliers...

t

[rpr]

I keep hearing about "tref", but have no idea what it is and how it should/can be tweaked to increase performance. Can anybody explain?

[The Nemesis]

The Nemesis: 1405 (and a few before) already can change multipliers...

t
The original 1405 bios would not change multi on the 1.85v mch boards though the 1404 would.

[taemun]

Ah... my bad then. My board is a 1.65V vMCH...

t

[tommyshango]

how do you know what version your board is? I am sending my board back RMA to asus. I might see if they can send that version to me instead.

[Kemp3]

Tommy they should send you a newer rev . They sent me a newer Rev when I RMA'd :)

[taemun]

What did you guys RMA them for?

t

[Kemp3]

My first board woudl randomly restart @ stock speeds, never while I was using it. But overnight I guess it would restart and while I was at work sometimes just shutdown. Verified it wasnt RAM , PSU or CPU so contacted ASUS and they arranged a RMA. Was surprisingly fast :) New board is newer rev and hasnt shutdown or restarted :)

[tommyshango]

my sata plug fell off or broke off from teh ez raid connecter. cheap piece of :banana::banana::banana::banana:.

[bullet2urbrain]

My first board woudl randomly restart @ stock speeds, never while I was using it. But overnight I guess it would restart and while I was at work sometimes just shutdown. Verified it wasnt RAM , PSU or CPU so contacted ASUS and they arranged a RMA. Was surprisingly fast :) New board is newer rev and hasnt shutdown or restarted :)

i have this problem too, for instance i just finished a nice 32M SPi, then bam restart after i walk away for like 50 minutes.

my board restarts alot throughout the night but i cant be without it.

[taemun]

bullet: try raising your PCIe freq by a little... I think that I had a similar problem (sata would stop responding), and it hasn't shown itself since i set PCIe freq to 105...

t

[bullet2urbrain]

FSB is @ 377... FSB has never been lower than 333...

i will try and raise the PCI-E Freq then.

[taemun]

Oops .. my bad I meant PCIe freq at the start of the post ... edited ...

t

[MacClipper]

My 3DFuzion 7300GS is supposed to swipe 256MB of system RAM to become a 512MB card, so maybe that's what's happening to you, too?...
No, the 7300GT DDR3 do not feature Turbocache.

Anyway, the missing RAM problem has been duplicated by the local Asus/Leadtek distributor service centre so I guess I just have to wait till they settle among themselves where it went wrong.

Courtesy of 11qwerty,
http://i35.photobucket.com/albums/d156/11qwerty/IMG_1357.jpg

[bullet2urbrain]

Oops .. my bad I meant PCIe freq at the start of the post ... edited ...

t


results still pending on this... PCI-E Freq @105 CPU @ 3.33 still stable, 32m and overnight SoB Stable

[hendrasystemind]

does anyone have try 4 dimm at the same time? i try it and my p5wdh wont post. can someone tell me what is wrong with my board. (bios 1405)
thx

[davexl]

does anyone have try 4 dimm at the same time? i try it and my p5wdh wont post. can someone tell me what is wrong with my board. (bios 1405)
thx

Yes, I have run 4 x 1G fine - it needs 2.3v with 4 DIMMS in to maintain the same settings that are stable on 2.1volts with 2 x 1G.

I would try 1 or 2 dimms, and get back to a stable BIOS to work out what is going on... enable timings by SPD etc etc

[hendrasystemind]

@davexl: i just try it, 2.4 v on the dimm. it just wont post.
my ram is gskill 6400HZ . both pair is the same.
should i rma this board ??

[davexl]

@davexl: i just try it, 2.4 v on the dimm. it just wont post.
my ram is gskill 6400HZ . both pair is the same.
should i rma this board ??

As I said, try just one or two dimms, and reset the BIOS to defaults - presumably you could post before you flashed it to 1405 right?

At what point did it stop working? I doubt the board is the issue... it could very well be 1405 is the problem - can you get back to 1305?

[J0d3r]

Got a p5wdh yesterday... FSB wall at 430, can spi 32m 430x9 with my 6600, but at 433x9 it restarts...

My 6600 was tested on a bad axe and could get 4120mhz 1m stable with 1.56v (watercooling). Now im stuck at 430 fsb.

I cannot run the memory 430 1:1 3-3-3 from Bios, it crashes. I Cant even run 3-4-4 from Bios, and i do it with my bad axe at 2.35v (vmod).

Flashed last oficial bios (130X) and i get nothing. I am at 1.65v mch and the board is a 1.65vmch max.

Changed the termal paste to AS Ceramique, and put a couple of fans around the mobo. No result.

Someone has any advice?

Damn... I LOVE my bad axe :(

[OneyedK]

Changed the termal paste to AS Ceramique, and put a couple of fans around the mobo. No result.:(

Dit you check the SB-sink for flatness??? Had to lap that one and put a small fan on it to go from 430 > 440.

vmch = 1.6V, couldn't pass 440 stable, 1.65V didn't help... (MCX159cu on NB)

and I couldn't get CAS3 at this speed 1:1 either... (bios 1401)

E6400 aircooled cnps9500cu

[J0d3r]

Uhmm I didnt check SB-sink at all, but Ill do, thanks you.


Seem that we are at the same point hehe, but you a little better (440). The Cas3 thing sux :(

Thanks

[Plester]

I just swapped out my Bad Axe for a P5W-DH and am starting to regret it. I can't quite get the P5W stable at 425 and even this requires 1.85 vMCH, and I HAVE to run mem at SPD or it will not boot no matter. I had the Bad Axe stone cold stable 24/7 at 437 and could run benches at 452, just had to get there with setFSB in windows. Thinking about doing the vMCH mod on the Bad Axe and putting it back in. Being able to set the 1333 strap in bios is a very nice thing too.

