Motherboard or CPU problem?

[pastymuncher]

Hi guys, a little help would be much appreciated. My main rig, specs in siggy plus Sony Sata dvdrw, WD 320Gb AAKS sata ll hdd (Vista 64 bit) and a Seagate 7200.10 320Gb sata ll hdd (XP 32 bit) has been having some problems since Wednesday.

It started off taking longer and longer to boot and then when i go to shut down it goes into standby instead. I then have to press the power button on the case and it turns back on from standby and then shut's down properly. In Vista's problem reporting section there was an entry for the CPU which went something like high power features have been disabled on core 0 and the same for core 1. Last night it also would'nt recognise the dvdrw drive. This afternoon it refused to boot at all. It got to the IDE/Sata controller part of the post sequence and displayed a load of gobbledegook after IDE 2 which is the dvdrw drive.

I have tried clearing cmos, using one hdd at a time, changed the sata cables for brand new ones, changed the dvdrw, ran memtest for four hours with no faults, ran at stock speeds and it does'nt matter if i use Vista or XP, it has the same problems. I am thinking maybe the chipset on the motherboard has gone.

The rig has been running at 3.8Ghz for several months now as the gains with 4.5Ghz are minimal. For 3.8Ghz i use a 400fsb and the board needs no extra voltage for this so it's not a overvoltage problem and it certainly is'nt heat related. Nothing has been changed, it just started out of the blue. I would appreciate any suggestions/thoughts you may have. Thanks for taking the time to read this.

 

[Chimerical]

Update the bios? Sounds like you've done pretty much everything already. Could try reseating the chipset heatsinks possibly.

 

[pastymuncher]

Already running the latest bios and it does'nt look like DFI will be releasing any more for this board. I ran Linx (similar to IBT) last night for a good couple of hours using max available memory. No errors and chipset temp did not exceed 24 degrees!! The joys of having a window mounted rad.

Going to try de-activating the sata ports one by one now but apart from that there's not a lot else i can do.

 

[westom]

It started off taking longer and longer to boot and then when i go to shut down it goes into standby instead. I then have to press the power button on the case and it turns back on from standby and then shut's down properly. In Vista's problem reporting section there was an entry for the CPU which went something like high power features have been disabled on core 0 and the same for core 1.

If you do not measure DC voltages with a meter, then every answer will be speculation.

Don't know why you are talking about increasing voltage. That is not relevant. Current is relevant.

Now, measure critical voltages when power supply connects to motherboard. Probe inside that nylon connector. Measure voltages on any one of red, orange, yellow, and purple wire. Note these voltages (to three significant digits) as the computer powers on, loads programs, and when it is multitasking to every peripheral device simultaneously. IOW play complex graphics, while downloading from the internet, while interfacing with a USB device, while searching the hard drive, while ...

Now report those numbers here since numbers can contain information you don't even realize.

 

[pastymuncher]

If you do not measure DC voltages with a meter, then every answer will be speculation.

Don't know why you are talking about increasing voltage. That is not relevant. Current is relevant.

Now, measure critical voltages when power supply connects to motherboard. Probe inside that nylon connector. Measure voltages on any one of red, orange, yellow, and purple wire. Note these voltages (to three significant digits) as the computer powers on, loads programs, and when it is multitasking to every peripheral device simultaneously. IOW play complex graphics, while downloading from the internet, while interfacing with a USB device, while searching the hard drive, while ...

Now report those numbers here since numbers can contain information you don't even realize.

Are you posting in the right thread? I did'nt mention increasing any voltages at all. It's actually back at stock at the moment.

It would seem that SATA channels 0 and 2 are dodgy from de-activating them one by one and moving the leads around. Look's like a motherboard RMA.

 

[westom]

It would seem that SATA channels 0 and 2 are dodgy from de-activating them one by one and moving the leads around. Look's like a motherboard RMA.

Or it can be what tends to many electronics slowly act 'dodgey'. As bad voltages get worse, various functions act marginal; then act defective.

