Damn Computers

[losikid]

So, I built up this bad ass high performance computer around the new year...Ok most would call it a gaming computer but I never installed a game on it. But I did have Solidworks, AutoCad, MatLab, Visual Studio, circuit board cad programs, a suite of Adobe products (photoshop, indesign, illustrator,acrobat pro, etc). I definitely overbuilt the computer for my needs. But I was sick and tired of everytime I wanted to run a game or heavy simulations on my laptop I'd start having a plethora of issues.

Anyways, halfway through my semester the solid state drive took a massive shit. Luckily all my school work was backed up on another drive.

Samsung was amazing and warrantied it quick and painlessly. But I haven't got around to installing the new drive back into my computer today because I didn't want to deal with it while classes were still in session because of the epically long process it takes to reinstall everything and do updates.

Well yesterday was my last final. So I started getting my desktop back up and running. And shit hit the fan again. While installing windows it suddenly went into the rapid on/off cycle. It would kick on for a second then kick off. Not even long enough for my fans to get to full speed.

I tried quite abit of stuff. Wiggled all the wires inside the computer to make sure everything was still plugged in, looked for any obvious issues (shorts, missing wires, etc) nothing. And it still had the booting issue after I tried all that.

After letting it sit all alone for 30 minutes, I tried again and it booted just fine. I had to restart the install process though...

Anyone know what could of caused the rapid on/off issues? The internet says anything from psu to cpu to the damn power button.

Everything is from well respected brands and from their premium product families. So I have a hard time believing shit is just failing left and right. My build is below if it helps.

ASUS ROG Maximus VII Hero Motherboard
ASUS GTX780 ROG Poseidon Graphics Card
ASUS Blu-ray reader/writer
Intel i7 4790k CPU
GSkill 16GB DDR3 RAM
CORSAIR AX860i PSU
Swiftec liquid cooling
Samsung 850 Pro SSD
WD Black HDD's
CM Storm Case

 

[R6MAXX]

i had that problem a couple years ago. turns out the mobo risers and screws were loose - asus p5b deluxe.

 

[binaryterror]

Yup check the case to mb screws and risers. Also check all the wiring to the switch from the motherboard. Another is the wiring to the PSU make sure no cuts and unplug and re plug everything in.

 

[THX_138]

I know just barely enough to get myself in trouble and my knowledge is limited, but it sounds like something in there is hunting for a - ground.

 

[BUBBA069]

Pull the power button switch plug off the mobo, jump the 2 pins with a flat blade of some kind. This simulates the momentary switch. If it stays running it's the switch. If issues stay swap the power supply. If it still persists start pulling components untill it stops. Start with all but the boot drive. Then add in cards, then ram.

Case switches can go bad. More so if the user is an aggressive button presser.

 

[MountainStorm]

It's a PITA but having built a lot of systems I developed a SOP to my TS regimen. I would remove everything and break it down into parts and examine each part and rebuild it on the bench and test it out of the case so I could try and reproduce the fault. If it seemed fine I would load test it and do a CPU burn and RAM stress test with error checking and run a drive checking utility. I would monitor the PSU voltage rails with SW or a multi while PC is loaded 100% and look for sag or droop. If everything checked out I would nut n bolt the case to make sure nothing is loose or misaligned and reinstall. If at this point the issue reappeared I would remove everything from the case and shoot that MFing case with my 12 gauge and get another case.

Just kidding about the shotgun part....although I have done that. The only thing that can be wrong about a case is a short against something or a bad switch or LED.

Like a said a real PITA but that's how I would do it.

 

[Peaker]

That seem to be good hardware.

My last problems was all RAM related, first my watercooled OCZ ram started to die one module at the time, and last year one of my 8Gb Corsair ram modules started to give me BSOD all the time.

Sadly i can not find the papers on that RAM so i cant RMA it so i just have to live with 3 X 8Gb.

I would start with the simple stuff, try to pull one RAM module at a time, and run without it for a while, thats how i isolated my faulty RAM blocks.

Since its so erly in the process you have experienced problems i doubt its GFX related, but if you have a spaer one or can borrow one it might allso be worth a try.