[davexl]

I just swapped out my Bad Axe for a P5W-DH and am starting to regret it. I can't quite get the P5W stable at 425 and even this requires 1.85 vMCH, and I HAVE to run mem at SPD or it will not boot no matter. I had the Bad Axe stone cold stable 24/7 at 437 and could run benches at 452, just had to get there with setFSB in windows. Thinking about doing the vMCH mod on the Bad Axe and putting it back in. Being able to set the 1333 strap in bios is a very nice thing too.

It might be something else like your RAM - FWIW both my P5W can do 425 at 1.65vMCH. And that is 48hr+ Orthos stable.

The SPD thing seems to be RAM related - I can boot at manual timings with the ram in my sig.

[OneyedK]

Uhmm I didnt check SB-sink at all, but Ill do

People tend to forget the SB is connected to the cpu too...
I noticed the problem when switching from 1.05V to 1.20V on the SB, things got worse... So I took off the sink and it's not remotely flat... Took me a lot of grinding and lapping to make the center touch the chip...

Might be that the ICH7 is the real reason why P5B Dx outperforms most x975-boards at the fsb-side... (pure MHz, not necessarily raw performance)

cas3 wouldn't bother me if I was capable of running 4:5... doesn't work either at this speed, but I really didn't have time to look into it, maybe I do something wrong... spd settings don't help me much with 5400UL...

BadAxe is hard to get (read expensive) in Belgium :-(

[J0d3r]

Thanks for all the info! Ill try at 1.05v, and Ill check SB. BTW I didnt removed the cover of heatsinks, Hope it helps.
P5wdh its a good mobo, with a lot of extras and 1.7vcore + 2.4vdimm. Bad axe only goes 1.60vcore (1.57 real) and 2.2vdimm.

But its more powerfull when you are talkin about pure FSB. It has less vdroop and clocks nice. I get mine from taiwan, but well... now its dead.

I hope I can finnaly get my 6600 ES at 4100mhz :(

Thanks,

[deadlock7]

I have mentioned earlier that i couldn't get 415+ with my mobo.I rmaed it and now i have a new one 68xxxx that came with 401 bios stick but it was flashed with 1305(maybe by retailer:confused: ).It has vmch 1.85 option at bios but there is no option for ecc!I flashed to 1405 but nothing again.wtf?
I tried to boot 415fsb that the previous one refuses but no luck again.I tried 420+ and it's ok..:clap: .
Another issue is with vmch.I used a multimeter and at "auto" reads 1.3v!At 1.85 option it reads 1.6v!Anyone knows smth about this?
Conclusion:1)Conroe platform has many many bugs..i'm pretty sure of this and you need time to find the way to solve them,2)where is the ecc option?,3)why does the mobo gives so low mch voltages?
I'd appreciate any answer of you guys..

[deadlock7]

Well the voltages error is because of my two multimeters' low battery.I checked it at mt friend's one and it's fine.:rolleyes:
The problem is with ecc option.There is no way to be in BIOS.I can see it at boot screen only if i put the battery off,but after reboot it disappears!It's a pity to rma because there is no ecc option.I'm unlucky..my ram is ecc fatties!:(

[Pecka]

I have an Issue with the q-fan on this board. I have connected two fans to the powerfan connectors and set them to silent operation in BIOS. The problem is they only stay quiet for a while.. as soon as I for example run Prime95 to load the system the fans start to run at full speed as they should, but they won't spin down when the temperatures decreases. I have to shut down the system for a while and then start again and the fans will run silent again. Anyone know a solution to this problem? I have the stock Intel cooler on the CPU.

[miochza]

Say ignore in the BIOS and let Speedfan automatically control it?

[iamhives]

Do you think the following storage tower with SATA port multiplier (model ST5X1PM) with work with the esata port of the P5W DH Deluxe?

Storage tower (http://www.addonics.com/products/raid_system/ast4.asp)


The Jmicron site says the JMB363 "Supports Port Multiplier with Command-based Switching on SATA II port" - see the spec at this link:


JMicron JMB363 spec (http://www.jmicron.com/JMB363.html)

Seems like a pretty cost effective external storage solution if it will work.

[Trellame]

Bios 1407:

http://mapesh-online.de/html/hwluxxspecials.html
or
http://taskforce.no-ip.info/pics/ASUS_P5W-DH/P5W-DH1407.zip

[Laughy]

Bios 1407:

http://mapesh-online.de/html/hwluxxspecials.html
or
http://taskforce.no-ip.info/pics/ASUS_P5W-DH/P5W-DH1407.zip

Hmm...any try this out and notice any changes?? Wish they had a change log in the zip file... :stick:

[kevsta112003]

My problem.

Bios doesnt recognize the timings of my RAM which are 4-4-4-12. It thinks it's 5-6-6-18, and I cant change them because it wont post when I set "Configure DRAM timings by speed" to Disabled.

[catscit]

My problem.

Bios doesnt recognize the timings of my RAM which are 4-4-4-12. It thinks it's 5-6-6-18, and I cant change them because it wont post when I set "Configure DRAM timings by speed" to Disabled.

you have increased vdimm?

[lawrywild]

you have increased vdimm?

it's nothing to do with vdimm, it's the board setting too low tRFC

[suio]

My problem.

Bios doesnt recognize the timings of my RAM which are 4-4-4-12. It thinks it's 5-6-6-18, and I cant change them because it wont post when I set "Configure DRAM timings by speed" to Disabled.

I have the same problem with Corsair memory.

[kevsta112003]

My memory is G.Skill DDR2 800 4-4-4-12, not the HZ kind though, just the regular.

Glad to know at least its not only me with the problem. Someone should add it to the list! :)

Oh and yes I tried increasing the vdimm, however I shouldnt have to, because its running at 375Mhz with the slower 5-6-6-18 timings. So its actually being underclocked.

