Noo! Restarts of doooom!

Toggle sidebar Toggle sidebar
B

booj2600

Distinguished
May 8, 2007
15
0
18,510
Well, just put together my brand new PC, first time build. Its:

EVGA nVidia 680i SLI mobo
EVGA nVidia 8800 GTX (stock)
Intel e6700 core 2 duo (stock fan, not overclocked)
Patriot DDR2-800 4-4-4-12
Samsung Spinpoint 500gb harddrive
OCZ gameXstream 700watt PSU
Windows Vista Home Premium

I believe I put it together decently enough. Everything got booted up and running, still does. But whenever its on for a while, even just idling, it reboots. No noises, no flashes, just turns off and instantly reboots, as casually as if it was supposed to do such a thing. However, I'm not following for that act! I'm fully aware it is NOT supposed to do that! Tricky machine!

So I'm asking for your help. I don't really know where to start. Now that I think about it though.. The nVidia Monitor program, to show system temps and the like, only shows stats for my GPU. And the Control Panel only shows settings for my physical LCD monitor, which I don't think is right. But, there was brief bit where it showed more than that. For that bit, it showed all of the voltages and temps running well and within "good" limits, but if I remember correctly there was a problem with the RAM voltage. I believe it showed the voltage as 1.8v, and I believe the RAM is supposed to be 2.2v. So, advice? I just wanted to way in here before I started n00b fiddling.

Thanks guys!
 
badge

badge

Illustrious
Jun 17, 2006
13,722
0
41,460
Well, just put together my brand new PC, first time build. Its:

EVGA nVidia 680i SLI mobo
EVGA nVidia 8800 GTX (stock)
Intel e6700 core 2 duo (stock fan, not overclocked)
Patriot DDR2-800 4-4-4-12
Samsung Spinpoint 500gb harddrive
OCZ gameXstream 700watt PSU
Windows Vista Home Premium

I believe I put it together decently enough. Everything got booted up and running, still does. But whenever its on for a while, even just idling, it reboots. No noises, no flashes, just turns off and instantly reboots, as casually as if it was supposed to do such a thing. However, I'm not following for that act! I'm fully aware it is NOT supposed to do that! Tricky machine!

So I'm asking for your help. I don't really know where to start. Now that I think about it though.. The nVidia Monitor program, to show system temps and the like, only shows stats for my GPU. And the Control Panel only shows settings for my physical LCD monitor, which I don't think is right. But, there was brief bit where it showed more than that. For that bit, it showed all of the voltages and temps running well and within "good" limits, but if I remember correctly there was a problem with the RAM voltage. I believe it showed the voltage as 1.8v, and I believe the RAM is supposed to be 2.2v. So, advice? I just wanted to way in here before I started n00b fiddling.

Thanks guys!
Click to expand...

I believe it showed the voltage as 1.8v, and I believe the RAM is supposed to be 2.2v.
Click to expand...

My suggestion. Set the RAM voltage to the RAM mfg.'s specifications and your crash/reboot problem may cease.
 
M

Myopic

Distinguished
Oct 25, 2006
38
0
18,530
Are you sure your cpu temps are OK? Use speedfan or Intels TAT to monitor the temps and see what happens. Sometimes it takes a while for the heat to build up if your case isn't ventilated well. Make sure your heat sink is seated well (not loose). I'm using a stock intel HSF and its given me loads of crap even with proper installation. :)
 
B

booj2600

Distinguished
May 8, 2007
15
0
18,510
Mmkay. Changed the voltage to 2.2. Now on boot-up though it's giving me timings of 5-5-5-15, instead of the 4-4-4-12 its rated for, so I don't understand that. Not a huge deal, but still.

Also, downloaded Speedfan. Been running it for, oh, 15 min now. Measurements:

System: -65C
CPU: 53C (changes by a degree or two, but I haven't seen it go about 54. Granted, this is at idle, but that's when my computer has been crashing.)
Aux: 36C
Core 0: 45C
Core 1: 46C

That's as of right this moment. How are they holding, are they acceptable? I'll be heading down for a nap now and check up on the temps/computer to see where they are and if it has crashed when I wake up. So, wish me luck!
 