Things start to get more nasty when you get down to a motherbord level, i have only experienced mobo problems 1 time and that was just issues with the brand/model of RAM used and that motherbord.
But that cant be your problem as it have been running for a while without problems.
Same goes for CPUs, i never had one fail thoung i have been really naughty with those since a 700 mhz CPU was king of the hill and still not fast enuff for me ( overclocking )

I am still inclined to think its motherbord isues you have, and even my favorite brand Asus can fail, even a higher end ROG bord.

 

[STANG KILLA SS]

Ps?

 

[losikid]

Ok so here's the current situation.

Last night, I got windows to reinstall and while letting the computer just chill in it's powered on state for awhile (I didn't install anything, i left it powered on after the windows instal). It started doing the stupid rapid on/off thing. So I thought maybe it was an sd card reader I installed while I had the computer tore down to replace the ssd. So I unpluged it and the computer booted right up after I plugged it back in. As to before I could unplug it and plug it back in a minute later and it would still ahve boot issues.

Then I got graphics card drivers installed and all the mobo drivers downloaded. Instead of installing all the mobo drivers/utilities I went to bed and left the computer on. About half hour to an hour after leaving it all by its lonesome it started doing the same shit again. I pulled the plug and went to bed.

I started to think maybe the computer going to sleep was the cause. So I've been fiddling with that all morning and it won't repeat. I tried pressing the sleep button inside windows and adjusting my automatic sleep timer down to a couple minutes to see if the automatic sleep was causing it. But it's not acting up.

I'm going to put the sleep timer back up to defaults (I'd assume there is a default button) and leave it alone for awhile and see what happens.


All my screws are snug and I don't see any wire that could be causing a short. If it was shorting out, I would think I should have issues at start up. It shouldn't run fine for a while and start acting up. And I haven't really moved any of the wires since it was running just fine before the ssd crash.

I wish I had a spare mobo and psu that I could test stuff with. Because I have a feeling its the psu or mobo

I'll try disconnecting the power button next time it goes into a psychotic breakdown

 

[Sarra]

So, the first thing I would do is look at what your board has on it, plugged into it, etc.

Then, I would unplug everything, strip it down to the bare necessities.

Board, CPU, one stick of RAM, and onboard video if you have it. Plug in your OS drive, too.

Go into the BIOS, and disable everything not required to just run it.

Then, start it up and let it sit there on it's own. If it doesn't do it, start swapping RAM sticks, plug in other drives, etc., until you can get it to do it. Then you'll know what's bad, and what's good.

If you can plug everything into the board and it will run just fine, then you have a BIOS config issue.

I have built, run, repaired, destroyed, and stress tested the ever living shit out of computers since 1996. One thing I have done since the very start, and will continue to do until I am unable to use a computer anymore, is disable automated sleep, hibernation, or other 'power saving' modes. While they might save you a little electricity, I find them to be a waste of time. Issues from sleep/hibernation have existed forever. Plus, I hate the idea of thermal cycling to the point that I just turn it on, and leave it on.

If electricity is really such a big deal, then... I don't know what to tell you. I could understand if you were running a total power hog like a PentiumD, but modern core i3, i5, and i7 machines use such low power when idle, it just doesn't make sense to shut it off over night. My i5 rig uses the same amount of power over a 24 hour period of typical use that my PentiumD used during boot. I could heat an entire house in the winter with the PentiumD, the i5 doesn't even generate heat.

I'm exaggerating a bit, obviously, but why bother with a feature that isn't exactly useful?

 

[losikid]

It won't do it, thats the issue. It's so damn random I can't troubleshoot it.

It's been running all morning without a hitch. And I started running stress tests to see if I could cause it to take a shit on me.

I ran a Inten Burn test on max settings, no issues there. Now I'm running a Furmark GPU stress test and the furmark cpu burner and all is working fine. Both my cpu and gpu are running max utilization.

Temps are a couple degrees higher than last time I ran this. But its also warmer in the house now and I haven't setup fan profiles. I'm assuming the mobo saves that info so it's on the quiet profile right now, which would explain the higher temps.

My 5 volt rail hasn't dropped below 5 volts, been holding steady at 5.00. But my 3.3v has been bouncing between 3.264 and 3.280. And 12v has been staing at 12.000v.


It's went into automated sleep multiple times this morning, so I don't know if last night was some sort of freak accident or if I have a vampire that won't show its face in the morning.