Are you sure its the board setting the tRFC too low? And do you know of a possible fix?

[miochza]

For the record, BIOS 1405 is really buggy. It reports an "overclocking/overvolting failure" everytime I reboot, even though I had been using that stable clock frequency and voltage for days... I'm too lazy to roll back.

[Falkentyne]

1405 or 1407 ?
1407 is the newest beta one...why are you using 1405?

[turbo]

I had problems using a bluetooth mouse with the 1405 bios, pointer jumped around and connection problems. 1407 seems ok.

[Methylphenidate]

1407 does not overclock as well as 1405 for me

[Prelim]

FSB Wall at around 360-370 fsb - Enable SPD timings option in Chipset section of bios.

is Asus capable of resolving this problem by newer BIOS?

i readed somewhere that its a memory issue that occur only with some kind of dimms. Is that true?

[lawrywild]

is Asus capable of resolving this problem by newer BIOS?

i readed somewhere that its a memory issue that occur only with some kind of dimms. Is that true?

The problem is tRFC, the board sets too low tRFC (for either the ram to handle, OR the board, we don't know really). SO yes, if ASUS add tRFC into memory timings options, then it's fixable..

[arrarro]

BIOS 1407 has now become OFFICIAL and FINAL

ftp://ftp.asus.com.tw/pub/ASUS/mb/socket775/P5W%20DH%20Deluxe/P5WDH1407.zip

cheers!:banana:

[taemun]

nice find arrarro! :D

hmm .. the beta is the same as the file, the CRC32 hashes from the zip files (one from asus, one from deluxx) match. Both 0EC1F93D. Nothing new here folks :(

t

[kayomani]

The randoms 256 degree CPU bug is fixed for me in 1407.

At what temp should I begin worrying about my mb? Its pretty much always at 42 degrees.

[famich]

Hello guys ,

I cannot get the eSATA onboard port to work. When I plug in eSATA to SATA cable for the storage Seagate HDD , WIN XP do not recognize it :confused:

However, with the USB 2.0 cable the HDD works flawlessly.
Apart from that , everything s OK.

THX.:)

[kevsta112003]

When I put my intake fan on high, my mobo reaches 34c. When my intake fan is on medium, it reaches around 43c. And with my intake set on low, it reaches 50c. I keep it on medium. I have two x1900xt's which run at 85c in games. The temp sensor is around the southbridge so it is kind of close to my videocards, which is why it can get a bit warm sometimes. But yeah, im curious as to what you guys are getting for temps on the mobo?

[Plester]

did anyone suggest a heatsink for the clock gen on this board, because it gets burning hot when overclocked.

[Timmaah]

im getting highs of 49, and i dont like it.. but apparently they put the sensor at the wrong place so anything up to 55°C should be fine, but dont quote me...

I also dont know which cpu sensor to trust, the ones from coretemp or asus probe. Seems weird that the cpu is as low as room temperature but all other components are ~40°C

At the moment...
Asus Probe:
Cpu: 26°C
Mobo: 42°C

CoreTemp:
Core1: 40°C
Core2: 40°C

Sisoft Sandra:
Motherboard: 42.0°C
Cpu: 26.0°C
Power / Aux Temperature: 37.5°C <--- is that case temp? and why would it be higher than cpu, doesnt make sense

AtiTool:
GPU: 44°C

setup is; e6600@3.0ghz (stock voltages), g-skill 2gb "hz" ram at 1ghz @ 2.1v and a x1900xt at stock

so basically, which temperatures should i trust? the temperatures are generally low bar the motherboard one but apparently the temps with this board go pretty high

[NghtShd]

When I put my intake fan on high, my mobo reaches 34c. When my intake fan is on medium, it reaches around 43c. And with my intake set on low, it reaches 50c. I keep it on medium. I have two x1900xt's which run at 85c in games. The temp sensor is around the southbridge so it is kind of close to my videocards, which is why it can get a bit warm sometimes. But yeah, im curious as to what you guys are getting for temps on the mobo?


I was getting 45c until I made a couple of modifications.

I pulled the SB sink and found that the thermal gunk on the southbridge was only touching the left and right ends of the chip. The middle swath was completely clear of any of the stuff. I cleaned the gunk off of the chip sink and I lapped the southbridge heatsink which was pretty un-flat. I also removed the shroud then reinstalled using AS5.

I cleaned the same glue-like hard thermal gunk from the northbridge chip and heatsink, but didn't lap because it would have been difficult with the heatpipe hanging off and I didn't want to risk damaging the pad that contacts the voltage regulators. I removed the shroud, applied AS5 and reinstalled.

After that, my MB temp had dropped from 45c to 37c.

The northbridge and southbridge heatsinks both felt really hot---I couldn't even hold my finger on the northbridge for more than a second or so--,so I decided to put a fan on each. That dropped my MB temp to 32c and now I could probably hold my toungue on the northbridge sink all day if I were so inclined (and gifted).

[Prelim]

The problem is tRFC, the board sets too low tRFC (for either the ram to handle, OR the board, we don't know really). SO yes, if ASUS add tRFC into memory timings options, then it's fixable..

its the most important problem found since the mobo has been released. I dont know why didn´t they fixed it yet...strange no??

i have Teamgroup pc5300 3-4-4 and have this issue, i was thinking to sell TG and buy Gskill HZ to fix the SPD bug...now i dont know what to do :( :(

[kevsta112003]

how long does it take you guys to post also?

it takes me a good 10-15 seconds.

[sniggle]

how long does it take you guys to post also?

it takes me a good 10-15 seconds.

Try disabling all Q-Fan as shown in the first post. If that doesn't work, set the temp sensors to Ignored in the hardware monitoring section.

