I recently built a budget HTPC from a Core2Duo, and Asus G41 based mobo. I've installed an ASUS HD4350 series graphics card and have twox2GB sticks of A-RAM memory, plus a 1TB WD HDD. Also a samsung DVD-ROM (HDD + DVD both SATA). I have a problem. Sometimes the computer boots fine. Other times I either get to the screen just before the first windows 7 screen (with the scrolling bar) and then the screen goes blank, or I get a BSOD, then the computer reboots.
Any advice where to start?
1) Remove GPU (completely) from the slot.
2) Re-insert GPU, position case so GPU is verticle and push straight down.
Also ... do the same with the RAM ... don't just wiggle them and push down ... pull them out and put them back in. Rotate sticks ... try running on just one stick ...
... Enter your ram timings and voltages manually ... key them into the BIOS (as documented with your RAM kit) ... Depending on your parts, it might help to boost your RAM voltages by a few hundredths. SAVE all settings entries before/upon exiting the BIOS main menu screens.
... I don'twant to jump to any rash conclusions, here, but i musta had a reason for asking, huh?
... reminds me of a few other, similar issues that i worked, in recent months ... after MUCH swapping and cajouling and thrashing about ... a drive replcmnt solved.
... it seems that a few of those drives might have intermittent data corruption issues, inside the first 90 days and makes troubleshooting an absolute nightmare !! Masquerades as many other poss fault-types ... much like marginal power or heat issues.
... too early to absolutely blame the drive ... but ... it is very high on *MY* suspect list.
= Al = (ain't it amazing how past trauma helps us remember stuff ... or forget).
If swapping the HDD is feasible, at this point, I might be tempted to try that "early" ... but you must rule out some simpler possibilities, such as marginal ram voltage (BIOS voltage settings) and poorly seated or flakey RAM modules ... first ... because that stuff is fairly quick and easy to rule out (relative to a full HDD replacement and OS load).
READ THIS VERY CAREFULLY ... ESPECIALLY ANYTHING TO DO WITH RAM ... BUT CONSIDER ALL STEPS HERE, AS POSSIBLEE CAUSES.
Firstly thanks for all your help guys. It's very comforting to have people like you out there who are eager to help!
I tried everything you suggested - removed & repl RAM, one stick at a time, raised ram voltage by the minimum increment, removed GPU and ran on system graphics, disconnected all fans except CPU, disconnected DVDROM, all to no avail. I even switched the power supply.
In the end I just did a system restore back to last week sometime, and BINGO! It worked. My system is running again. Which presents us with another issue... What was causing the problem that could be fixed by a simple rollback? Drivers?
Could it feasibly be the hard disk causing it? I made a mistake in my previous reply.. I said it was a WD GreenPower... it's not ...it's a regular Seagate. And its well more than 90 days old (but probably no more than 180days).
huh ... and I was about to start telling fortunes and reading palms.
To answer ... "Who knows?" ... But it was obviously software related and possibly HDD related.
If the prob should manifest again, with similar symptoms, then DO re-establish ALL your HDD data and power cable connections, at both ends, and start looking at new, replacement HDDs.
Set restore points frequently, until you are certain that you are out of the woods.
Might be a good idea to do a "drive-clean-up and de-frag", at this point ... just to "move and refresh all data" ... Sometimes corruption can be caused by reading and writing to the exact same spot on a drive, over and over. ... Defrag will re-locate and re-write all data on the drive so, it is advisable, once a year, or so.
prolly ... reseat/rotate your ram and boost your voltage modestly ... run prime95 (freeware) to stress it ... could also be one of the following ...
Intermittent PSU fail
Bad HDD cable or poor data cable connection
SATA DATA cable too close to power cables.
Marginal fail or intermittent fail of mobo VRM parts (capacitor, etc.)
Thermal issues (CPU Thermal Junction) under extreme load.
I hate these random/intermittent issues ... very hard to nail down ... could even be a marginal BIOS setting or a flakey driver version ...
Do the ram thing ... you can run on one module, to test (will be a bit sluggish in single ch mode).
The memory subsystem is the easiest to tweak and the most likely culprit.
I am not especially suspicious of the HDD, at this point. ...
Clear your bios and re-enter your ram timings manually ... save them and also boost the ram voltage a notch or two ... save/exit. ... rotate/reseat ram sticks ... try running on just one stick, for awhile ... then, the other.
Watch your temps (all), running Prime95, over-night.
Stay in touch ... ... ... Man ... this sort of issue just drives me batty.
Do all the stupid stuff, too ... make sure PSU pwr cord is snugly seated, etc.
... Reseat ALL cards and cables very firmly ... change power strips ... etc.
It would be good to know exactly what the cause is ... but you might fix it without ever finding out what the cause was ... just remove and reseat all cards and cables very firmly.
If there is an MRI scanner, near your system ... move your PC to another outlet/location.
= Al = Keep at it ... be methodical ... there's only 10 major parts, in there and only half of them are the potential cause..
A 63 cent wire killed 3 of our best Apollo mission specialists.
A chunk of loose foam insulation killed another seven and a poorly seated "O"-ring killed another seven ... That is 17 specialists ... killed by cheap or poorly seated parts.