ASUS P5B-Deluxe; Problems & Fixes

[Big.Wayne]

Hello,

I just put my conroe system back together a few days ago and its been running great, however the last 24 hours something strange has been going on?

I was sitting doing my surfing and the screen went black, almost like the power-saving kicking in any I wiggled the mouse and a moment later the picture came back?

Just a few hours ago the same thing happened but this time the screen stayed blank and I had to restart the machine.

Upon reboot it went:

BEEEEEEEEP . . . . . . . beep, beep, beep (thats one long beep and three short beeps) then switched off again, then rebooted and went all the way to windows.

What do these beeps mean please?

Once in windows I had the feeling that something wasn't quite right so I unistalled the drivers and tried the newer versions (CAT 7.8) however things still seemed a bit sluggish on the desktop so I opened CPU-z for a peek. . .



It seems my PCI-E had set itself to 1x which is something I did read about 6 months ago but it never happened to me before. I restarted the machine a few times and it didn't go away, I had a look in BIOS and everything seemed in order. I then decided to shutdown the machine and give it a minute to rest. Upon reboot the PCI-E setting had corrected itself (back to x16).

I've used this puter for 7 months at least and never had this happen before. As I said I did rebuild it a few days ago and added in a Creative X-Fi XtremeGamer but apart from that nothing has changed?

I'm running:

ASUS P5B-Deluxe
Core 2 Duo E6300 : 2800MHz (7x400)
Crucial tenth anniversary sticks (2x1GB)
Powercolour 512MB Radeon X1950XT
Antec Truepower TRIO 500w

All the volts are on manual/minimum except vCore which is 1.3250v and NBCore which is 1.45v (purple).

Its been ok for the past few hours and I have since re-seated the card.

I know this has happened to people before but I cant remember the fix? should I change my PCI-E Freq from 100Mhz to 101Mhz or what?

[Lou Snutt]

All the volts are on manual/minimum except vCore which is 1.3250v and NBCore which is 1.45v (purple).

Its been ok for the past few hours and I have since re-seated the card.

I know this has happened to people before but I cant remember the fix? should I change my PCI-E Freq from 100Mhz to 101Mhz or what?

Some people report success upping the PCI-E Freq. Only taking it down worked for me - stable for 2 months at 98 (again some report requiring it as low as 96). And keep the NBCore at 1.45v.

Hope it works for you.

F.

[Big.Wayne]

Some people report success upping the PCI-E Freq. Only taking it down worked for me - stable for 2 months at 98 (again some report requiring it as low as 96). And keep the NBCore at 1.45v.

Thanks for the reply,

Yes I did read through this whole thread and others when I first setup the board but I cant quite remember what the issue was, now there are so many thousands of posts in this thread its a daunting prospect to sift through them

Why did you stipulate to keep the NBCore at 1.45v?

Also have u got any idea what the beeps mean? I have had this before with various Radeon cards?

[Lou Snutt]

Why did you stipulate to keep the NBCore at 1.45v?

Also have u got any idea what the beeps mean? I have had this before with various Radeon cards?

From what I have read on various fora, it seems to be the combination of PCI-E and NBCore that provides the stability and the 1.45v provides the best chance of stability; changing either can throw it off again. It also seems that the PCI-E setting that provides stability may vary from board to board to it's a case of find the one that works for you. 98 and 1.45v work for me so I am keeping them locked down.

As for the beeps, I have found only on reference to that combination on Tom's which suggests re-seating the graphics card.

[mashakos]

Disable QFan.

done that. No go.

[Lou Snutt]

done that. No go.

I'm currently running 4.33 beta 9 quite successfully on the P5B Dlx Wifi App. As has already been mentioned, there are only 2 independent speed controllers on this board: one for the CPU (as long as you are using a CPU fan with a 4-pin header e.g. stock with most CPU's) and a second which controls all fans connected to connectors CHA-FAN 1 to CHA-FAN 3. This second one is controlled by the Speed04 setting. So, your thermal monitors need to be set to control Speed04 and the fans that you want to control need to be connected to the CHA-FAN 1, 2 or 3 connectors.

[mashakos]

I'm currently running 4.33 beta 9 quite successfully on the P5B Dlx Wifi App. As has already been mentioned, there are only 2 independent speed controllers on this board: one for the CPU (as long as you are using a CPU fan with a 4-pin header e.g. stock with most CPU's) and a second which controls all fans connected to connectors CHA-FAN 1 to CHA-FAN 3. This second one is controlled by the Speed04 setting. So, your thermal monitors need to be set to control Speed04 and the fans that you want to control need to be connected to the CHA-FAN 1, 2 or 3 connectors.

* checks manual * now that's just great! The loudest 120mm fan I have has a short wire that can only be connected to either CHA_FAN2 or CHA_FAN4.
Guess I have to open up the case to check where it's connected...

thanks for the heads up, I wouldn't have found out this is the way the P5B was designed otherwise.

[Lou Snutt]

* checks manual * now that's just great! The loudest 120mm fan I have has a short wire that can only be connected to either CHA_FAN2 or CHA_FAN4.
Guess I have to open up the case to check where it's connected...

thanks for the heads up, I wouldn't have found out this is the way the P5B was designed otherwise.

No probs.

I've found the 4.33 beta indicates both cores on my E6400 running at the same temp whereas the older version suggested that Core 1 was 2 deg C cooler than Core 0.
Also the speed control of the beta works better than the earlier version on this board. I have set min speed = 50% (Noctua fans don't turn at all at less than this) and max speed = 100%. The earlier version tended to switch from the min to the max settings but the 4.33 will hold them at an in-between value (60% or 80%) if that's all that's needed. Makes for a much quieter box.

[Big.Wayne]

one for the CPU (as long as you are using a CPU fan with a 4-pin header e.g. stock with most CPU's)

I have a 120mm Sharkoon '2000' cooling my processor and three 80mm Sharkoon '2000' cooling the chassis.

The 80mm fans are being slowed down by the Q-Fan thing to about 1300rpm which makes them all pretty quiet, however the 120mm fan always runs at full blast?

All the fans are three pin so I wonder why Q-Fan can't work its magic on the CPU fan?

I see it has a 4-pin header (PWM) connection but that should make the Q-Fan work a bit more accurately than a 3-pin, it shouldn't mean a 3-pin can't have its speeds adjusted at all?

Anyone got a 3-pin fan to work with Q-Fan on the CPU header?

[Lou Snutt]

All the fans are three pin so I wonder why Q-Fan can't work its magic on the CPU fan?

I see it has a 4-pin header (PWM) connection but that should make the Q-Fan work a bit more accurately than a 3-pin, it shouldn't mean a 3-pin can't have its speeds adjusted at all?

Anyone got a 3-pin fan to work with Q-Fan on the CPU header?

Bad news, I'm afraid. The CPU fan can be controlled ONLY via the 4th pin. If you want to control the speed of your 3-pin CPU fan then you will have to connect it to one of the CHA-FAN 1, 2 or 3 connectors and set BIOS to ignore the CPU fan warning. This is how I am running the 120mm Noctua on my Scythe Ninja cooler. I haven't tried this with Q-fan running as I let Speedfan control it all.

[Big.Wayne]

Bad news, I'm afraid

Yup I realised that about 6 months ago, just wanted to find out why?

I haven't tried this with Q-fan running as I let Speedfan control it all.

Thats cool? you use Speedfan to control the case fans? didn't think of that

[Lou Snutt]

Yup I realised that about 6 months ago, just wanted to find out why?

Thats cool? you use Speedfan to control the case fans? didn't think of that

Why comes from the different methods of speed control between 3 and 4 wire fans. 3-wire fans are controlled by varying the effective voltage on the 12V wire (usually by PWM). A 4-wire fan has the 12V wire tied to the 12V supply and the control is managed internally within its associated electronics according to the signal input on the 4th wire.

I'm lucky with my Antec P180 case and Ninja cooler that I can fit a 120mm fan on the Ninja to pull air across the RAM and then across the CPU and there is a rear case fan directly in line to take the air out of the case. So by using Speedfan to control these 2 fans depending on the Core 0 and Core 1 temps the whole system is suitably ventilated.

[ChaosMinionX]

Strange problem with a B3 quad, using 1216 bios on P5B deluxe.

I set FSB to 400mhz and the multi to 8... yet the board still displays 3.6ghz and not 3.2ghz? Any fix for this?

Thanks

[unclewebb]

ChaosMinionX: Lot's of software reads the default multiplier for your cpu which is 9X for a Q6600 and then reads the FSB, 400 MHz, and multiplies the two together and assumes you must be running at 3600 MHz.

CPUz doesn't have this problem and works excellent on my P5B Deluxe. The latest beta version finally reports the core voltage correctly as well.

[ChaosMinionX]

ChaosMinionX: Lot's of software reads the default multiplier for your cpu which is 9X for a Q6600 and then reads the FSB, 400 MHz, and multiplies the two together and assumes you must be running at 3600 MHz.

CPUz doesn't have this problem and works excellent on my P5B Deluxe. The latest beta version finally reports the core voltage correctly as well.

Well I know that, but the bios itself isnt even reading correctly....I will have to further troubleshoot it Thanks for your reply though.

[Patriote]

Anybody having problem with their Q6600 on a P5B Deluxe with bios 1216 ?

It seems that whatever vcore i set, i can't get pass 3.6Ghz. It does 438x8 = 3.5Ghz but not 400x9 = 3.6Ghz Whatever vcore, NB FSB Termination voltage i try, it keeps rebooting when i stress using 4 instance of Orthos.

[smsmasters]

Anybody having problem with their Q6600 on a P5B Deluxe with bios 1216 ?

It seems that whatever vcore i set, i can't get pass 3.6Ghz. It does 438x8 = 3.5Ghz but not 400x9 = 3.6Ghz Whatever vcore, NB FSB Termination voltage i try, it keeps rebooting when i stress using 4 instance of Orthos.

Try FSB Term set to AUTO and only set NB Voltage and CPU VCore.

[Patriote]

Try FSB Term set to AUTO and only set NB Voltage and CPU VCore.

Tried that my friend! I tried all sort of FSB Termination and NB voltage combination and still, it keeps rebooting randomly during stress with Orthos. Yesterday night, i removed my G.SKill RAM kit and hooked a 1GB stick of OCZ PC2-8000, set all timing stock for 800Mhz, i set 3.6Ghz 400*9 1.45v and still, when i woke up this morning, the computer was onto windows log-on screen.

Now it's either the mobo, the mobo bios, the PSU or the CPU itself. But how could it be the CPU if it does 438*8 (3.5Ghz) but not 400*9 (3.6Ghz) ?

[karateo]

try playing wth sb voltage.
i had the same problem with my E6300@3500 (500fsb)

I thought that there was a power cut( we have plenty in our town) and it took me two weeks to realize the problem!

[Patriote]

SB Voltage huh ? Ok ill give it a try! I really wonder how it could help but at the point im at, i'd try anything!

[karateo]

i also thought that it wasn't related with high fsb/clock but it turned out that this was the reason that i couldn't make stable my c2d.

i hope i helped

these are my voltages i think ( i am not sure because i am @ stock right now)
vcore 1.35
vdimm 2.25
fsb 1.4
nb 1.45
sb 1.6
ich 1.057

for this result:
http://www.xtremesystems.org/forums/...0&d=1186523872

[Patriote]

i also thought that it wasn't related with high fsb/clock but it turned out that this was the reason that i couldn't make stable my c2d.

i hope i helped

these are my voltages i think ( i am not sure because i am @ stock right now)
vcore 1.35
vdimm 2.25
fsb 1.4
nb 1.45
sb 1.6
ich 1.057

for this result:
http://www.xtremesystems.org/forums/...0&d=1186523872

Thnaks i will try to play with SB and ICH voltage and will post back!

[Patriote]

Well, i tried 400*9 with 1.50v (real), 1.60v for SB, 1.057v for ICH and it still rebooted after 6-7hours of stressing with Orthos!

This is pissing me off so much... Stuck @ 3.5Ghz... I'm almost better to just hook my E6600 back and run it at 3.8Ghz ...

[unclewebb]

The most important voltages for me were ICH and NB when testing at high FSB MHz.

fsb 1.4
nb 1.55
sb 1.6
ich 1.215

The NB cooler on the P5B Deluxe works excellent. I had to remove mine and straighten out the heat pipe a little so it made good contact. It works far better than the one on the DS3 so don't be afraid to give the NB some extra voltage.

[Patriote]

The most important voltages for me were ICH and NB when testing at high FSB MHz.

fsb 1.4
nb 1.55
sb 1.6
ich 1.215

The NB cooler on the P5B Deluxe works excellent. I had to remove mine and straighten out the heat pipe a little so it made good contact. It works far better than the one on the DS3 so don't be afraid to give the NB some extra voltage.

As i said, i tried everything combination of voltage! Do you see my avatar ? It's the comp im running ATM. The NB is watercooled as well. I tried to give max voltage for FSB and NB and it didn't chnaged a darn thing. It still will reboot when stressing. Im not trying to achieve high FSB! Just 3.6Ghz with 400*9. It does 438*8 (3.5Ghz) so i really don't understand why it won't do 400*9 (3.6Ghz).

But Thanks anyway!