[zoob]

Hmm..

I updated my BIOS to 1407 (from 0801) and it looks like 400 FSB is now stable.
Been blending my E6700 at 400 x 9 @ 1.52 vCore (BIOS) for 1 hr 22 min now, where before it'd crap out randomly between 0 to 35 minutes.

I approve! :)

[Khaotic]

The 1407 also lets me adjust my CPU Ratio now on my L624A E6600. I second the approval.

I've been slowly working with the CPU. My first wall seemed to be at 356FSB with 1.5 volts. I just bumped up to 376 @ 1.52ish and its holding its own crunching on a small Orthos for 3387. But with my luck it will BSOD right after i hit the SUBMIT :eek:

[zoob]

Ahh man, the CPU ratio adjustment exploits EIST. So much for that idea. :|

Sitting with a comfortable lapped E6700 @ 320 x 10, Auto everything, GSkill ZX @ DDR-800 4-4-4-12, 2.1V. C1E and EIST enabled for power savings.

Big Typhoon also lapped, P180 with two Yate Loons, all at 640-670 RPM for some very quiet computing...

The exhaust fan from the building across the street is louder than my computer, even with the window closed! :)

[DigitalTrance]

Hey guys. My sytsem will not go into stanby. Sometimes it locks going into standby, and sometimes it boots as if the system was off if it does manage to go into standby. This occurs with 1305 and 1407 Bios. Any suggestions?

[cmthomson]

I installed 1407 yesterday, and this morning Probe 2 is still (once again) showing 128/256 as the CPU temperature. Everest works fine.

[Dani]

470MHz ORTHOS stable with my P5WDH "1.65vMCH version", copper cooler on NB :)

[jagt]

470MHz ORTHOS stable with my P5WDH "1.65vMCH version", copper cooler on NB :)
Which bios are you using?

[revenant]

ok - I am going to have to give the 1407 a try now.. seems like it may help with my minor instability issues at 400x9...

[chron]

my problem lately is that the hardware monitor has been acting REAL slow... Like PC Probe takes forever to load and messing with the settings in it has a very long response time...

Any ideas? All Q-Fan controlls are disabled.

[boba23]

Hey Folks,

i got my ASUS P5WDH for about 3 weeks and am still fighting to get it absolutely stable. Right now i am almost running out of ideas what my problem ist.
First of all, i am able to get my E6400 Dual Prime stable (been running for 11-12 hours, no errors) at 3,2ghz, 8x400mhz with 1.3850 Vcore in Bios and about 1.36 i'd say real, as far as CPU-Z shows it under full load. Anyways, Vcore doesn't seem to be the problem, since more of it, let's say 1.415 doesn't help either.
Anyways, i am running to Samsung S-ATA's in Raid0 on the Intel ICH7 controller. PCI locked to 33mhz. I use to compile my own Firefox and Thunderbird builds quite frequently, no problems before on my A64@2.6ghz, same raid0 setup btw on a K8N Neo2.
What happens now, is that my compiler runs bail out about every 2nd or 3rd run with errors on corrupted libraries or internal compiler errors.
Now i read in the Anandtech Forums, that some people have trouble with data corruption and overclocking on the P5W-DH.
Anyone else experiencing this? Or might it still be only the CPU at it's limits and not able to do 3.2ghz?
On the other side, it seems absolutely Prime stable, which makes me wonder. Since on my A64 i had really trouble getting Prime stability, but on the other side, even with Prime bailing out, i had no problems what so ever compiling my stuff.
What is more, max. cpu load during building is 50%, since the compiler and make do not use both cores simultaniously. This again makes be think that it isn't a cpu issue but maybe more a motherboard or raid setup related thing.
Next thing i'll try is to do the compiler tests on a single, old IBM DTLA drive that i still have left here.

Thanks for any ideas or suggestions on my problems ....

boba

[pee4]

Lock PCI Express bus at 100

Set MCH voltage to 1.55v

[suio]

Hey Folks,

i got my ASUS P5WDH for about 3 weeks and am still fighting to get it absolutely stable. Right now i am almost running out of ideas what my problem ist.
First of all, i am able to get my E6400 Dual Prime stable (been running for 11-12 hours, no errors) at 3,2ghz, 8x400mhz with 1.3850 Vcore in Bios and about 1.36 i'd say real, as far as CPU-Z shows it under full load. Anyways, Vcore doesn't seem to be the problem, since more of it, let's say 1.415 doesn't help either.
Anyways, i am running to Samsung S-ATA's in Raid0 on the Intel ICH7 controller. PCI locked to 33mhz. I use to compile my own Firefox and Thunderbird builds quite frequently, no problems before on my A64@2.6ghz, same raid0 setup btw on a K8N Neo2.
What happens now, is that my compiler runs bail out about every 2nd or 3rd run with errors on corrupted libraries or internal compiler errors.
Now i read in the Anandtech Forums, that some people have trouble with data corruption and overclocking on the P5W-DH.
Anyone else experiencing this? Or might it still be only the CPU at it's limits and not able to do 3.2ghz?
On the other side, it seems absolutely Prime stable, which makes me wonder. Since on my A64 i had really trouble getting Prime stability, but on the other side, even with Prime bailing out, i had no problems what so ever compiling my stuff.
What is more, max. cpu load during building is 50%, since the compiler and make do not use both cores simultaniously. This again makes be think that it isn't a cpu issue but maybe more a motherboard or raid setup related thing.
Next thing i'll try is to do the compiler tests on a single, old IBM DTLA drive that i still have left here.

Thanks for any ideas or suggestions on my problems ....

boba

Either the memory or the power supply is the issue in your rig. Did you test your memory after a reboot and before Windows loads?