Stewartwi

Stewartwi

Distinguished
Aug 5, 2006
275
0
18,780
What is the full specs on your mobo like A1 or T1 versions? I know when I bought my board there was a lot of talk of problems when running ram voltage over 2.1 you might want to change it to 2.1 if the 2.2 didn't fix it and see how it goes.
 
P

pongrules

Distinguished
Dec 28, 2006
301
0
18,780
I have pretty much the same setup as you. I've had it for about three months now. The 680i motherboard has gone through some rough spots but they've been pretty much ironed out by now. It's still kind of a sketchy board, though as I still have issues from time to time. Mostly, it's great, but here's what I've run into and what I've done to solve it - hopefully it can help some.

On first boot, noticed that the board was detecting 5-5-5-18 timings instead of the 4-4-4-12 my memory is rated. It also detected the memory running at 1.8 volts instead of the 2.1 it's rated at. Went into the bios, chose the expert option under timings and manually set the timings and voltage. No problems for awhile, ran fine under memtest.

First problem, random restarts. Checked memtest, memory is fine. Thinking it may be a PSU issue, but decided to update to the latest bios and see what happened. Random restarts stop.

Second problem, turn on the computer - fans are on but no post, MB reads -- . Uh oh. So I cleared the CMOS and it posts. For awhile.

Third problem a little later, turn on the computer and not post again. No big deal, I clear the CMOS, turn the computer back on again and...no post. Uh oh. Try it three more times and no post. So, I take the memory out and try booting one stick at a time in the alternative memory slots. Works fine for both sticks. Both sticks placed in the blue slots now instead of the black slots, boots up just fine and has been stable for about two weeks now.

Go to Nvidia's website, they have a great community there that helped me greatly in resolving these issues.

I got this board because it had the best reviews at the time for OC'ing and was the only affordable board (shame on the Asus Striker) that could do 16x in both pci-e slots, so I thought it was going to last me awhile. Well, I don't OC currently because there's no need to yet, and there's no point really in going with SLI, so when Vista hits SP1 I'm probably going to get a different board if this one doesn't remain stable.
 
B

booj2600

Distinguished
May 8, 2007
15
0
18,510
It is an A1 version. Anything else you need for specific specs?

And what is a good length of time to run CPUBurn? I don't want to hurt my little baby! :(

So I went to bed, slept for 2 and a half-ish hours. Woke up, got worried as all of the lights were turned off, but it was in sleep mode. That was not what had happened last night when I left it on all night, it just rebooted then. This may or may not be a hopeful sign.. At least it didn't reboot!

With CPU burn running for a few minutes, no errors yet (Not that I'm entirely sure how long I have to run it before I might see some, even if the heatsink is bad... ;))..

Case: -65C
CPU: 70C
AUX: 35C
Core 0: 65C
Core 1: 65C
 
P

pongrules

Distinguished
Dec 28, 2006
301
0
18,780
You can do your system a few more favors and invest in a better heat solution. The best air cooling solutions are keeping CPU's down in the 40's under full load. I use the Scythe Infinity and the highest temp on the CPU I see is 45C. You're still within acceptable ranges, however, so that's not your problem. Just an FYI.
 
B

booj2600

Distinguished
May 8, 2007
15
0
18,510
Alright, ran CPUBurn for 20 minutes straight. Temps stayed in the range I posted above, give or take a degree or two. Computer didn't restart. So.. Yeah. It's possible I fixed it by just switching the voltage at BIOS, right?

And thank you a lot, PongRules! You do have about the same set-up, and thank you for outlining your other issues too.. Just so I know what to look for. Would it be a good idea, in your opinion (Or anyone else's?) to update to the latest BIOS? Also, I was planning on getting a better heatsink, soon as I get my next paycheck. But thanks for the suggestion!

If anyone else has any suggestions, go for it.. And should I update the BIOS? Other than that.. I suppose I'll just come back if it happens again, it's looking alright so far!