It may not be DC voltages. But when electronics get strange, checking DC volts is a first thing done. 30 seconds of labor. It may save you how much money?

No, you were not discussing voltages. But your symptoms were.

 

[Mattus]

If you do not measure DC voltages with a meter, then every answer will be speculation.

Pretty much every post you've made has said this? If there are problems during the POST, it's not going to be a PSU issue unless the PSU has given up completely, which it doesn't seem like it has.

pastymuncher, is it working now you're using alternative SATA ports?

 

[pastymuncher]

Pretty much every post you've made has said this? If there are problems during the POST, it's not going to be a PSU issue unless the PSU has given up completely, which it doesn't seem like it has.

pastymuncher, is it working now you're using alternative SATA ports?

Yes. Well sort of anyway. It cured the boot problem but the pc still goes into standby instead of shutting down. Unfortunately i have a need for 3 working ports (2 hdd, 1 DVDRW) and now i have lost Sata 0 master and slave, and the same with Sata 2, i only have Channel 1 giving me 2 out of the six ports on the board. RMA time for a board that only lasted 3-4 months. I have tried updating the chipset drivers but there are no newer ones available. Think i will try for a refund/swap and get a P45 board instead.

 

[westom]

It cured the boot problem but the pc still goes into standby instead of shutting down.

Further information that might have solved this problem long ago still cannot be posted - because those numbers were not provided.

Most people have little appreciation for how much information is obtained - in 30 seconds - if voltage numbers are taken during power up and when the system is under maximum load. So much information that a post to describe it would probably be the longest one here.

For the OP - the fact that you have no idea how important those numbers are suggests how useful that information is. Now, how many items are on your list of "definitively known good"? If that list is still empty, you have accomplished nothing after so much work and posting. By this time, you should have a list of components and subsystems that are good - without any doubt.

 

[Mattus]

Definitely time for a motherboard RMA. In a way it's good that the SATA ports are dead because if it was just the shutdown problem it'd be harder to prove that the mobo was faulty.

Further information that might have solved this problem long ago still cannot be posted - because those numbers were not provided.

Most people have little appreciation for how much information is obtained - in 30 seconds - if voltage numbers are taken during power up and when the system is under maximum load. So much information that a post to describe it would probably be the longest one here.

So you reckon that a PSU problem is making the rig go into standby instead of shutdown, despite the fact that the motherboard already has one fault, and that the PSU is coping OK whilst the rig is running?

I appreciate the importance of the PSU, but every post you make says the same thing and half the time it's nothing to do with the issue. Intuition and a process of elimination beats a multimeter most of the time.

 

[westom]

I appreciate the importance of the PSU, but every post you make says the same thing and half the time it's nothing to do with the issue.

Those failures are typical of unstable power. Since the problem can be eliminated or identified in 30 seconds - and done so without doubt, why then spend massive time and money swapping parts? Part swapping (shotgunning) has a history of even creating more failures.

Nothing is stable and makes sense in electronics until voltage stability is confirmed. Normal is to have a computer working for months with a defective supply. Strange infrequent problems that got fixed by changing things were eventually traced to an always defective from day one power system (which is more than a supply). 30 second to read meter numbers from a fully loaded system - not even disconnect a single wire. Amazing how many will fear to do that only because they never did it before.

OP's problem sounds very much like a power supply controller problem. Note how many others do not even discuss that controller - do not even know what it is. Why have I not posted a single useful reply? The next post could be extremely informative. Replies to the OP will only be as useful as the numbers he provides. No numbers means only wild speculation.

 

[Mattus]

Half of the SATA ports are dead. The computer works fine on the other ports (aside from the shutdown problem.) Do you really think that's a power issue?

I do agree that testing the 5v and the 12v with a meter is often a good idea. But on this occasion I don't think it's going to help much with pastymuncher's motherboard issues.