[revenant]

maybe vdimm (memory voltage) or mch (NB chip/chipset) voltage? most ddr2-800 modules like about 2.1v and at 400 on the fsb you should be at 1.55v or higher on the mch.. maybe more like 1.65v if you put a fan on the NB sink or have good air flow in your case. that NB chip gets hot! :) A lot of people here have pulled both the NB and SB sinks cleaned the factory thermal paste off and put AS5 on those.. some have even lapped the SB sink because it's been VERY bad... not even making contact with chip top in many cases.. that could be the root of your instability.. possibly.. could be a lot of things though.. but if you're still on auto for chipset voltages and/or om 1.8v for the memory I would start there. good luck!

[boba23]

Thanks for the suggestions so far guys, but i also played around with voltages as well. had MCH on 1.55, VDIMM up to 2.25 (though i ran memtest86 to check the dimms on that speed for hours). but right now i am on auto on all the voltages cept for cpu and dimm, but gonna try upping some again.
what i didn't do so far is removing the nb/sb sinks ... that might be my next step. what is more i wanted to get a new power supply anyways going 500W+ . though i am not too sure about the enermax being the problem. it's supposed to be a very good and stable psu, and the voltage readings don't look bad at all, even under full load.
still, what do you guys say, that this one task is basically the only application i get those errors? i can even play HL2 EP1 for hours and hours without any issues at 3.2ghz.
you sure, that this couldn't be in any way raid/intel driver related? anyone of you running raid0 with a cpu and the mobo at 400mhz fsb?

boba

[revenant]

I kind of doubt it's IO related.. hrm... but stranger things have happened.. I know there was a rumor about nforce4 nvraid IO file corruption issues for a while.. only on certain motherboards.. like the msi neo4 plat sli.. anyways..

I would leave your mch at 1.55v at least.. and hard set your pci-e frequency at 100mhz.. those two are known to for sure improve stability at 400mhz fsb speeds.. maybe try some -full- 24 hour dual prime validation tests would be prudent to make sure the cpu and mem are happy at current vcore/vdimm.. just rule those out for good..

[mefromfl]

will these rams work on this mobo

http://shop3.outpost.com/%7BOss3yULW2EOgoPjE2DBPrw**.node3%7D/product/4158473;jsessionid=Oss3yULW2EOgoPjE2DBPrw**.node3? site=sr:SEARCH:MAIN_RSLT_PG

[X7007]

what are PEG Root Control and PEG Buffer Length

PEG Root Control = Auto , Disabled,Enabled ?
PEG Buffer Length = Auto, Slow,Normal ?
Link Latency =Auto Long,Slow ?
Slot Power = Auto ? or one of the rest, light-er,normal,heavy-ier?

[taxman]

My final OC:

e6700 @3.5 Ghz 1.55v (350 FSB)
Corsair Value Ram: 1:1 4-4-4-12 @ 700Mhz 2.1v
Can't do 3.6 even with 1.6vcore...

My Core Temps are ~50C idle and ~70C load
PC Probe II says my CPU is 27C idle 35C load

my e6700 is a week 27.... I think I got a crappy one since I expected @ ~3.7-3.9

[NghtShd]

so what about the shut down after saving problem while overclocking, is this still in the p5w-dh bios :rolleyes: ? My abit does not do that anymore after bios 1.2 :)

If you mean the cold booting instead of warm booting thing, it still does that with 1407.

[OneyedK]

My final OC:

e6700 @3.5 Ghz 1.55v (350 FSB)
Corsair Value Ram: 1:1 4-4-4-12 @ 700Mhz 2.1v
Can't do 3.6 even with 1.6vcore...

My Core Temps are ~50C idle and ~70C load
PC Probe II says my CPU is 27C idle 35C load

my e6700 is a week 27.... I think I got a crappy one since I expected @ ~3.7-3.9

just my 2 cents, heat is your problem I think... 70°C loaded is too much to OC higher... what cooling???

[catscit]

just my 2 cents, heat is your problem I think... 70°C loaded is too much to OC higher... what cooling???

heat indeed, my e6700 is 36 idle and 51 load.
But most e6700 are bad overclockers.
Mine wil never go past 380x10, stable at 360x10, i can lower the mp but then fsb increase isn't much, so cpu speed is still lower.

[Falkentyne]

That's coretemp speed.
Tends to be pretty normal for people to average 65-70C load even with high end air cooling. I've found that the Intel TAT reports a bit lower temps than coretemp. Maybe about 4C at load, *however* identical temps at idle (huh?).

Happens even with lapped heatsinks.

I think the only way to reduce temps like that is to lap the heatspreader, but you can say byebye to your warranty. (and most people buy retail chips for the 3 year warranty, otherwise people would just buy OEM if they didn't care about that).

[Mykou]

so what about the shut down after saving problem while overclocking, is this still in the p5w-dh bios :rolleyes: ? My abit does not do that anymore after bios 1.2 :)
Hi

what is your 24/7 set up ?

[jozzy]

The Nemesis: 1405 (and a few before) already can change multipliers...

t

how bout the 1407 ??

[mefromfl]

will these rams work on this mobo

http://shop3.outpost.com/%7BOss3yULW2EOgoPjE2DBPrw**.node3%7D/product/4158473;jsessionid=Oss3yULW2EOgoPjE2DBPrw**.node3? site=sr:SEARCH:MAIN_RSLT_PG


bump

[taxman]

just my 2 cents, heat is your problem I think... 70°C loaded is too much to OC higher... what cooling???

Yeah I know that heat is what's holding me back. I'm using a lapped Scythe Infinity and a lapped IHS... up to 2000 Grit and a razor blade amount of AS5.

Its definitely my shizzy chip :(
too much voltage early on. Fails Prime 95 @ 1.5x (instead of 1.55) after 1 hr.

Damn I wish I bought from newegg instead since they have more traffic -- in other words, I might have received a week 28 instead. (?)

[Dani]

Which bios are you using?
1407, it's for my setup most stable so far (1.65MCH version)

[Falkentyne]

Yeah I know that heat is what's holding me back. I'm using a lapped Scythe Infinity and a lapped IHS... up to 2000 Grit and a razor blade amount of AS5.

Its definitely my shizzy chip :(
too much voltage early on. Fails Prime 95 @ 1.5x (instead of 1.55) after 1 hr.

Damn I wish I bought from newegg instead since they have more traffic -- in other words, I might have received a week 28 instead. (?)

um.....
don't use a razor blade amount of AS5 ! No wonder you have such high temps! Air bubbles galore ?

go to the as webpage and read carefully on how to apply it, and then follow those instructions carefully.

[thudo]

My specs..

o CPU Used: Conroe E6700 Retail B2
o Motherboard: Asus P5W DH (1301 Bios)
o Ram: OCZ DDR2 PC2-8000 Platinum Extreme Edition Dual Channel 1000MHz DDR2
Recommended: CL 4-5-4-15 2.2 Volts (use a VDIMM of 2.25V ± 5%)
o Power Supply: OCZ GamerXtreme 700W
o Video Card: XFX 7600gt @ 620/830

CPU temp @ 3.66Ghz (Idle 35-36c, Load: 41-45c.. MegaLoad: 51-55c)
Cooler: Scythe Mine + Arctic Silver 5 Thermal Paste
Motherboard temp @ 3.66Ghz (Idle 33c, Load ~35c)
Video temp: (Idle 41c, Load 47c)

My Current Settings:
AI Overclocking: Manual
CPU Frequency: 366
DRAM Frequency: DDR2-915Mhz
Preformance Mode: Auto
PCI Express Frequency: 100
PCI Clock Synchonization Mode: 33.33Mhz
Memory Voltage: 2.30V
CPU VCore Voltage: 1.5250V
FSB Termination Voltage: 1.50V
MCH Chipset Voltage: 1.65V
ICH Chipset Voltage: 1.20V

MicroCode Updation: Enabled
Max CPUID Value Limit: Disabled
Execute Disable Function: Disabled
Enhacned C1 Control: Disabled
CPU Internal Thermal Control: Auto
Virtualization Technology: Disabled
Intet SpeedStep Tech: Disabled

My Current Settings:
Configure Dram Timing by SPD: Disabled
Dram Cas# Latency: 4
Dram Ras# to Cas# Delay: 4
Dram Ras# PreCharge: 4
Dram Ras# Activate to Precharge: 12
Dram Write Recovery Time: 4
DRAM ECC Mode: Disabled
HyperPath 3: Disabled
DRAM Throttling Threshold: Disabled

I run various tests with a PC RTS game (Dawn of War) I am testing and normally the game would die during the 1st round OR during the 2nd. Now it never dies with the above while keeping my original specs I want. Orthos still dies after 2mins but other benchmark apps work fine.

Edit.. have downclocked the video card to stock.. so I lose about 1-2 fps.. shesshh. Better to have system stability then the damn thing dropping to desktop every 15-30mins.

Still.. VERY happy with the specs above.. been running for now a month doing various tasks and its solid. Still.. always watching and waiting.. :rolleyes:

[Timmaah]

just thought id mention that in the latest 1407 bios under advanced -> chipset, you can now chose whether youre using vista for the hd-audio encoder :/

not sure what it does just thought id mention it.

[thudo]

1407 doesn't seem worth updating too.. Asus still has not added the key overclocking tool which the P5B has which unlocks all Conroes. I'm running 1401 and its fine for now. COME ON ASUS! Get with the P5B = P5W DH program and synch the bioses. P5W DH is your flagship. MAKE IT SO! ;)

[phile]

will these rams work on this mobo

http://shop3.outpost.com/%7BOss3yULW2EOgoPjE2DBPrw**.node3%7D/product/4158473;jsessionid=Oss3yULW2EOgoPjE2DBPrw**.node3? site=sr:SEARCH:MAIN_RSLT_PG

Yes, absolutely. Though, you may want to go for a 2GB kit if you're a gamer.

-phil

[boba23]

hey,

me again ... getting a step closer to my solution. right now i am doing my tests on a stoneage IBM DTLA 46gb pata drive, which is connected to the 3rd sata port of the intel controller via a pata-sata converter. and guess what? i am doing my 5th build run in a row now and had no errors so far. running 8x400mhz @ 1.3875 Vcore in Bios. on my samsung raid0 array i would have had 2 crashes while building for sure now.
next thing i am gonna do is check the samsungs with a dos tool and also change their sata cabling.
still no one here who might be running raid0 on a heavily overclocked system to confirm or deny my raid0 problems?

thanks

boba

[boba23]

1407 doesn't seem worth updating too.. Asus still has not added the key overclocking tool which the P5B has which unlocks all Conroes. I'm running 1401 and its fine for now. COME ON ASUS! Get with the P5B = P5W DH program and synch the bioses. P5W DH is your flagship. MAKE IT SO! ;)

unlock all conroes? does this also include e6400 allendale? or real "conroes" only?

boba

[Ic3man]

1407 doesn't seem worth updating too.. Asus still has not added the key overclocking tool which the P5B has which unlocks all Conroes. I'm running 1401 and its fine for now. COME ON ASUS! Get with the P5B = P5W DH program and synch the bioses. P5W DH is your flagship. MAKE IT SO! ;)

The 1407 bios DOES unlock the multi but only downward, only the x6800 and a few ES's are truly unlockable (up & down).

