Sign in with
Sign up | Sign in
Your question

975XBX acting funny

Last response: in Motherboards
Share
December 19, 2006 4:23:52 AM

Okay, I'm on the verge of throwing my PC away. My PC won't POST, and when I get it to work, then I can either run for a short while and then it hangs, or like it is now, my girlfriend has been playing NWN2 for 3 hours non stop, and no issues.

My Spec:

CPU : E6600 - Stock Cooling - No OC - (Idle 44-46C) -
(Max Load 63C S&M @ 100% 15min)
MoBo : 975XBX Rev 304
SoundCard : Audigy 2
RAM : 2gig (512x2+1) DDR 533 stock (Corsair Value)
GCard : GeCube X1900XTX with Accelero X2
Idle temp 46-49C / Load in game 88C
HDD : Seagate 160 SATA 2 + Western Digital 80 SATA 1

I did a complete rebuild last Wednesday, and then played games for 4 days non stop - NO issues.

Sunday Morning, I start the PC and the damn thing will not initialize the Screen. Everything starts up, yet it seems as though the BIOS is not working. I fiddled around a bit, and got it to start up. When I play a game, the PC freezes (all temps are normal) sometimes and other times not.

I decided to take everything apart, and run with:
1.) CPU + PSU (mobo obviously) : At first no result, but after resetting a bit the BIOS gives beep for memory
2.) Memory in : Nothing happens - CPU Fan running yet no beeps.
3.) Graphics card in: Initializes and complains about PCIx power
4.) PCIx power in : boot to no HDD
5.) HDD in : boot to windows

All seems fine, I reassemble all other components, all the while testing if the PC will boot. With everything complete, I start and all seems well. I play 72 holes of Tiger Woods 2007, which can stress the PC to some extent and all is well.

I exit, reboot and start TW 2007, and then play some more - but this time it freezes and I reset only to have a no POST situation.

Things I have done to go from no POST to working:
1.) Reset switch - hold in and allow the BIOS to initialize and then releasing : worked once
2.) Reset : repeatedly pressing Reset also worked once
3.) Remove Reset connection : No good result
4.) Fiddling with 24pin power connection can freeze the PC in windows, or cause the non POST pc to POST.
5.) During the testing last night, I left the BIOS batt out for more than 1 hour (while I was away) - yet my settings were not removed.
6.) Replacing the side cover seems to make the PC go to a no POST situation - I suspected it might push down on the 24pin Power connection cable, so I rerouted it - results are still the same.

Funny thing is, if something is broken, the likelyhood of the PC failing in X amount of time is quite big. Yet, on occasions it runs for 3 hours pushing all facets of the PC - temps are in check and everything okay.

So I really do not know what is going on, and would appreciate some other help or angles to look at my problem. I'm thinking PSU, 0V grounding the mobo, or gcard.

Thanx

More about : 975xbx acting funny

December 19, 2006 4:42:44 AM

Thanx for the reply!

I'm running a Antec TP2 430W (yes, it's below suggested wattage but I've been running this through my X1800XL and X1900XTX for a long time, plus the Core 2 has a 06 power rating, that is marginally better than my P4 640 at 05)

How do I clear the BIOS? I shorted pin 2/3 and left for a while, but the settings never changed. I removed the battery for an hour, yet the settings were the same. I even reverted to the 1378 BIOS by doing a BIOS flash from Windows. I'm thinking of leaving the battery out over night.

Will test the memory tonight.

I failed to mention that we had a big thunder storm here on Saterday - so i've thought of the memory being hurt or the bios being hurt. But, as soon as I think the mobo is broken I get it to game without any issues. And I also checked all the caps (visually) and the mobo in general aswell, no shorts, no burned pins.
December 19, 2006 5:00:49 AM

Do you have the latest Bad Axe Bios? You can download it here: Latest Bad Axe Bios

Are you using Intel's IDCC or IDU for looking at Voltages and Temperatures?

I'm running a Bad Axe with a 3.46GHz EE Presler and have not run into any problems like you are experiencing.

The easiest way to clear cmos is to pull the batter and either us the batter or any conductive device to short the positive and negative battery terminals in the battery holder. Ensure power is removed from the board when you do this. If you do need to update the Bios remember to enter bios after you update and hit F9 to set defaults then you can change them to suit your needs.
Related resources
December 19, 2006 5:11:57 AM

Quick test you can do to see if the board is at least reaching the memory POST code is to pull all of the memory and power the system up. The board should go through POST and when it hits the memory test POST it will beep 3 times if it does not detect any memory.

For a board to not display anything on the screen I think it could be one of these things. 1) Monitor might be going out. 2) Video card is having issues. 3) Bios is corrupted but that would usually kill the board. 4) Your PS is causing a under voltage condition during POST and it is enough to through the POST process off.

You can also try turning everything to disable in the Bios and do some burn in testing and see if that changes anything.

If the memory is suspected only put one stick in and stress it. Do the divide and conquer.

It would help if you had a POST card so we could see if the board is actually making through POST or getting stuck on a particular POST code. POST cards are what the Intel Board Validation and Bios developers use to help debug failing boards.
December 19, 2006 5:25:29 AM

I have the latest BIOS (last week or the week before) - I reverted back to the previous version yesterday in the hopes that I could reload the bios.

I have IDCC, and all seems okay. Speedfan also shows normal volts - I'll check with a multimeter on the volts.

I did remove the memory, and was also checking for the memory absent beeps - but somehow it didn't do it first time around. I'm starting to think the BIOS is all messed up - but why does the computer freeze in windows - the bios once loaded can't have an impact on the OS?

When I remove the 24pin power connection, I can upon putting it back get to a post state almost always. What does the 24pin supply to the mobo in an off state? Does it supply anything to the BIOS?

My abit had a 7 segment display, that ran thorugh the whole BIOS sequence - I wish everyone would put it on the mobo. Can't one use the debug mode on the 975xbx to check the bios?

I guess the easiest way is to swap out parts, but the problem is that I do not have any spare, and buying a mobo or psu or gcard is going to be expensive. Once again, the fact that I can get everything to run at some occasions tells me that the problem is either based on a condition being reached, or just some stupid small problem giving me
December 19, 2006 5:59:49 AM

Quote:
I have the latest BIOS (last week or the week before) - I reverted back to the previous version yesterday in the hopes that I could reload the bios. Please install the latest bios but use the Bios recovery method. Use either a CDROM with the .bio file on it or use a USB thumb drive and have the .bio file in the root directory. Pull the Bios jumper off the board and power the board up. It will automatically look either for a CD or a USB thumb drive and will load the Bios. When it is done just re-install the Bios jumper and then go back into Bios and use the F9 to ensure Bios defaults are set.

I have IDCC, and all seems okay. Speedfan also shows normal volts - I'll check with a multimeter on the volts. If the voltage is changing to rapidly you might not see anything with the multimeter.

I did remove the memory, and was also checking for the memory absent beeps - but somehow it didn't do it first time around. I'm starting to think the BIOS is all messed up - but why does the computer freeze in windows - the bios once loaded can't have an impact on the OS? That is correct. Once the Bios gets to init19 and hands off to the OS it does not effect the OS. The issue might be that the Bios hands off corrupted system information and you trip over that later.

When I remove the 24pin power connection, I can upon putting it back get to a post state almost always. What does the 24pin supply to the mobo in an off state? The board provides a a 5V Standby voltage that powers the CMOS clock. It is what allows the Bios to retain its settings even if you pull the CMOS battery. Does it supply anything to the BIOS?

My abit had a 7 segment display, that ran thorugh the whole BIOS sequence - I wish everyone would put it on the mobo. Can't one use the debug mode on the 975xbx to check the bios? The cost for the POST Code display and surrounding logic is around $1.35 or more. That is a lot for the board company to suck up on each board sold. Sorry, not sure what you mean by debug mode??? There is a second series of pins that can be jumpered that allows further Bios features to be unlocked but I don't remember any debug features that come up with it. But I only did it once and I don't do much over clocking.

I guess the easiest way is to swap out parts, but the problem is that I do not have any spare, and buying a mobo or psu or gcard is going to be expensive. Once again, the fact that I can get everything to run at some occasions tells me that the problem is either based on a condition being reached, or just some stupid small problem giving me
Using the Bios Recovery method at least will put you back to a stable Bios image. That is what I do if I suspect a boards Bios has been corrupted. If the boards still used the socketed Firmware part then the best is to use the .hex or .rom file and burn the bios with a chip burner.
December 19, 2006 6:11:51 AM

I think the BIOS will be my saviour, and then it's the PSU I will tackle. Somehow I suspect the PSU might be dipping for some obscure reason (like mentioned before).

Thanks for your suggestions! Much appreciated! I'll give it a go tonight - my girlfriend says the PC is still going and she's been playing NWN2 now for 5 hours.
December 19, 2006 6:16:23 AM

You sir have a very interesting problem. :?

My Bad Axe is a -302 that I reworked to support Conroe and Kentsfield processors. Since the uCode for Conroe and Kentsfield processors are the same all that was really needed for my board to support a Kentsfield is what is called GTLREF. There is a GTLREF circuit for each core. Until the -304 boards they were not stuffing the board to support Conroe or Kentsfield.

I check back tomorrow and see how you came out. :) 
December 19, 2006 11:35:26 AM

One more question please: what can make a PC hang in windows, if all temps are okay? I say hang, meaning the PC just freezes and onlt a power switch held in or ac mains disconnect will get the PC out again.
December 19, 2006 4:39:06 PM

For a system to hang in windows is usually caused by corruption in the OS or a program running in the OS. This can be caused by bit errors in memory, a bad windows installation or something on the board that causes a bit error. This could be caused by one of the power rails going low so a digital one or zero cannot be determined correctly.

From what you have been saying it sounds like a memory or Windows OS issue. I believe you said you re-installed windows so I have been ruling that out.
December 20, 2006 4:00:29 AM

Okay, so last night I tried a different PSU (500W). Still no post.

I removed the gcard, and everything else except the CPU and the new PSU. I was trying for the BIOS to beep on no RAM present, as I'm sure it's one of the first things the BIOS wishes to load. No beeps. I first clear the CMOS, and then retry with no ram. Still no boot. So, I retry after removing the CMOS battery for 45min. I get a long beep - shutdown and startup again - shorter beeps. Put memory in randomly between the three sticks I have and still no post. Memory out, no beeps.

I'm thinking mobo might be damaged - what do you think? With just a CPU and power, you should be able to get some sort of beep sequence? Even if the CPU is buggered, BIOS should report no CPU present. I also ruled out the CPU initially, seeing that I could run TAT and S&M for 10min at full load with reasonable temps. Like you mentioned, something let's windows hang, and I have a funny feeling the mobo is the culprit.
December 20, 2006 4:44:04 AM

Please clear the CMOS by pulling the Lithium ION Coin battery with the Power to the board removed and then short the two terminals in the battery holder. You can also just flip the coin battery and reinsert it into the battery hold do this for ~15 seconds and the CMOS will be reset.

Next try doing the bios recovery I suggested last night.

If you have another CPU like an older Prescott, Presler or Smithfield you can try powering the board up with one of them. If the board works then I would say the processor is the problem if the board still won't boot then RMA the motherboard. Intel warranties their motherboards for 3 years if memory serves me correctly.
December 20, 2006 4:55:54 AM

I removed the battery, shorted the two terminals and left the battery out while I went to fetch my dad's PSU.

Still the same.

Will the BIOS always report a faulty CPU, or is there a chance that the BIOS might not pick up CPU faulty, and hang?

I'm going to RMA it - I just want to get to a point where I'm almost certain that the CPU will not hang the BIOS.

My reasoning goes like this:
BIOS won't initialize screen - I can get it to post, by doing something which I don't know what. If I do get it to boot, games can be played and CPU can be stressed with FP calcs. It may freeze, or it may not. Removing all but CPU yields no beep codes, unless BIOS has been cleared or battery removed and +/- shorted. With CPU being able to run in game, aswell as stressed for 1 to 4 hours, the CPU I cancel out as faulty. The only thing left is mobo (psu tested).

I guess, we can't really pin point the problem. So, with the X1900XTX warantee voided and CPU suspected to be healthy, I'd recon a new mobo might yield the solution. At least I'll have two mobo's... :lol: 

Thanx for all your help and time!
December 20, 2006 4:58:58 AM

You're welcome.

I still think you should try the Bios Recovery before giving up.

:) 
December 20, 2006 5:03:31 AM

That was my next step last night - but it won't POST at all now, no matter what I do.
December 20, 2006 5:05:14 AM

You can also try putting the Bios jumper in the Maintenance mode Pins 2 and 3 and see if it POST's into Bios.
December 20, 2006 5:16:42 AM

Tried it with pin 2 and 3, and with nothing aswell - no luck.

What is the sequence of the POST?

If the CPU can not do it's function or be called upon, the POST should report it as faulty? POST is independant of the CPU, as it has to diagnose the system, and that includes the CPU.
December 20, 2006 5:28:41 AM

What happens is when the power switch is toggled this causes the PS to start turning on. When it come up to and the various rails come up to operating values it provides a signal to the motherboard called power good.

Next the Bios ie(firmware) gets moved across the bus and uncompresses and put into the main memory. This can't happen if the CPU is bad or not in the board. This also can't happen if the main memory is bad or not in the system. This is when you should receive the 3 Beeps alerting you to the fact that you need to check your memory is installed.

After the Bios is up and running in the main memory it starts the POST processes for the rest of the board. The POST process starts enabling and varifying that the various onboard devices are enabled. The Bios hands off to the OS at what is called init19. This is the point that bios provides the OS with all of the information about the motherboard that it needs to identify and load the drivers for the motherboard.
December 20, 2006 5:29:30 AM

If I remove the CPU, and start up I should get a non stop beep for Processor Problem?
December 20, 2006 5:30:51 AM

Nope, the board will not even turn on. The power good signal goes through the processor and then onto the MCH and ICH and then to the firmware.
December 20, 2006 7:08:00 AM

Yip - 10/10 for being an a** - only realized that the CPU out will not start the mobo :oops:  .

Went home quickly - Took the CPU out, reseat it. No memory, start - 3 beeps. Memory 1 for 1 in, all okay. In goes the gcard, starts up okay. Did S&M for 15 min at full, passed. Played Tiger Woods 2007, no problems.

I think there's a loose contact on something, or a component is heating up to a point of failure. Maybe the PEG slot is buggered.

NEcase, going to get a new mobo and RMA the XBX - I'll use the return for my girls rig...

Thankyou greatly once again! I'll get back on what happened next, season 2 if you wish :D 
December 20, 2006 7:09:39 AM

Okay, I will stay tuned. :D 
December 20, 2006 7:37:08 AM

Quote:
Yip - 10/10 for being an a** - only realized that the CPU out will not start the mobo :oops:  .

Went home quickly - Took the CPU out, reseat it. No memory, start - 3 beeps. Memory 1 for 1 in, all okay. In goes the gcard, starts up okay. Did S&M for 15 min at full, passed. Played Tiger Woods 2007, no problems.

I think there's a loose contact on something, or a component is heating up to a point of failure. Maybe the PEG slot is buggered.

NEcase, going to get a new mobo and RMA the XBX - I'll use the return for my girls rig...

Thankyou greatly once again! I'll get back on what happened next, season 2 if you wish :D 


I'm not saying this is your problem, but I had similar issues for a couple of days with an Asus P5GDC-V motherboard with Prescott CPU intermittently not booting, sometimes it would freeze during use.. I tried the usual suspects, power supply and memory...The problem,.. I had a Western Digital 80 gig drive as a secondary master, data only, and the I/O chip was going bad! It was taking down the system or preventing proper POST. I removed that 80 Gig WD and all was working well again.
December 20, 2006 10:09:54 AM

Thanx for the reply. My testing was always done with the HDD as last thing to connect and sometimes with the HDD's right through

If a bad SATA chip can cause a post problem, then I'm even more convinced that my mobo got some damage either by me (rebuild), the lightning or heat or just bad luck.

Due to your post, I'm feeling a bit better with my new Mobo that I ordered :lol: 

Oh well, if Intel will fix/replace the 975XBX, I still win. I've been meaning to build my woman a PC - now's as good a time as any, with the R600 around the corner... :lol:  :lol:  :lol: 
December 21, 2006 11:18:48 AM

Problem solved - was the mobo. Just reinstalled everything, and the system boots perfectly. Will however check if it stays this way. I really like Asus mobo's.....
December 22, 2006 3:03:39 PM

I've seen you ask about shorting the 2/3 pins a couple of times, and I don't think you got an answer on that... though I didn't read every post in this thread thoroughly. :) 

I've had to short 2/3 on the bios configuration jumper about half a dozen times while tweaking my overclock settings. Whenever I can't post I short the 2&3 pins, power down completely (either flip the switch on the PSU or unplug it), then reboot with it still shorted. This will boot with default settings and allow me back into the bios to correct whatever setting I borked. When you save the corrected settings the bios will instruct you to power down and revert the jumper back to its original position.

I doubt this will make any difference for you though, since you're not tweaking anything.
December 22, 2006 3:15:20 PM

I guess I buggered up the XBX BIOS wise, or I damaged something onboard. Mind you, the XBX is renowned for being stubborn with wrong bios settings.
!