Odd freezing during games.

BlueDestiny

I've asked before in H/A but to no avail, so I'll try here. When I play a game on my computer, after a while it will start to freeze and unfreeze, either doing that until the whole system crashes and reboots or just severely lowering the framerate. The games that this has happened in are: Crysis, Dawn of War, Garry's Mod, Trackmania Nations Forever, and Titan Quest, along with a couple of small indy games. I've run two seperate full virus scans and a spyware scan, with everything coming up clean. I've used a registry cleaning utility to no avail, but I think it made TQ work a little better. I've cleaned off and reapplied thermal paste to my CPU and heatsink, dusted my tower and gotten new drivers for my soundcard and video card. I used a SMART scan of my HDD with everything fine, defragmented and used checkdisk. All of which has done exactly nothing, and the problem seems to have only gotten worse. My current rig is as follows:

CPU: AMD Athlon 64 3700+
Mobo: ECS KN1 Lite
RAM: Corsair XMS 2GB (Brand new, got it after problem started.)
HDD: Western Digital 80GB SATA
Video: GeForce 7600GT 256mb
Sound: Soundblaster Audigy
PS: Antec 430W w/3 12volt rails

fightinfilipino

could be your video card is overheating...

try reseating your ram, any expansion cards in your system, and all of the cables first too. sometimes these things are caused by simple bad connections.

Neaky

Remove the sound card and see if the problem continues to happen.

BlueDestiny

Thanks guys, I'll try testing my video card with a benchmarking program.

BlueDestiny

Well, no luck with the benchmark. I don't think the video card is the problem, mostly because it ran Crysis for a good hour at least before the problems cropped up while Dawn of War never makes it past 5:45. Trackmania runs for a pretty long time, but eventually it too locks up and the game tells me my framerate is too low, after unfreezing.

BlueDestiny

Sorry for the triple post, but I think my video card is fried, at least partially. Several of the little cylinders sticking out of the board next to the heatsink are opened, with some kind of brownish-orangish stain around the openings on top. The heatsink probably got clogged with dust at some point without me noticing, so I guess I'll have to get a new card.

Idoliside

BlueDestiny wrote: »

Sorry for the triple post, but I think my video card is fried, at least partially. Several of the little cylinders sticking out of the board next to the heatsink are opened, with some kind of brownish-orangish stain around the openings on top. The heatsink probably got clogged with dust at some point without me noticing, so I guess I'll have to get a new card.

Check your power supply also. Dodgy power supplys can cause alot of problems similar to what you've been having.

BlueDestiny

My first PS failed a while ago, so my current one is still pretty new. I guess I'll post what SpeedFan sees for the voltage, since I don't really know what to look for. These are all while idling/looking at webpages:

Vcore1: 1.36V
Vcore2: 3.26V

+3.3V: 2.51V
+5V: 4.95V
+12V: 5.95V
-12V: -10.88V
-5V: 0.58V
+5V: 4.92V
Vbat: 3.07V

I probably should have said that before the problems happened I heard a fairly loud popping noise coming from my tower, which was probably the cylinders on my video card opening from the heat.

EDIT: Now I just need to decide on a 7900GS or an 8600GT...

Mordrack

I really hope that isn't your actual +12V rail voltage.

BlueDestiny

Mordrack wrote: »

I really hope that isn't your actual +12V rail voltage.

I'm not well versed in power supplies, but it's far newer than the video card (only a few months old) so I'm going to guess it isn't at fault. Also, I just learned that the cylinders I was mentioning before were the capacitors on my GPU. And a few are blown open and browned.

Transparent

I think you've found the problem... Still under warranty? Also, those voltages are crazy wrong. Your system shouldn't be bootable if -5 and +12 are correct.

BlueDestiny

Sadly no. I'm pretty loaded with spending cash though so a hundred or so isn't a big problem for a replacement. I think that low voltage reading was caused by the blown out capacitors which reduced the power of the video card due to overheating caused by dust which then caused it to seize up during heavy load. Sweet! Problem solved, thanks MSTT!

Transparent

All you know is that the capacitors on your video card blew. The video card's bad now, but that doesn't rule out the power supply IMO.

zanetheinsane

Also worth checking out: http://en.wikipedia.org/wiki/Capacitor_plague

While failing capacitors typically result in the aforementioned system instabilities, occasionally, failed capacitors will lead to a failure of the voltage regulators on the motherboard. There are two common theories on why this happens.

Replace motherboard with video card in the quote.

Transparent

I knew there could be collateral damage, I've seen so many power supplies fail I don't trust them, but it was an uninformed assumption. The voltage regulator makes sense. At any rate, I wouldn't consider this computer fixed until the voltages are +/- 10 percent of what they should be.

El Guaco

Transparent wrote: »

I knew there could be collateral damage, I've seen so many power supplies fail I don't trust them, but it was an uninformed assumption. The voltage regulator makes sense. At any rate, I wouldn't consider this computer fixed until the voltages are +/- 10 percent of what they should be.

I'll second that. Even if your vid card had problems, it still ran. The vid card could not possibly cause enough loading (resistance) problems to drag down the voltage on ALL of the rails on your power supply. Most likely, you have a bad power supply that ruined your video card! I think the busted vid card is a symptom of a greater problem. I think if you put a new vid card in there without testing your power supply with a crappy $10 vid card in there, you take the risk of ruining more stuff.

BlueDestiny

El Guaco wrote: »

Transparent wrote: »

I knew there could be collateral damage, I've seen so many power supplies fail I don't trust them, but it was an uninformed assumption. The voltage regulator makes sense. At any rate, I wouldn't consider this computer fixed until the voltages are +/- 10 percent of what they should be.

I'll second that. Even if your vid card had problems, it still ran. The vid card could not possibly cause enough loading (resistance) problems to drag down the voltage on ALL of the rails on your power supply. Most likely, you have a bad power supply that ruined your video card! I think the busted vid card is a symptom of a greater problem. I think if you put a new vid card in there without testing your power supply with a crappy $10 vid card in there, you take the risk of ruining more stuff.

Using a calculator showed that most of lines are within 10%, barring the few obvious ones. I currently have this PSU installed on the first or second of November from last year, so it's about 5.5 months old now. In any case, the video card and PSU will be under RMA warranty from newegg, so even if it the PSU is the real culprit I can get both replaced. Thanks for pointing that out, though.

Dixon

I haven't read all the posts but he cylinders are the capacitors that regulate the voltage and current through the card. If they are blown then either the bus for the card was taking in to much power or the PSU is not running properly. If the capacitors are blown on the vid card you will want to get the RMA started as it will only get worse. Check your motherboard as well to see if any if the Caps on there are blown. That will give you some idea as to whether or not the PSU is the problem or the Bus it self. I don't know if you have a volt meter to check the numbers but that would be something good to have. Usually with blown caps, the PSU is the problem or the bios is set to draw more power then it can handle.

BlueDestiny

Visual inspection of the mobo shows that all the caps are fine. My wireless internet card is also fine, along with my sound card. If the PSU is fine, would newegg just send it back to me after inspecting it?

EDIT: Alright I've decided to RMA that power supply. If this plus a new GPU doesn't fix it, then I guess it's time for a new computer.

Transparent

BlueDestiny wrote: »

El Guaco wrote: »

Transparent wrote: »

I knew there could be collateral damage, I've seen so many power supplies fail I don't trust them, but it was an uninformed assumption. The voltage regulator makes sense. At any rate, I wouldn't consider this computer fixed until the voltages are +/- 10 percent of what they should be.

I'll second that. Even if your vid card had problems, it still ran. The vid card could not possibly cause enough loading (resistance) problems to drag down the voltage on ALL of the rails on your power supply. Most likely, you have a bad power supply that ruined your video card! I think the busted vid card is a symptom of a greater problem. I think if you put a new vid card in there without testing your power supply with a crappy $10 vid card in there, you take the risk of ruining more stuff.

Using a calculator showed that most of lines are within 10%, barring the few obvious ones. I currently have this PSU installed on the first or second of November from last year, so it's about 5.5 months old now. In any case, the video card and PSU will be under RMA warranty from newegg, so even if it the PSU is the real culprit I can get both replaced. Thanks for pointing that out, though.

With blown capacitors the entire electrical chain is now suspect. I certainly hope it's just the video card and not a bad power supply and / or motherboard for your sake, but unfortuantely there is the posablility they all got damaged.

I have the same power supply, and one thing to note is that 430W is only if you're maxing all three rails, they're only 16amps each. Your system's within specs for just having the ATX and ATX 12v rails hooked up though (I'm assuming your 7600gt doesn't have a PCIe power connector to put it on its own power rail, and that you have all 24 pins plugged in to the ATX connector even though the little clip doesn't quite line up with the extra 4 pins).

BlueDestiny

Once my card and replacement PSU come in, I'll keep an eye on the voltage with speedfan for a while. If it's still messed up, then I guess I just have to bite the bullet and get a new mobo. Ugh, at least I'm becoming skilled at minor computer maintenance. And being forced to upgrade would get me out of socket 939 and into a dual core or something.

Idoliside

Keep in mind, Speedfan is notorius for showing incorrect voltages on the 12v rail. Ive had three PSU's in this computer, and all of them showed my 12v rail to have 7v going through it.

I recently had freezing problems with a second hard drive being installed. Replaced my Q-Tech PSU with a PowerX one (same wattage) and the latter seems to be the most reliable.

Transparent

Yeah, see if the numbers your BIOS setup reports are more normal, the number speedfan is giving you is so low I don't believe you'd be able to boot into windows.

BlueDestiny

Transparent wrote: »

Yeah, see if the numbers your BIOS setup reports are more normal, the number speedfan is giving you is so low I don't believe you'd be able to boot into windows.

Well, I already RMA'd that PSU, so I'm out of luck there. In any case, Newegg accepted the thing so I'll get a replacement back in a few days,probably on saturday/friday. Never hurts to play it safe.