Sign in with
Sign up | Sign in
Your question

Same Problem under Win 7 and Ubuntu, so suspecting HW issue

Last response: in Systems
Share
April 26, 2012 3:48:52 AM

Have been running my Asus CG5290 Desktop (Core i7 920, 9 Gigs Ram - 3 2GIG DDR3 and 3 1GIG DDR3, nVidia GeForce GTX 260, 1TB Internal Drive) on Ubuntu 12.04 beta 2 64-bit for a few weeks because it had been having problems with Windows that seemed specifically timed to an nVidia driver update. In retrospect I'm glad I switched anyway, but I may have blamed Windows for a problem that wasn't it's fault.

In any event, two days ago I had some disk problems where the main drive kept going into Read-Only mode. I finally shut down the system, and opened it up to notice that the video card was in the slot nearest the processor and a couple other heat sinks. Given that I had seen continuing issues with applications freezing (gray screen and then later coming back) I suspected overheating. I went to move the card... and evidently pushed down a little hard breaking off the "latch part" of the PCI-slot. Which meant now I HAD to move the video card.

There is one other slot on the machine where the card can fit, but it isn't a perfect fit as it latches down, but appears a bit loose. There is another "regular" PCI slot right next to it, and I suspect it is just high enough to be keeping the video card from completely seating.

By going into "manual" mode I was able to reboot the machine, and had to clean up a lot of messed up inodes. The computer will work as long as it's laying on it's side (which minimizes any stress on the video card), but it's kind of a big box to leave that way. Ran Palimpsest and the HD keeps coming up clean with no S.M.A.R.T warning or any further bad sectors. Ran Memtest for several hours, and no issues there either. However last night it seemed that my screen kept turning grey every time I had to hit the network or the hard drive. So that made me suspect that even if the hard drive tests indicated it was just fine, that possibly there was still some issue. It's a 7200 rpm drive with 32 mb of cache, and so it occured to me that it could also be that perhaps there's a cache problem... or maybe even something odd with the motherboard and the built-in SATA controller. Part of the reason I keep suspecting the motherboard is that the on-board audio died about a year ago. No big deal since i got a USB external card from Creative, but given that there was no obvious reason for it to stop working...

Before I went to bed last night I turned the computer off. When I got home this evening I turned the machine on, and the video kept blinking off and then on for a minute or so, and then flick off and back on, etc. Shut it down, opened the box, confirmed nothing seemed to have sparked and the card wasn't too hot to the touch, turned it back on, and it's been rock-solid since.

I can only afford to replace one component at this point. Which one would you recommend as the best first choice? I'm thinking motherboard, then HD, then video card... but I could also argue HD, then motherboard, then video card. Or maybe... ?
April 26, 2012 11:30:27 PM

OK, so I left the machine running all day with the System Monitor on to see if I could see any overheating. Temps at the time I got home looked fine, BUT I noticed that the SB Max temp claimed to hit 255C at some point during the day. Any idea what that could be about when none of the other temps went over 65C? I'm thinking that REALLY means MB issues, but am I right to think that?
m
0
l
April 27, 2012 1:13:21 AM

255C lol.

255 is the decimal version of a 1 byte (8 bit) field of all ones. usually written x'FF'

I'd guess the temp monitoring program you are using does not correctly pick up the temp from the right location. Try a different temp monitor program.

m
0
l
Related resources
April 27, 2012 2:15:55 AM

The temp is showing fine most of the time. So while I'm willing to believe that at some point it "spazzed" out, that would still leave me curious as to why?
m
0
l
April 28, 2012 4:33:03 PM

Setup the machine with a house fan literally blowing straight into it. Brought system temps from 50s to 60s down to 40s to 50s. Let it run for 24 hours, and sure enough still had a 255C reading from the southbridge, and evidently it stopped talking to the USB port where the Wireless LAN plugs in. I suspect some of what I was seeing previously was probably the Wireless LAN port degrading as well. Reboot and everything started up just fine, which leads me to believe I've just got a bad MB chipset. So I've ordered a new LGA 1366 MB (specifically - http://www.newegg.com/Product/Product.aspx?Item=N82E168...) which also gives me some USB 3 ports. Also ordered some Arctic Silver and some Arctic Silver cleaner. My intention is to replace just the MB, and see how far that gets me. I am still suspecting the video card as well, but I couldn't afford to buy both right now, and hopefully just this change will be enough.
m
0
l
!