May I ask what you do for a living? I'm guessing some kind of electronics engineer?

 

[pastymuncher]

Further information that might have solved this problem long ago still cannot be posted - because those numbers were not provided.

Most people have little appreciation for how much information is obtained - in 30 seconds - if voltage numbers are taken during power up and when the system is under maximum load. So much information that a post to describe it would probably be the longest one here.

For the OP - the fact that you have no idea how important those numbers are suggests how useful that information is. Now, how many items are on your list of "definitively known good"? If that list is still empty, you have accomplished nothing after so much work and posting. By this time, you should have a list of components and subsystems that are good - without any doubt.

Not everyone has a multimeter you know. If you read things properly you will have seen i have tested the ram and cpu. Swapped the drives around which are only a couple of months old anyway. I know for a fact that the psu is fine. If that was dodgy then the system would be unstable, especially when overclocked and on full load. The dvdrw has been swapped just to check.

As i said before, i have dead sata ports and now the last pair have are playing up as well so it could be a problem with the ICH9R controller. It also won't shut down at all now. I have to power off by holding down the power switch on the case so i am going to RMA the board now. It appears i am not the only one to have this problem on this board. There are a couple of others who have had the same problem over at the DFI forums. Pity as it's a good board otherwise.

 

[westom]

Half of the SATA ports are dead. The computer works fine on the other ports (aside from the shutdown problem.) Do you really think that's a power issue?

It clearly could be. Remember his earlier posts referencing unexplained power modes and strange controller behavior.

Chasing strange problems in digital electronics was quickly terminated once power anomalies were identified. Digital electronics do strange things when voltages are marginal. I have literally seen machines open all valves in combinations that could never exist (in software) only because voltages were marginal.

30 seconds to confirm that input to a SATA controller is just too simple and too informative to ignore.

OP's problems have symptoms of a supply controller problem. Just another thing those numbers would report - in 30 seconds.

And no, testing only of the 5 and 12 volts is insufficient information. You did not grasp yet what those requested numbers are - further not understand what was being asked.

Why so much time arguing ‘could be this’ reasoning? Simpler is to take those numbers, actually learn something, have something on a list of “definitively good’, and therefore have accomplishments.

What is on that list of ‘definitively good’? Nothing. Nothing has been accomplished because that list and the list of ‘definitively bad’ are empty. So much discussion. So much work. And still not even one component or subsystem listed ‘definitively good’. The list of ‘unknowns’ remains long and untouched. Nothing has been accomplished.

 

[Servo]

pastymuncher

Goto power options

Then, Change advanced power options

goto power buttons and lid
change, power button to shut down, see if that helps.

Lol, I bet you tried that.

 

[PaulGreyhead]

It clearly could be. Remember his earlier posts referencing unexplained power modes and strange controller behavior.

Chasing strange problems in digital electronics was quickly terminated once power anomalies were identified. Digital electronics do strange things when voltages are marginal. I have literally seen machines open all valves in combinations that could never exist (in software) only because voltages were marginal.

30 seconds to confirm that input to a SATA controller is just too simple and too informative to ignore.

OP's problems have symptoms of a supply controller problem. Just another thing those numbers would report - in 30 seconds.

And no, testing only of the 5 and 12 volts is insufficient information. You did not grasp yet what those requested numbers are - further not understand what was being asked.

Why so much time arguing ‘could be this’ reasoning? Simpler is to take those numbers, actually learn something, have something on a list of “definitively good’, and therefore have accomplishments.

What is on that list of ‘definitively good’? Nothing. Nothing has been accomplished because that list and the list of ‘definitively bad’ are empty. So much discussion. So much work. And still not even one component or subsystem listed ‘definitively good’. The list of ‘unknowns’ remains long and untouched. Nothing has been accomplished.