Thanks!
 
warezme

warezme

Distinguished
Dec 18, 2006
2,450
56
19,890
You can do your system a few more favors and invest in a better heat solution. The best air cooling solutions are keeping CPU's down in the 40's under full load. I use the Scythe Infinity and the highest temp on the CPU I see is 45C. You're still within acceptable ranges, however, so that's not your problem. Just an FYI.
Click to expand...

You are seeing 45C on the die not the cores, the cores will likely be in the high 50's to low 60's with that temp showing on the die. You'll know its your cores because there are two separate temperature readouts and they are not equal. One core usually runs a bit hotter than the other.

He says his is 70C which is high no matter if its die (scary) or core's, not so scary but troubling.

But yes, he DEFINITELY needs a better cooling solution.

When my 680i overheats, (has only happened once because I forgot to turn on the water pump), it just shuts the system down completely, no restarts like he is exhibiting.

If I were him I would get better cooling, and check for shorts. I had a floppy causing me grief on this board at one time, until I unplugged it.
 
P

pongrules

Distinguished
Dec 28, 2006
301
0
18,780
You sure about that? Because I read 45C on one core and like 42C on the other core under full load (I just gave the highest temp of the two). I always thought the die was a higher temp.
 
warezme

warezme

Distinguished
Dec 18, 2006
2,450
56
19,890
You sure about that? Because I read 45C on one core and like 42C on the other core under full load (I just gave the highest temp of the two). I always thought the die was a higher temp.
Click to expand...

I just noticed, you are running stock, so your temps are surely correct then. He is overclocking and will likely need to bump up his voltage so temps like yours will not be achievable especially if he's already hitting 70 8O .

My temps have hit 65c with orthos running full load at 3.7Ghz but I have backed down the speed a little as there is not much difference between 3.6 - 3.7 and my temps are cooler.
 
chuckshissle

chuckshissle

Splendid
Feb 2, 2006
4,579
0
22,780
I thought it's a cpu temps and ram issues here. Run some test on that cpu like Orthos and stress it out and see if it's going to overheat. And for the rams test with MemTest for errors.
 
P

pongrules

Distinguished
Dec 28, 2006
301
0
18,780
I was thinking T-Junction, actually. My T-Junction is around 80C, but I think that's probably normal. Hell, I don't even know what a T-Juntion is. :eek:
 
A

asdasd123123

Distinguished
Feb 16, 2006
415
0
18,790
The rams only need those 2,2 volts when running at the specified max overclocked mode. Wich it would seem you are not attempting to do.

I would attempt to run Memtest in dos mode first, and see if they produce errors.
If not I would just blame the PSU, although as others in this thread suggested, that motherboard is crap, and you might want to return it as defective.
 
P

pongrules

Distinguished
Dec 28, 2006
301
0
18,780
I don't know if I'd call the motherboard crap, but it is sketchy and I have certainly been frustrated with it on occasion. Enough that if it keeps up I will switch to the DS3. However, when it works as advertised, it's a very nice board and more people have had success with it than not. The latest BIOS's have ironed out most of the issues.
 
Surferosa

Surferosa

Distinguished
Aug 18, 2006
189
0
18,680
I don't know if I'd call the motherboard crap, but it is sketchy and I have certainly been frustrated with it on occasion. Enough that if it keeps up I will switch to the DS3. However, when it works as advertised, it's a very nice board and more people have had success with it than not. The latest BIOS's have ironed out most of the issues.
Click to expand...

What youve got to remember though is how long the P965 and Gigabyte DS3 have been on the market compared to the 680i (what revision / bios update we upto with the DS3.. 3 & bios 12 or someink?). And the DS3 wasn't such a star when it was first released- there were many problems similiar to those you have described 12 months ago.

Ive recently built rigs with DS3, 650i, and 680i. They are all good boards- it really comes down to what you want from/on them. Personally, Ive found the 680i very stable (got the similiar memory issue you described)- only thing I dislike is the wine that the northbridge fan makes. That reminds me, need to buy an after-market northbridge cooler...
 
P

pongrules

Distinguished
Dec 28, 2006
301
0
18,780
That's true. I got it because it overclocked very well, could upgrade to the quadcore and can do SLI in 2 x16. I viewed it as the most future-proof board I could get at the time. I still think it can be, I'm just worried about the memory slot issue.
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom