ASUS P5B-Deluxe; Problems & Fixes

[chuckbam]

I use the Marvel Yukon controller........so that I can take the stress off the North-Bridge (a.k.a. PCI-Lan controller), or is it south? Can't remember but I use the Yukon.

I am using the PCI one too. But for a kind of dumb reason. I do not like that the other is called Generic.

Nasgul

Thanks a bunch 2x

[alitayyab]

I have had this board (Asus P5b-Deluxe Rev 1.03G) for about 2 years. the problem i am going to describe has been there for about 8 months.


At times the system would just stop responding at all. that means that nothing short of turning the psu switch to off works. pressing the 'reset' button does nothing, neither does the 'power' button. no blue screen of death, nothing. the machine just stops responding to everything.

turning off the power, turning it on usually gets the system up and running and the chances of the samething happening are extremely varied. at times it boots up okay and the problem doesnt come back for what seems like an age, other times it would do the same thing again immediately.

The event viewer shows up nothing (no errors what so ever).

Things that i've noticed:

1. There is no guaranteed way to reproduce the error, however one way that occasionally does this is to change any bit of hardware in my system (be it a HDD, a DVD drive etc). specifically any hardware that need a wire to connect to the mobo. most commonly happens when using a Sata cable.

2. Everytime this happens, the system clock sorta goes to sleep. when the computer starts to run normally, the clock picks up EXACTLY where it left off when the problem began


Things that i've done:

0. Bios updates. no difference

1. Checked Power supply: using a volt meter. all voltages are pretty okay. checked psu in another rig (p5k-vm + 8800gtx) works great

2. Checked Processor in another board (again a p5k-vm) no problems

3. Chekced RAM (memcheck) no errors. checked ram in another system, no problems

4. Undo my oc settings. the problem still strikes if i run my system at default

5. checked gfx cards (4870, 8800gtx). get similar issues with both cards.

6. checked multiple hdds in the system (greatest stability with IDE HDD)


The problem is that if i am to RMA this board i have to write in exactly what goes wrong and how to reproduce it (that is the way things go here ) . as this error seems to just come on its own, i have no sure way of reproducing it.

I checked asus's forums and came up with this:

http://vip.asus.com/forum/view.aspx?...Language=en-us

this describes the problem EXACTLY. but the solution doesnt work for me atleast. i've tried 4 different sets and i get problems with all of em. the last set i tried i pulled out from another system where they were working fine.


I really am running out of ideas as what to do. the simplest to me is to get another board and try my luck with an RMA (takes upto 3 months to process one).

I really dont want to go the new board way as i am tempted by the core i7.

Please if anyone has any ideas please lemme know!


sys specs:

1. Asus p5b-d rev 1.03G (bios 1236) + Intel Q6600 @3.0Ghz + Tuniq Tower 120
2. Corsair 2x1Gig DHX XMS2 RAM
3. HIS HD4870-512
4. Corsair HX620 PSU
5. Seagate 80GB Sata-2 drive
6. Sony DVD writer S-ata
7. Logitech CCT (single Blue)

[chuckbam]

[QUOTE=alitayyab;3323360]I have had this board (Asus P5b-Deluxe Rev 1.03G) for about 2 years. the problem i am going to describe has been there for about 8 months.
...................................

QUOTE]

I would first think Power Supply, but you have a good one. And, you said you checked it. Do you ever run a program like Everest that post the core volts and rails. There is a good Vdroop with this board. I run the CPU at the high end of the specs.

What about the sound card? Did you ever test it without it?

I will go i7 eventually also. Just want the prices and the tech to mature a little.

[chuckbam]

Any good for my settings? I need to look to see if there is a 64-bit. This is a 1 time run with some system load.

Not that long ago, I reduced my setting to 3.28 GHz/ FSB 1640 MHz Mem 820 MHz, 5-6-6-18, Also I have C1E enabled.

I was at 3.36 GHz/ FSB 1680 MHz Mem 840 MHz, 5-5-5-15

[alitayyab]

[QUOTE=chuckbam;3323384]

I have had this board (Asus P5b-Deluxe Rev 1.03G) for about 2 years. the problem i am going to describe has been there for about 8 months.
...................................

QUOTE]

I would first think Power Supply, but you have a good one. And, you said you checked it. Do you ever run a program like Everest that post the core volts and rails. There is a good Vdroop with this board. I run the CPU at the high end of the specs.