Can you explain to me how a PSU can affect one SATA port and not another?
I am a PC Technician, and diagnosing a computer problem is comparable to a Doctor diagnosing an illness. One has to hear all of the symptoms before deciding what the problem could be, or where to start looking. Experience, logical thinking and an open mind are good traits to have when working with faulty PC's. Your fixation with checking voltages can lead you away from the problem cause, as it has done in this situation.
This is nothing personal, but I would also question your attitude and the way you construct your posts. I dare say it is unintentional, but your posts come accross as aggressive.

 

[pastymuncher]

pastymuncher

Goto power options

Then, Change advanced power options

goto power buttons and lid
change, power button to shut down, see if that helps.

Lol, I bet you tried that.

First thing i tried.

Applied for an RMA yesterday to the place i got it from and was approved yesterday. Sent off yesterday too. No hanging about here. They should get it today. They said that if found faulty they will send me a new one. Not exactly what i wanted but at least i won't have to reinstall both OS's. Hopefully it will last longer than this one did.

 

[pastymuncher]

Can you explain to me how a PSU can affect one SATA port and not another?
I am a PC Technician, and diagnosing a computer problem is comparable to a Doctor diagnosing an illness. One has to hear all of the symptoms before deciding what the problem could be, or where to start looking. Experience, logical thinking and an open mind are good traits to have when working with faulty PC's. Your fixation with checking voltages can lead you away from the problem cause, as it has done in this situation.
This is nothing personal, but I would also question your attitude and the way you construct your posts. I dare say it is unintentional, but your posts come accross as aggressive.

That's exactly the way i read his post's too!! No doubt Mattus came to the same conclusion. Trying to help someone is good but not when the answers/suggestions come across so aggressively and almost arrogantly.

 

[westom]

Can you explain to me how a PSU can affect one SATA port and not another?
I am a PC Technician, and diagnosing a computer problem is comparable to a Doctor diagnosing an illness. One has to hear all of the symptoms before deciding what the problem could be, or where to start looking.

So you ignore a major input to digital electronics - voltage? Actions inside a digital IC are analog. Threshold and timing changes occur with voltage and temperature changes - even variations across the chip. Not everything in a digital IC fails simultaneously. For example, one VLSI controller chip selectively changed valve setting only because voltage had dropped. Enough to start doing strange things on one port but not others.

Of course, you also know grounds between different ICs are never same. This does not apply to the OP's problem. But another analog problem that, if you have fully learned this stuff, causes analog misbehavior in digital ICs.

Before deciding what the problem could be, one must hear all the symptoms. You said that. So he ignores a symptom of SATA misbehavior - marginal voltages. Other symptoms also exist that are synonymous with margin DC voltages. Taking 30 seconds to obtain numbers is too difficult. Easier is to RMA a board using shotgun logic? That contradicts experience, logical thinking, and an open mind - your statement. To automatically blame the motherboard is classic fixation on shotgunning. Not know what has failed - so just replace something on speculation.

You mistake aggressive with significant experience and knowledge. The OP's posts suggests power problems that also include strange behavior from a supply controller. So the decision is to ignore a major input - voltage variation - that can also cause SATA port misbehavior?

I hope they trained you on how digital electronics are analog inside; as we often must do with our techs. Ever see a signal on a digital IC output appear on its input - go right through the IC? A digital IC can work normally without any connection to Vcc? These are not relevant to this OP's symptoms. But is knowledge and experience available when one understands how digital ICs work. Analog behavior of SATA ports is one reason why responsible techs routinely confirm DC voltages. Other OP posted symptoms suggested a supply controller problem.

 

[PaulGreyhead]

If you have one chip or controller controlling four or six sata connectors, how can a voltage drop affect one connection and not another? With a parallel supply each channel should receive the same power input. If it isn't, it is not a power supply unit fault but something on the board that is stopping/changing the supply.
Experience and knowledge are totally different to aggression and attitude.
I think I've had enough training thank you. It started roughly 27 years ago with an electricians apprenticeship. Worked in computer repairs since the early 90's. I certainly don't know everything, and would never claim to do so. I still learn something every day.