 

[losikid]

The only difference right now compared to last night is that my case covers are off. But I don't see how those could be causing a short

 

[losikid]

Threads are useless without pics right?

 

[losikid]

Ok, so I just got it to go into an automated sleep while the stress tests were running.

I wiggled the mouse, it came right back on and kept stressing the gpu/cpu. There was some temp fluxations in the gpu while it was in the monitor off state (before it went to sleep)

So I think I can rule out it being a sleep issue? I'll end up turning sleep/hibernate off...but I want to rule out that being the issue.

It might be the power button, it's always been really sensitive. But I want it to happen again so i can unplug the button and see if it stops causing the rapid on/off cycles (it never even gets to the bios screen)

 

[losikid]

So, I've been trying to install the asus mobo software. Because HW-Monitor was reading a very high vcore (like 1.8xx volts). So I wanted to check it with the asus ai-suite. Make sure my mobo isn't trying to overclock the piss out of my cpu.

For some reason, the files I downloaded from asus's website won't start up. I double click on the setup.exe and/or the asussetup.exe and both will create a process but no windows pop up.

So I tried the CD instead. Using autorun, A program starts up in my bottom bar, I can hover over it and see a preview but no where on my screens is this window showing up.

So I plop the cd into my laptop, autorun boots the cd up just fine.

I'm running windows update now. Hopefully one of the many hundred updates will fix this issue.

 

[losikid]

Ok, so I just reinstalled windows 7. All the driver cd's worked like a charm everything seems to be working good.

So the install issue is fixed. Hopefully the reinstall fixed the booting issue? I read afew threads I found on google and os reinstall fixed the issue. Although all of their computers were running fine and started acting up one day. Mine had started having issues while installing the OS.

I'm going to hope this fixed the issue and continue installing stuff.

 

[losikid]

Ok, so it ran perfect all day yesterday. Got all the drivers installed, software installed, anti-virus installed...

This morning, it froze up solid while streaming a video, the sound coming out of the speakers went to an annoying two or so tones alternating back and forth. I hard Booted the computer and it froze on the screen that the password bar should be on, but no password bar. I let it sit for a minute or two and all was fine.

Couple hours later I got up, walked away. And came back. The computer was frozen again. No programs were running. It was just on the desktop. The clock showed 12:13 and my phone was telling me the time was 12:46. So it was frozen for 33 minutes. I hard booted again and it did the same freeze up procedure on the password screen. Keep in mind, the computer normally boots to a usable user interface in 10-15 seconds. But it freezes on the screen for a minute or two.

Any ideas what hardware could be causing this? This freezing issues was the exact issue I had before my SSD crashed. One time, after the hardboot the ssd became no longer recognizable. (BUT all my other drives worked fine, and i tried the ssd in multiple sata ports.

It's a fresh install of windows and since this was happening before and after my ssd failure, I'm thinking it's another piece of hardware in my computer that took out my ssd, caused the rapid start/stops, and is causing this freeze up.

I'm thinking it's the CPU or Mobo. Both could be causing a post issue and the freezing. The mobo is directly hooked up to the ssd.

I did make another observation this morning. My sata ports are 90* off the board but they sit directly underneath my graphics card. They are very cozy with the graphics card, maybe even touching. The sata cables (supplied with the mobo) have little metal spring disconnect things on top that may or may not be touching the graphics card. You can see what I'm talking about in this pic

 

[losikid]

It has froze again.

And it's not because of the graphics card. I insulated the card from the sata parts with some plastic.

The Q-code from my mobo is showing A0 which means the IDE initialization has started and according to the web that means the bios has given the OS control of the computer. And it's a normal code to be showing.

The hdd light on my computer is not lit up or flashing. So there is no signal of the cpu working right now.

So I'm thinking the cpu is taking a dump and just quits working randomly. This would explain why the mobo kept resetting during post? Or the Mobo is failing and not updating the Q-code display?

 

[losikid]

Or it could be a psu...I just wish I knew someone with a corsair axi that I could just slide in.

 

[Noll11]

if you have a spare cpu to drop in (even if it isn't very good), I'd give that a go for sure. Failing that, then PSU would be the next thing to try. Have you tried just having one stick of ram in and letting it sit like that for a while? dying ram can cause some weird issues sometimes.