What about the sound card? Did you ever test it without it?

I will go i7 eventually also. Just want the prices and the tech to mature a little.

hi,
thanks for the reply.
i used asus pc probe the volts were good.
Vdroop i know of, and is pretty much taken care of. when the system runs it is orthos (1 hour) and OCCT stable.

the only problem is when it starts acting up.
i am thinking of picking up a used x38 board if i dont find an answer soonish!

[chuckbam]

[QUOTE=alitayyab;3323535]

hi,
thanks for the reply.
i used asus pc probe the volts were good.
Vdroop i know of, and is pretty much taken care of. when the system runs it is orthos (1 hour) and OCCT stable.

the only problem is when it starts acting up.
i am thinking of picking up a used x38 board if i dont find an answer soonish!

Did you ever think of an ASUS p35 vanilla to hold you till the i7?

[[XC] Hicks121]

Any of you guys running a yorkie on the p5b-d? I just walked into a great deal on a x3330.

[unclewebb]

alitayyab: I was using a Q6600 - G0 for a while on my P5B Deluxe and it was far harder getting it to run fast and stable compared to any Dual Core I've tried.

I've found that even with the Dual Core, when you have random instability issues you can improve things with more chipset voltage. I definitely wouldn't set anything in the bios to AUTO for voltage or anything else for that matter on this board.

What are your settings on the voltage tab in the bios? When I first started overclocking the P5B I could run for hours, Prime stable and then a day or two later at the same settings I would have weird problems, usually at 0% load, surfing the net or something and my computer would lock up or reboot. It made no sense but extra voltage fixed pretty much everything.

The only issue I have now is that sometimes when I reboot, my wireless keyboard isn't recognized in the bios. I can hammer away on the Delete key when it is booting up in the bios but nothing happens. After it gets to Windows the keyboard works fine. I think it is designed like this by Asus so if you are using Hibernate mode it won't let you make any bios changes but sometimes it gets confused and even if you use ShutDown it still locks you out of the bios and you can't push F8 either to bring up the boot menu.

I thought this was just a wireless keyboard issue but I plugged in a PS2 keyboard and same thing. I've read in forums that I'm not the only one with this issue so I've learned to live with it. It doesn't happen very often and if I wasn't trying to overclock the chipset 100% it probably wouldn't happen at all.

Turning off the power supply switch for 30 seconds usually fixes this.

My E8400 C0 has no issues with voltages and overclock settings being saved or lost with bios version 1237. Hopefully they fix this issue with the next bios release for your E0 Quad.

[alitayyab]

alitayyab: I was using a Q6600 - G0 for a while on my P5B Deluxe and it was far harder getting it to run fast and stable compared to any Dual Core I've tried.

I've found that even with the Dual Core, when you have random instability issues you can improve things with more chipset voltage. I definitely wouldn't set anything in the bios to AUTO for voltage or anything else for that matter on this board.

What are your settings on the voltage tab in the bios? When I first started overclocking the P5B I could run for hours, Prime stable and then a day or two later at the same settings I would have weird problems, usually at 0% load, surfing the net or something and my computer would lock up or reboot. It made no sense but extra voltage fixed pretty much everything.

The only issue I have now is that sometimes when I reboot, my wireless keyboard isn't recognized in the bios. I can hammer away on the Delete key when it is booting up in the bios but nothing happens. After it gets to Windows the keyboard works fine. I think it is designed like this by Asus so if you are using Hibernate mode it won't let you make any bios changes but sometimes it gets confused and even if you use ShutDown it still locks you out of the bios and you can't push F8 either to bring up the boot menu.

I thought this was just a wireless keyboard issue but I plugged in a PS2 keyboard and same thing. I've read in forums that I'm not the only one with this issue so I've learned to live with it. It doesn't happen very often and if I wasn't trying to overclock the chipset 100% it probably wouldn't happen at all.

Turning off the power supply switch for 30 seconds usually fixes this.

My E8400 C0 has no issues with voltages and overclock settings being saved or lost with bios version 1237. Hopefully they fix this issue with the next bios release for the E8x00 E0 processors.

all voltages are set to their lowest value except for Vcore which is 1.3125.
the issue is very very random, if i can reproduce it consistently than i could file an RMA or maybe work towards a conclusion.


the reason i am gunning for a x38 is that a friend of mine is selling his. about 50% off market price.

[chuckbam]

Asus in the US has a very small office for customer service. They can ship you an advanced replacement if you have a credit card.

I don't know if you are in the US.

Up them volts. I run my CPU 24/7 for a long time now at 1.39v at idle. I test it by looping prime95 and 3DMARK06 simultaneously. The 3Dmark really brings out the Vdroop.

The bios setting CPU volt is much greater. But, it gives me 1.39V. I think I have the bios at 1.45+ as a setting.

I would up some other volts as well

I use bios 1219 for the Q


http://www.thetechrepository.com/sho...&highlight=p5b


http://www.xtremesystems.org/forums/...d.php?t=115217

http://www.thetechrepository.com/sho...&highlight=p5b

[unclewebb]

all voltages are set to their lowest value except for Vcore which is 1.3125.

Why? Some of these boards are very sensitive to chipset voltage settings and I can almost guarantee you that if you start bumping those up, your problems will go away. Here's a good place to start:



I had to bump the northbridge to 1.65 volts to get the FSB up to this speed:



I've been using the voltage settings in the screen shot above for the last year and a half if not longer and I haven't hurt anything yet. Stability is excellent.

[alitayyab]

okay i'll try higher nb volts.
but shouldnt the board atleast respond to a hard reset EVEN if this is a chipset issue???
in my 20 years i've never encountered a problem like this.
unfortunately i am not in the US, thus my slightly off RMA process where i have to show the distributor what is wrong with the board.!

[mine]

Any of you guys running a yorkie on the p5b-d? I just walked into a great deal on a x3330.

look @ my post above

got one of the first Q9550 E0 retail

you will get fine benchmarks out of the box .....

but sadly enough after struggling for two days (1236 and 1237 beta)
with my trusted P5BDLX (first gen.-first rev.) I think its time to jump to a new generation of boards

The E 0 is recognised fine but board and the buggy Asus beta bios gave me a lot of headaches :

a. Vcore every time defaults to Auto when entering Bios...same happens with the multi -- changes from 6 to 8 and 8 to 6 and 8.5 nicely

b. 8,5 multi is only working @ default 333 Fsb (+ - 5 ) , no OC possible

c. 8 multi has a FSB wall @ 400 (you need a very insane and high Vcore for a 45 nm CPU)

d. CPU runs quite stable for 6 h with Prime 95 in all instances with only 1, 1 V @ 3 GHz (8x 374) with 2 x 2 GB Apogee 1150 RAM (1.45 V NB )

but fails immediately with IBT and OCCT (CPU /RAM) in seconds


best

m

[unclewebb]

mine: What speed and voltage were you running your Apogee memory at? The P5B has a strap change at 400 MHz so it is a lot easier to run stable at a FSB of 401 MHz than it is to run at 400 MHz, especially with a Quad. I assumed you tried using lots of voltage for the northbridge, etc. to try to get your computer stable. What sort of temps do you get up to with Prime small FFTs?

The P5B isn't exactly top priority for the bios writers at Asus these days but they seem to keep making some slow progress with each release. My E8400 - C0 is pretty happy on the P5B Deluxe now. It needs more chipset voltage than my original E6400 ever did but whatever makes it happy, I give it.

With lots of FSB MHz for my E8400 - C0 I was having trouble in the bios with it locking up while I was making adjustments but 1.65 volts for the northbridge sure cured that! If you plan to replace your board then don't be afraid to give it some juice. Slower, more conservative, memory speeds with CL4 timings works more reliably for me than DDR2-1100+ speeds and CL5 timings. With the tighter CL4 timings, the performance loss is minimal.

I also pulled off my northbridge heatsink a long time ago. It wasn't sitting squarely so I bent the heatpipe a little and used some AS5. If a person was serious and didn't mind some extra noise then bolting a small fan to it would probably be a good idea too.

[mine]

mine: What speed and voltage were you running your Apogee memory at? The P5B has a strap change at 400 MHz so it is a lot easier to run stable at a FSB of 401 MHz than it is to run at 400 MHz, especially with a Quad. I assumed you tried using lots of voltage for the northbridge, etc. to try to get your computer stable. What sort of temps do you get up to with Prime small FFTs?

The P5B isn't exactly top priority for the bios writers at Asus these days but they seem to keep making some slow progress with each release. My E8400 - C0 is pretty happy on the P5B Deluxe now. It needs more chipset voltage than my original E6400 ever did but whatever makes it happy, I give it.

With lots of FSB MHz for my E8400 - C0 I was having trouble in the bios with it locking up while I was making adjustments but 1.65 volts for the northbridge sure cured that! If you plan to replace your board then don't be afraid to give it some juice. Slower, more conservative, memory speeds with CL4 timings works more reliably for me than DDR2-1100+ speeds and CL5 timings. With the tighter CL4 timings, the performance loss is minimal.

I also pulled off my northbridge heatsink a long time ago. It wasn't sitting squarely so I bent the heatpipe a little and used some AS5. If a person was serious and didn't mind some extra noise then bolting a small fan to it would probably be a good idea too.

Hi unclewebb

thanx for your kind words

In fact I had no troubles getting the 2x2GB Apogees stable up to 1068 DDR @ 2V with the former E 6600

The 9550 E0 changed a lot ..you are right with the timings : you have to be careful with these sticks , Asus bios allows a trfc of 42 only and this is way to low for Hi-End 1200 DDR 2 RAM (I ended with SPD timings ..you get trfc 52 or I used memset)

Temps with the noctua in a push-pull config. (Scythe Kama )
are excellent 36-36-36-36 idle -- Prime load 49-48-48-48 ...8 degrees lower
than my first gen. E 6600 !.

Right now NB is @ 1,45 V . i got some kind of lockups in bios just as you reported and weird "no boots" even with bios @ default.
(1,65 V mmmhhhh .. should I ?????)

Biggest shock are the stability tests in IBT and OCCT :

Failure in seconds even @default 333 FSB @2,83 Ghz

Even with Vcore 1.3V (from 1,12)

I like the way Asus is supporting a 2 year old board but please not in that way .. This is just frustrating ...


Best and regards

from Black Forest Germany

M


EDIT : So I followed Uncles advice and upped the voltage on the NB to an insane 1,65 V ! Temps were incredibly high ..I measured 84 C .. . So i installed a 80mm Yateloon silent fan @7v , pointing in the NB direction. And guess what ? Cured a lot of problems . Not all . ..but got my first clean IBT run (256 MB RAM)

Prime 95 custom runs for 6 hours without any problems with a mere 1,09 V (CPUZ) 1,12 in bios. Amazing.

Thanx

M

[unclewebb]

mine: I have no idea why the northbridge requires so much voltage with 45nm but it does seem to help overall stability. I found 65nm Quads were similar needing more chipset voltage than 65nm Dual Cores. Maybe with the next bios release they will have figured something out and I'll be able to drop it from 1.65 volts but until then, I'm leaving it.

As you said, it cures a lot of problems but it doesn't fix everything. I still have a resume from Stand By issue when the FSB is at 500+ MHz but with the 9.0 multi on my E8400, I can survive with 450 x 9.0

Most people are running their new ATI 4850 cards at 80C so I don't get too worried about temp numbers as long as things are stable. The shutdown temperature for Intel Core CPUs is approximately 125C so even though 80C or 85C sounds like a big number, it probably won't hurt anything. What did you use to measure the temperature? I just checked the heatsink on the NB with an IR thermometer and it's presently showing only 48C. I know it can get a lot hotter than that. Next time I feel the heatsink with my magic finger and some skin comes off I'll try another IR measurement. Do you have any software that reports NB temps? I didn't think Asus Probe reported that last time I checked.

Using a program like SPDTool might help give you some control over your SPD memory timings. It's pretty easy to use. Minimum Refresh to Active/Refresh Command Period... is where you can make adjustments to tRFC in the SPD table. I'm not sure if the Asus bios will respect your request or if it will be ignored if it is out of the range that it can handle.

Edit: I just ran Prime Blend for about half an hour. The NB voltage was at 1.65 volts and the IR measured temp of the heatsink was about 55C. I was really surprised when I pulled my heatsink off and noticed what poor contact it was making. Pull it off, put it on a flat surface and bend the heat pipe a little until it allows both ends of it to sit square. I don't have any before and after numbers but a heatsink that is making full contact with a chip has to be better than one only making contact on the corner of a chip. I'm usually not the type of person to check stuff like that out but I'm glad I did.

Now that I stopped Prime, the NB measured temp is about 49C to 50C and the southbridge heatsink is measuring 58C to 59C. No worries.

Edit#2: 57C measured on the NB heatsink is equivalent to being able to hold two fingers on the flat side of the heatsink closest to the graphics card for about 10 seconds. Your mileage or fingers may vary. Try holding your fingers on a Gigabyte DS3 NB heatsink and you'll find out what hot is all about!

[mine]

mine: I have no idea why the northbridge requires so much voltage with 45nm but it does seem to help overall stability. The shutdown temperature for Intel Core CPUs is approximately 125C so even though 80C or 85C sounds like a big number, it probably won't hurt anything. What did you use to measure the temperature? I just checked the heatsink on the NB with an IR thermometer and it's presently showing only 48C. I know it can get a lot hotter than that. Next time I feel the heatsink with my magic finger and some skin comes off I'll try another IR measurement. Do you have any software that reports NB temps? I didn't think Asus Probe reported that last time I checked.

Using a program like SPDTool might help give you some control over your SPD memory timings. It's pretty easy to use. Minimum Refresh to Active/Refresh Command Period... is where you can make adjustments to tRFC in the SPD table. I'm not sure if the Asus bios will respect your request or if it will be ignored if it is out of the range that it can handle.

Edit: I just ran Prime Blend for about half an hour. The NB voltage was at 1.65 volts and the IR measured temp of the heatsink was about 55C. I was really surprised when I pulled my heatsink off and noticed what poor contact it was making. Pull it off, put it on a flat surface and bend the heat pipe a little until it allows both ends of it to sit square. I don't have any before and after numbers but a heatsink that is making full contact with a chip has to be better than one only making contact on the corner of a chip. I'm usually not the type of person to check stuff like that out but I'm glad I did.

Now that I stopped Prime, the NB measured temp is about 49C to 50C and the southbridge heatsink is measuring 58C to 59C. No worries.

Edit#2: 57C measured on the NB heatsink is equivalent to being able to hold two fingers on the flat side of the heatsink closest to the graphics card for about 10 seconds. Your mileage or fingers may vary. Try holding your fingers on a Gigabyte DS3 NB heatsink and you'll find out what hot is all about!

Thanks a lot Unclewebb
the 84 C were measured after 50 runs IBT stress test (half memory) and 1,65 V NB.

Now after getting all primestable for 8 hours

only my IBT failures @ mild overclocking (FSB 375-8x) give me a lot of headaches

I got 50 runs with 512 MB flawlessly
and 5 failures with 1024 MB RAM chosen.. (NB from 1.45-1.65 V)

EDIT: not even IBT stability @ default FSB/VCore and setup default in Bios

[unclewebb]

How are you measuring the temp of your northbridge at 84C? If you are using software then I'd like to try running the same program just to compare. If you're using an IR thermometer like I have then I'd suggest pulling your heatsink off and having a look at the contact it's making.

If you're stable when using 512 MB but not stable with 1024 MB then I'd try adding a little more memory voltage. I've found that a bios setting of 2.10 to 2.15 volts works great for many different types of DDR2 ram. Even modules that are only rated at 1.8 volts will run stable at a faster speed with some extra voltage. That seems to be a common theme for me.

You might burn your ram out at 2.5 volts but 2.15 volts long term won't hurt anything. There usually isn't much to be gained going beyond 2.15 volts in the P5B Dlx bios for 24/7 type use.

Does the 1024 MB test run longer and create a higher CPU core temperature? When well overclocked, most CPUs will lose long term reliability when the core temperature hits 65C to 70C. Every CPU is different and the amount of temperature head room you have varies depending on how hard you're pushing your CPU. Use a nice program like RealTemp which can keep track of your maximum core temperature and you might see a pattern develop between core temperature and stability.

There are some FSB speeds that are easier to run stable at. You lose performance when the FSB goes over 400 MHz on the P5B due to the strap change but it's easier to run reliably at 410 MHz then it is to run at 390 MHz. I'm not a big fan of the 366 MHz to 400 MHz range. Try running with a FSB just over 400 MHz and then use a smaller CPU multiplier so your overall MHz are about the same. This might help isolate your problem.

Personally, I find IBT to be a little bit overkill. It's great to be IBT stable but do you ever use any real world software that heats your CPU up to within 10C of what IBT heats your CPU up to?

IBT is sort of like putting a brick on the gas pedal of your car and then going to sleep for the night. If your engine blows half way through the night then I guess it wasn't "brick stable" but since I never drive around with a brick on my gas pedal then not being "brick stable" isn't really a problem for me or my car. I like being Prime small FFT stable but even that is borderline overkill because I don't use any real software that runs the floating point units of both cores of my CPU at full load for extended periods of time.

Edit: I just tried the LinX front end for linpack and the bigger the test, the more the core temperatures go up so you might be hitting the wall due to excessive CPU heat.

[mine]

thanx unclewebb , appreciate your efforts

Unfortunately (not for me , but for my computerparts) I use a small Silverstone HTPC case .

Though the 9550 E 0 was incredibly easy to cool. I use the small Noctua (9 vers.) heatsink in a pushpull config. with 2 scythe Kama 92 mm PWM fans.

idle : highest temp.: 36 C (core 1) @ 22 C room temp.


load : highest temp : 58 C (core 1) @ 22 C room Temp (IBT-max memory)


So even with small FFts I never saw temps exceeding 54 degrees.


I measured the NB temps again with an IR Thermo. @ 1,55 V after 30 min. IBT and got 62 degrees.

Will reseaten it and apply new Noctua TP .


I tried adding some more juice to the Apogee RAM before (2,2V) but , sorry to say,that was contraproductive. (no Micronchips inside ) they run best @ 1,9-2,0 V)

very interesting : FSB 400-410 @ 8x multi gave me much more problems
than 380x8,5 , thats quite unusual , isn´t ?


Just for fun I reinstalled my trusty old E 6600 (first gen. ES) ....whow what a stable configuration

Really no problems at all ...rockstable with Prime-and IBT

though interesting ...same problems with FSB 400 +

I remember for sure , there were no such problems at all 2 years ago ... I just look at screens with
Superpi 13,0 sec and FSB 9x 420 and 8x 440.

So I think the board may have some kind of degradation or is getting old , Just doesn´t like young speeedy cpu girls inside the house or 2x2 GB 1150 sportscars


like all of us , very sympathic .. in a way ....

May be my young speedy blond girlfriend should get it as a present , playing solitaire on it.

and yes : mmmmhhhh... No Time and Hope for a functional Asus bios

Best

M

[snnoopypw]

Hello!

I'm planning to buy 8GB of DDR2@800MHz for my ASUS P5B-Deluxe BIOS v1236.
The RAM in question is this:
G.SKILL F2-6400CL5Q-8GBPQ, DDR2 800 MHz, KIT 4x2 GB
details:
http://newgskill.web-bi.net/bbs/view...d=g_ddr2&no=95

I'm going to run a 64bit OS with memory remap enabled. I'm pretty sure that I'll be able to allocate all 8GB BUT...the question is will I be able to run it @ 800MHz (which is the default max for P5B-Deluxe)?
I'm asking you experts this becaue I stumbled upon this:
http://forums.hexus.net/pc-memory/10...5b-deluxe.html
to be precise, post #13:

You probably found this out by now, like i did just recently when I bought a Asus P5B Deluxe board... but the 8Gb possiblity of the server can only be done on DDR2 PC5400 or DDR PC-5200 RAM (677 or 533 respectively), as the board doesnt support anything more then 4Gb total in PC-6400 RAM.

Also the above configurations will only work with specific manufacturer chips...I put 4 x 1Gb PC-6400 Rambo ram on my board, and it only detects 3Gb.... which sucks big time..... seems the board manufacturer had a marketing agreement with specific ram manufactuers.... no offense to Corsair, as I'm now looking for the required 4 x 2Gb Corsair RAM for the board to get it to 8Gb but my current problem is that ASUS dont provide the details of the supported RAM modules for 8GB.

If you have done this already Agrippa, any advice would be appreciated!

so, what say ye?

P.S. please no "why do you need 8GB of RAM?"

[Ancient1]

Hi

I think my MarvelYukon died . one RJ45 blank the other orange . using PCI NIC .

Does anyone know where the chip is located ? and what voltage can be changed in BIOS to add juice to it ?

[unclewebb]

mine: Your CPU temps look good and shouldn't be causing a problem. Most memory scales upward with more voltage but from what I've read about the Apogee, it doesn't need extra voltage like the Micron D9 chips do. TweakTown didn't really gain anything by going beyond the rated 2.1 volts when they tested some DDR2-1150 Apogee GT 2x2GB modules last month. Have you ever tried slowing your memory down and running it 1:1 with CL4 or CL5 timings when checking stability? I've always preferred that to big memory MHz numbers on this board.

Maybe the next bios release will help with E0 compatibility. I'm pretty happy with how my E8400 - C0 runs now on the P5B Deluxe but I haven't tried using a 45nm E0, Dual or Quad on the P5B so I guess I can't help you. One board has lasted me for the entire Core CPU generation so I can survive now until Core i7 comes out.

[mine]

mine: Your CPU temps look good and shouldn't be causing a problem. Most memory scales upward with more voltage but from what I've read about the Apogee, it doesn't need extra voltage like the Micron D9 chips do. TweakTown didn't really gain anything by going beyond the rated 2.1 volts when they tested some DDR2-1150 Apogee GT 2x2GB modules last month. Have you ever tried slowing your memory down and running it 1:1 with CL4 or CL5 timings when checking stability? I've always preferred that to big memory MHz numbers on this board.

Maybe the next bios release will help with E0 compatibility. I'm pretty happy with how my E8400 - C0 runs now on the P5B Deluxe but I haven't tried using a 45nm E0, Dual or Quad on the P5B so I guess I can't help you. One board has lasted me for the entire Core CPU generation so I can survive now until Core i7 comes out.

Hi Unclewebb

Was an extraordinary experience with such a nice and skilled person like you !!

and yes I tried all options , even 400 FSB / DDR 2 800 @ 4-4-4-15
in the end I think this may be a combination of problems with my board , which in fact was one of the first P5B-DLX that entered Germany 2006 (no revison), the current Asus bios
and the new Quad E0 revision.

I will jump the i7 waggon a little bit later next year , so a P 45 will be my final solution ( just ordered a Gigabyte P45 DS4 (my first Gigabyte in years !) because of its good MacOSX compatibility :-) and the many success stories with E 0 and Quad CPUs .

Liked as well the German Gigabyte forum with excellent support. Like so many I am really tired of the Asus online support .

Once again thanks for your help and yes ...your tolerance for my reactivated , rusty schoolenglish

Greetings from BlackForest
Germany

Mine

[Agail]

Hi Guys!

I'm having a problem with my P5B dlx, I hope you guys can help me.

I apologize for the long post, but please bear with me:
I had a pair of Crucial ballistix (2x1gb, 800mhz) which I had running at 1200mhz, 5,5,5,12 @ 2.15v- they ran perfectly at those speeds for a couple of months - then after a overclocking session, I pushed them even higher and increased the volts to 2.4 (If I remember correctly), anyway, this fried the ram.

So I got another set of ballistix - (unfortunately the single sided ones) and now I cant get them to run at the rated speeds (They are supposed to be able to run at either 800 @ 4,4,4,12 or 1000 @ 5,5,5,15 - STOCK) I've tried different voltages ect, but it doesn't seem to help.

Now I originally thought that there is something wrong with the ram, so I tested them in a friends pc and the work perfectly at 1000mhz without having to change anything?

At the moment I've got a E0 e8400 445 x 9 @ 1.25 and the Ram at 890 @ 5,5,5,15 - this is highest they will go.
I'm running the latest bios 1237 - I've also tried 1236, still no luck.
My NB Vcore Voltage to 1.45V.
Both my NB and my CPU are watercooled - and the only mod I've done is a Vdroop pencil mod

Now I've gone over my bios settings again and again, tried different timings, volts, sub timings, sub voltages still no luck.
I've tried resetting the bios to stock and only changing the ram speeds - it still wont boot.
Is it possible that I've damaged something on the board?
Any ideas - this is driving me crazy!

[unclewebb]

Agail: As I mentioned above, I think this board appreciates more NB voltage when using the newer 45nm CPUs. My NB is only air cooled but it seems to like having the NB set to 1.65 volts in the bios. On the previous page I posted some bios voltage settings that are working good for my E8400 - C0.

http://www.xtremesystems.org/forums/...postcount=4925

I think the facts of life are that this board is a couple of years old now and it is not really optimized for some of the newer memory. Some new sticks that can run DDR2-1200 speeds on a newer board with a newer chipset probably won't be able to run that same speed on a P5B Deluxe. Asus makes a little bit of progress with each bios release but it just isn't that high a priority for them.

When you say that you can't get them to run at the rated speed does that mean they aren't Prime stable or you can't even boot up?

I was having some boot up issues before (it would hang) but I was running some mismatched sticks. One was a Ballistix 1GB and the other was a Team. With the extra voltage and some new OCZ DDR2-9200 Reapers it's been trouble free lately.

It's also possible that your CPU might need a little more voltage now. After the initial burn in some 45nm CPUs need a little more voltage to run the same overclock.