[lawrywild]

hey,

me again ... getting a step closer to my solution. right now i am doing my tests on a stoneage IBM DTLA 46gb pata drive, which is connected to the 3rd sata port of the intel controller via a pata-sata converter. and guess what? i am doing my 5th build run in a row now and had no errors so far. running 8x400mhz @ 1.3875 Vcore in Bios. on my samsung raid0 array i would have had 2 crashes while building for sure now.
next thing i am gonna do is check the samsungs with a dos tool and also change their sata cabling.
still no one here who might be running raid0 on a heavily overclocked system to confirm or deny my raid0 problems?

thanks

boba

I have 2 seagate 200gbs in raid 0 and a maxtor 16mb on sata4

3.5ghz, 1.6v no raid problems

[miochza]

Wow 1.6V on air lawrywild? You are much braver than I.

[Timmaah]

The 1407 bios DOES unlock the multi but only downward, only the x6800 and a few ES's are truly unlockable (up & down).

so the e6600 cpu multiplier can only be clocked downwards?

[Monroe]

Has anyone read this:

http://support.asus.com/faq/faq.aspx?no=D7F52826-545A-E665-58CE-CB2AC36794D9&SLanguage=en-us

quote: "It should be normal if the motherboard temperature is within the range of 40~65 degrees."

I run my 6600 at 3.2Ghz (9x356) stable on stock(!) voltages.
Idle temperatures are:
Motherboard 40 °C (104 °F)
CPU 31 °C (88 °F)
CPU #1 / Core #1 40 °C (104 °F)
CPU #1 / Core #2 39 °C (102 °F)
Loaded tems are 10°C higher everywhere.

[kevsta112003]

Does anyone know on Speedfan, what temp3 is?

Temp 1 is the mobo, i am positive on this
Temp 2 for me is the CPU, i know this also
temp 3 I have no idea. Its hanging around 42c now, but maybe a month ago it used to be like 80-120c on idle and would fluctuate randomly. But now all of a sudden its like 40-60c. And sometimes it will increase when CPU increases in temps, but it doesnt always.

HD0- HD temp

[Ic3man]

so the e6600 cpu multiplier can only be clocked downwards?

Correct

[Pjoeloe]

Hi
what is your 24/7 set up ?
I answered your PM,

If you mean the cold booting instead of warm booting thing, it still does that with 1407.
the problem I mean is that if you power on your pc from a cold boot, the fans and hd`s are going on and off like 3 times, this is killing your hardware if you do this 365 times a year :rolleyes:

[Mykou]

I answered your PM,


thanks m8 , just tell me what ram facturer is yours ?

[mefromfl]

Yes, absolutely. Though, you may want to go for a 2GB kit if you're a gamer.

-phil


I have 1 gb of these rams already, if i upgrade to 2gb do i need faster speeds? or is this good enough?

[mefromfl]

Is there a heat issue witht his mobo? What other duo 2 mobo is very cool?

[X7007]

what are PEG Root Control and PEG Buffer Length

PEG Root Control = Auto , Disabled,Enabled ?
PEG Buffer Length = Auto, Slow,Normal ?
Link Latency =Auto Long,Slow ?
Slot Power = Auto ? or one of the rest, light-er,normal,heavy-ier?

and do u have like Shutdown when u restart in windows and then back on when u go above 320FSB ?
^
does it ok ? the system like going off and then back on after 2sec alone.

[JoeBar]

Quick question. Do 1.65v MCH boards have a different revision?

[taxman]

um.....
don't use a razor blade amount of AS5 ! No wonder you have such high temps! Air bubbles galore ?

go to the as web page and read carefully on how to apply it, and then follow those instructions carefully.

Well, since my IHS and infinity are so flat, the more I add the less contact between the 2 surfaces... kinda like filling in a flatland as opposed to a mountain range.

Any one think my logic is flawed?

To rephrase: "the more you lap, the less AS5 you should use"

---------------------------------

e6700 [Lapped] @ 3.5 Ghz 1.55v [Scythe Inf. Lapped]
Asus P5W DH 1407
x1900xtx @ 675/1600 [Accelero X2]
x1900 CF @ 675/1600 [Accelero X2]
2 GB Corsair VS 4-4-4-12 @ 700Mhz 2.1v
74GB WD Raptor
250 WD Caviar
BenQ 1655
BenQ 1655
5x 120mm Scythe 1600RPM Fans w/Controller
Enermax Noisetaker 600w
NEC 20wmgX2

[Falkentyne]

Well, since my IHS and infinity are so flat, the more I add the less contact between the 2 surfaces... kinda like filling in a flatland as opposed to a mountain range.

Any one think my logic is flawed?

To rephrase: "the more you lap, the less AS5 you should use"

---------------------------------

e6700 [Lapped] @ 3.5 Ghz 1.55v [Scythe Inf. Lapped]
Asus P5W DH 1407
x1900xtx @ 675/1600 [Accelero X2]
x1900 CF @ 675/1600 [Accelero X2]
2 GB Corsair VS 4-4-4-12 @ 700Mhz 2.1v
74GB WD Raptor
250 WD Caviar
BenQ 1655
BenQ 1655
5x 120mm Scythe 1600RPM Fans w/Controller
Enermax Noisetaker 600w
NEC 20wmgX2

Your logic has nothing to do with how you apply the AS5.
It's true that the more you lap the less you need, but you still need to apply the dab. Just a slightly smaller dab instead. It still has to spread out so the core part that radiates the heat directly to the Heatsink is covered.

Trust me.
please.
DO NOT spread AS5 on a IHS !
It does NOT matter how flat it is !!!
Air bubbles *WILL* get trapped when you apply a surface to another one covered with AS5.

By applying the "rice dab" method, in the middle, it starts in the middle then gets squeezed out by pressure evenly. And since the CPU die is in the middle of the IHS......well... the part that needs the contact gets it----without air bubbles.

The outer perimeter of the IHS is inconsequential. Virtually no heat transfer occurs there.

Now, applying AS (ceramique is safer) to the southbridge is a different story, as the chip covers the entire surface; there the 'dab' method won't work.

[phile]

I have 1 gb of these rams already, if i upgrade to 2gb do i need faster speeds? or is this good enough?

PC-5300/PC-5400 (DDR2-667) is fast enough unless you plan to significantly overclock.

-phil

[phogger]

Wow 1.6V on air lawrywild? You are much braver than I.

I too put vCore to 1.60v, following the lead - not as brave.

Core Temp Beta temps are 48/47 C right now on air.

That's been the only way to get me past 3.2GHz in BIOS.

[syne_24]

Quick question. Do 1.65v MCH boards have a different revision?

I assume you meant 1.85vmch? All of the older boards should have 1.65vmch but only those with serial 68MOAG will be 1.85vmch. Also below the ram slot it should be atleast 0628

[Falkentyne]

I assume you meant 1.85vmch? All of the older boards should have 1.65vmch but only those with serial 68MOAG will be 1.85vmch. Also below the ram slot it should be atleast 0628

Amen.
My board barely made the cut. But it did...
0628, 68moag, i think the serial was like 113000 or 117000, borderline....

[walaustin]

i have an E6600 L626 running at 3510Mhz vcore 1.5V on p5wdh dlx 1.85vmch version :D

http://img476.imageshack.us/img476/9449/390gods2.jpg

:banana:

[chron]

Do the 1.85v mch versions run hot?

[The Nemesis]

Do the 1.85v mch versions run hot?
my 1.85v mch board ran no hotter than my 1.65v mch board even when overclocking. My 1.65v version could pi @ 420-422 FSB and 3DMark up to 412FSB. My 1.85v could pi up to 440 FSB with stock "9" multiplier & 450 FSB with "8" multiplier & 3DMark06 up to 423 FSB.

[davexl]

my 1.85v mch board ran no hotter than my 1.65v mch board even when overclocking. My 1.65v version could pi @ 420-422 FSB and 3DMark up to 412FSB. My 1.85v could pi up to 440 FSB with stock "9" multiplier & 450 FSB with "8" multiplier & 3DMark06 up to 423 FSB.

Interesting, my 1.85v board runs (or reports at least) 4-5 degrees higher than my 1.65v - at same settings, same CPU, same cooler and fan etc.

[Mykou]

I assume you meant 1.85vmch? All of the older boards should have 1.65vmch but only those with serial 68MOAG will be 1.85vmch. Also below the ram slot it should be atleast 0628
Still VDroop I presume on new rev ?

[chron]

vdroop is there on purpose. i'd assume it will always be there unless u mod your board

[Timmaah]

what is vdroop?

[hot_fifty]

it is when your cpu is loading the vcore will droop

[hadron]

hi guys,
well, i have like cpu with bad mojo or i'm just doing something terribly wrong. i got e6300 on p5b dlx with 2x1gb of goodram, 7600gt and a wireless card. the thing is i just can't get any serrious speed out of this thing now that i got oc. the last stable speed is at 330fsb, i tried like everything. got the bios stuff like in that thread, i'm runing 1:1 with ram which i think is not the problem since i had them runing stable at 500ddr, put the vcore to 1.4, cpuz reads like 1.352, bumped up the other voltages too, temperatures are nice since i got big typhoon cooling the thing, disconected everything i don't need in bios which is 0507, i could upgrade it but thisone should do it too. i can't get even 1m of pi at >335 fsb, the last idea i had was just to give it more juice but then that is quite high so i first decided to ask here what to do if anyone can do magic or i'll just have to settle for puny 2310mhz. thanks for all the help.

[Falkentyne]

You're using 0507 ?? Does that even -support- C2d ? I thought you had to have 701 -minimum-.

Flash it to 1305. That version is trouble free for everyone, just use the ezflash in the bios.

And the only reason you're limited to 333 fsb is because I bet you didn't disable hyperpath 3. Disable that.

[Mykou]

I love changelogs related to P5W .. improving each passing day

P5w DH

1407 2006/10/02 Mise à jour

Support new CPUs. Please refer to our website at: http://support.asus.com/cpusupport/cpusupport.aspx
Fix EN7900GTX PCIE VGA Card compatibility issue when PEG Link Mode is set at [Faster]

the last changelog for P5B ..

P5B

Version 0711 2006/10/04 Mise à jour

Description P5B Deluxe Release BIOS version 0711

1. Fix DRAM frequency incorrectly display problem under auto, standard and AI N.O.S mode
2. Add check ATI CrossFire VGA card routine
3. Add ASUS Cross Graphics Impeller (ASUS C.G.I) feature
4. Reduce system full reset of times
5. Patch some VGA card can''t found when NB Vcore value has been set too high
6. Fix DDR frequency 889 unstable by SPD
7. Enable support for NVIDIA 7950GX2 cards under recovery mode
8. Fix CPU frequency not correct when use ES CPU and lock free fuction enable
9. Fix memory display message not correct under multi-language
10. Add new CPU Ucode to support Kentsfield B3 stepping CPU
11. Fix chassis fan 2 speed not correct when fan speed too low
12. Fix ATI CrossFire can''t work when using with 800FSB Prescott CPU
13. Enhance Chassis Q-Fan sensitivity
14. Improve memory compatibility

P5W DH >the only and same description line " support new cpu"
P5B >14 lines ...

Awesome , asus keep going on !:clap:

seems P5W doesn't have so many issues for you .. :slapass: :slapass:

May give a try with abit AW9 ( doesn't like the only pci too close to 2nd PCI-e ) or the RD600 or the next DFI nForce 680i ...., please come quickly ...:p: