Nvidia GeForce GTS 250 1GB and system locks up while playing

Status
Not open for further replies.

CefVil

Distinguished
Jun 16, 2009
93
0
18,640
Greetings to everyone in this great community of Tom's Hardware, this is my first post so I hope I don't screw up anything.

Just a week ago I got this new computer that was sent to me already assembled by the vendor. I installed a Win7 beta and attempted to use it for gaming, but my screen is so old I needed to buy a DVI to VGA adapter to plug it into the nvidia 250 gts.

I played Gears of Wars, Quake 4 and Kane & Lynch for some days using the integrated Radeon HD 3200, with no incidents at all. Finally I got the DVI to VGA plug last Friday and used the nvidia for the first time with 185.85 drivers.

Then it happened. While playing Gears of Wars on highest and at a resolution of 1280x1024, suddenly the sound went into a loop, the game froze up completely and finally the OS locked up. I had to push restart button, I thought it was a first time error... I'm a bit of a noob on computers, and opened again GoW, from that point on I was able to play four straight hours with no errors or lock ups.

Yesterday the story kinda repeated. While playing Gears, again the sound loop came, and the system locked. This time I googled my issue and found a thread about 185.85 drivers not working properly, so I switched into 182.50 as recommended. Afterward I played Jericho on highest for some hours with no incident.

Finally today while thinking all was fine I launched Quake 4, this time I saw some weird bars traveling horizontally up and down on the screen, I panicked and closed it, then tried Kane & Lynch, it only showed some kind of white vortex and strange lines... my last option Gears of Wars ran fine.

About two hours later I played Quake 4 and the system locked up as usual. Restart, played again and I just finished the game with no further ado.

This system lock up affair seems to happen only once a day and only while playing, it has no previous choppiness or any kind of symptom before just happening.

These are the machine's specs:

Computer Type ACPI x64-based PC
Operating System Microsoft Windows 7 Ultimate
OS Service Pack [ TRIAL VERSION ]
Internet Explorer 8.0.7127.0
DirectX DirectX 11.0
Computer Name CEFVIL-PC
User Name CefVil
Logon Domain [ TRIAL VERSION ]
Date / Time 2009-06-16 / 03:04

Motherboard:
CPU Type TripleCore AMD Phenom II X3 710, 2454 MHz (13 x 189)
Motherboard Name Gigabyte GA-MA78G-UD3H (2 PCI, 3 PCI-E x1, 2 PCI-E x16, 4 DDR2 DIMM, Audio, Video, Gigabit LAN, IEEE-1394)
Motherboard Chipset AMD 780G, AMD K10
System Memory [ TRIAL VERSION ]
DIMM1: Kingston 2 GB DDR2-800 DDR2 SDRAM (6-6-6-18 @ 400 MHz) (5-5-5-15 @ 333 MHz) (4-4-4-12 @ 266 MHz)
BIOS Type Award Modular (12/30/08)
Communication Port Communications Port (COM1)
Communication Port Printer Port (LPT1)

Display:
Video Adapter NVIDIA GeForce GTS 250 (1024 MB)
Video Adapter NVIDIA GeForce GTS 250 (1024 MB)
3D Accelerator nVIDIA GeForce GTS 250
Monitor Dell E176FP [17" LCD] (CC6395BD4VKS)

Multimedia:
Audio Adapter Realtek ALC889A @ ATI SB700 - High Definition Audio Controller

Storage:
IDE Controller Standard Dual Channel PCI IDE Controller
IDE Controller Standard Dual Channel PCI IDE Controller
Storage Controller AC1QW8U0 IDE Controller
Disk Drive WDC WD5000AAVS-00ZTB0 ATA Device (465 GB, IDE)
Optical Drive TSSTcorp CDDVDW SH-S222A ATA Device (DVD+R9:16x, DVD-R9:12x, DVD+RW:22x/8x, DVD-RW:22x/6x, DVD-RAM:12x, DVD-ROM:16x, CD:48x/32x/48x DVD+RW/DVD-RW/DVD-RAM)
Optical Drive VKLONW HM38PIBOX SCSI CdRom Device
SMART Hard Disks Status OK

Partitions:
C: (NTFS) [ TRIAL VERSION ]
Total Size [ TRIAL VERSION ]

Input:
Keyboard HID Keyboard Device
Mouse HID-compliant mouse

Peripherals:
Printer Fax
Printer Microsoft XPS Document Writer
FireWire Controller Texas Instruments TSB43AB23 1394A-2000 OHCI PHY/Link-Layer Controller (PHY: TI TSB43AB23)
USB1 Controller ATI SB700 - OHCI USB Controller
USB1 Controller ATI SB700 - OHCI USB Controller
USB1 Controller ATI SB700 - OHCI USB Controller
USB1 Controller ATI SB700 - OHCI USB Controller
USB1 Controller ATI SB700 - OHCI USB Controller
USB2 Controller ATI SB700 - EHCI USB 2.0 Controller
USB2 Controller ATI SB700 - EHCI USB 2.0 Controller
USB Device USB Input Device
USB Device USB Input Device

DMI:
DMI BIOS Vendor Award Software International, Inc.
DMI BIOS Version F1
DMI System Manufacturer Gigabyte Technology Co., Ltd.
DMI System Product GA-MA780G-UD3H
DMI System Version
DMI System Serial Number [ TRIAL VERSION ]
DMI System UUID [ TRIAL VERSION ]
DMI Motherboard Manufacturer Gigabyte Technology Co., Ltd.
DMI Motherboard Product GA-MA780G-UD3H
DMI Motherboard Version x.x
DMI Motherboard Serial Number [ TRIAL VERSION ]
DMI Chassis Manufacturer Gigabyte Technology Co., Ltd.
DMI Chassis Version
DMI Chassis Serial Number [ TRIAL VERSION ]
DMI Chassis Asset Tag [ TRIAL VERSION ]
DMI Chassis Type Desktop Case
DMI Total / Free Memory Sockets 4 / 3

All powered by a 600 watt Pixxo Jaguar PW PSU with 2 fans.

Ran OCCT GPU test for about 10 minutes, the approximated time when lock ups occur and it showed no errors.

I really hope you can help me solve the problem, as the rig is new and I really don't want to go through all the nasty process of returning any kind of hardware to the vendor.

Many thanks in advance.

Last Minute Up-Date:

Now I can't play Kane & Lynch, graphics look corrupted by a vortex-like effect. Strange enough I just played Left4Dead on highest with no problem.

.C.
 

hundredislandsboy

Distinguished
Run Real Temp in the background and after a crash or playing for hours, check its log for high temps. Your problem sounds like a heat issue. What brand is the videocard?

I have a GTS 250 also and if I overclock it to 800 GPU core, it'll give me artifacts that go away if I lower the clock speed. It'll also get artifacts if the card goes to 78 - 80 C. this happened during my initia testing with the GPU only at 50%.

It doesn't happen anymore because the card never goes above 72 C with the fan set at 80% prior to starting Crysis, FC2, or CoD4.

 
@ CefVil: I'm a little confused: Is this a single card or SLI? You are showing two 1Gb graphics cards in your post.
If this is a SLI setup, remove one card and retest, because I think the power supply is the culprit here; 36A is plenty for a good unit, but I am not too sure about yours:

http://images.pcenlinea.com/show/?id=4992227cb48a4#
 

CefVil

Distinguished
Jun 16, 2009
93
0
18,640
Many thanks for the answers, now:

@HundredIslandsBoy, the card is ECS, it's not overclocked at all.

Here's what Everest says:

Video Adapter nVIDIA GeForce GTS 250
BIOS Version 62.92.6D.00.75
GPU Code Name G92GTS
PCI Device 10DE-0615 / 1019-1A0B (Rev A2)
Transistors 754 million
Process Technology 65 nm
Die Size 315 mm2
Bus Type PCI Express 2.0 x16 @ x16
Memory Size 1 GB
GPU Clock (Geometric Domain) 738 MHz (original: 740 MHz)
GPU Clock (Shader Domain) 1836 MHz (original: 1836 MHz)
RAMDAC Clock 400 MHz
Pixel Pipelines 16
TMU Per Pipeline 1
Unified Shaders 128 (v4.0)
DirectX Hardware Support DirectX v10
Pixel Fillrate 11808 MPixel/s
Texel Fillrate [ TRIAL VERSION ]

Memory Bus Properties:
Bus Type GDDR3
Bus Width 256-bit
Real Clock 999 MHz (DDR) (original: 1000 MHz)
Effective Clock 1998 MHz
Bandwidth [ TRIAL VERSION ]

Miscellaneous:
Fan Speed 41%

nVIDIA ForceWare Clocks:
2D/3D GPU: 740 MHz, Shader: 1836 MHz, Memory: 1000 MHz

How can I set the fan to 80% prior gaming and such? it always shows running at 79% after I close GoW.

@coozie7, it's a single card, I only paid for one as far as I know, SLI is disabled. Is this PSU insufficient or just too crappy?

@mousemonkey, if I can't use Real Temp on my CPU, what could I use to monitor it's temperature? How to know if the PSU is the problem?

By the way, a lock up just happened while I was listening to music, I think this is getting worse. Should I switch back to the HD 3200 IGP?

Thanks again

.C.
 
@ CefVil: Was not certain, if it was SLI then the PSU would, indeed, be a likely suspect but I think it'll be fine, unless it is faulty.
Try Everest; It's a free download and can check motherboard, CPU and graphics card temperatures and, to a degree, PSU voltages, although it gives some strange readings on my GA 965P S3 board.
Either GPUZ or Rivatuner will monitor the graphics card temperatures as well and both can record to a log for further reference.

EDIT: D*&( beaten by the Monkey;)
 

CefVil

Distinguished
Jun 16, 2009
93
0
18,640
Hi again guys, I'm now running CPUID, and it threw this:

CPUID Hardware Monitor 1.1.4.0
-----------------------------------------------------

Mainboard Vendor Gigabyte Technology Co., Ltd.
Mainboard Model GA-MA780G-UD3H (0x36A - 0x997285F6)

LPCIO
-----------------------------------------------------
Vendor ITE
Model IT8718
Vendor ID 0x90
Chip ID 0x8718
Revision ID 0x5
Config Mode I/O address 0x2E

Dump config mode register space, LDN = 0x4
0 1 2 3 4 5 6 7 8 9 A B C D E F
00 00 00 00 00 00 00 00 04 00 00 00 00 00 00 00 00
10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 87 18 05 00 1A 80 50 00 41 00 00 00 01 00 00 00
30 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60 02 28 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70 00 02 00 00 04 04 00 00 00 00 00 00 00 00 00 00


Hardware monitor
-----------------------------------------------------

ITE IT87 hardware monitor

Voltage sensor 0 1.04 Volts [0x41] (CPU VCORE)
Voltage sensor 1 1.98 Volts [0x7C] (VIN1)
Voltage sensor 2 3.31 Volts [0xCF] (+3.3V)
Voltage sensor 3 5.16 Volts [0xC0] (+5V)
Voltage sensor 4 11.97 Volts [0xBB] (+12V)
Voltage sensor 5 -12.80 Volts [0xC8] (-12V)
Voltage sensor 7 3.23 Volts [0x78] (+5V VCCH)
Voltage sensor 8 3.14 Volts [0xC4] (VBAT)
Temperature sensor 0 39°C (102°F) [0x27] (TMPIN0)
Temperature sensor 1 33°C (91°F) [0x21] (TMPIN1)
Temperature sensor 2 40°C (104°F) [0x28] (TMPIN2)
Fan sensor 0 2352 RPM [0x11F] (FANIN0)
Fan sensor 2 1266 RPM [0x215] (FANIN2)

Dump hardware monitor
LPC Register space, base address = 0x0228

0 1 2 3 4 5 6 7 8 9 A B C D E F
00 11 10 F0 00 FF FF 00 37 FF 87 54 09 27 1F FF 16
10 FF FF FF 73 D7 81 80 7F 01 FF 02 FF FF FF FF FF
20 41 7C CF C0 BB C8 FF 78 C4 27 21 28 80 23 72 72
30 FF 00 FF 00 FF 00 FF 00 FF 00 FF 00 FF 00 FF 00
40 7F 7F 7F 7F 7F 7F 5F 74 2D 40 90 22 FF FF FF FF
50 FF 2A 7F 7F 7F 50 05 05 90 05 08 12 60 00 00 00
60 00 14 41 27 90 03 FF FF 00 14 41 23 90 03 FF FF
70 00 14 41 20 90 03 FF FF FF FF FF FF FF FF FF FF
80 00 00 FF FF 00 00 FF FF 00 00 00 00 00 00 02 00
90 FF 00 00 00 FF 00 00 00 FF FF FF FF FF FF FF FF
A0 00 00 00 00 00 00 00 FF FF FF FF FF FF FF FF FF
B0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
C0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
D0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
E0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
F0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF



Hardware monitor
-----------------------------------------------------

AMD Phenom II X3 710 hardware monitor

Temperature sensor 0 29°C (83°F) [0xE6] (Core #0)
Temperature sensor 1 29°C (83°F) [0xE6] (Core #1)
Temperature sensor 2 29°C (83°F) [0xE6] (Core #2)

Dump hardware monitor



Hardware monitor
-----------------------------------------------------

GeForce GTS 250 hardware monitor

Temperature sensor 0 51°C (123°F) [0x33] (GPU Core)

Dump hardware monitor



Hardware monitor
-----------------------------------------------------

WDC WD5000AAVS-00ZTB0 hardware monitor

Temperature sensor 0 33°C (91°F) [0x21] (HDD)

Dump hardware monitor

Also I'm using EVGA Precision to increase the GPU fan speed up to 80% as default, some hours ago I made some testing on Kane & Lynch, which worked without artifacts. I saw a peak of temperature of 72 celsius degrees at around 2 pm, after the fan speed up it's working at a stable margin of 51 degrees.

Finally I uninstalled some remaining ATI drivers that I missed. I'll go and run some tests leaving GPUZ working and logging to file while on background, hope that keeps record of the temperature.

Thanks again

Edit: Played for about 10 minutes and artifacts began to appear as well as mini lock-ups of about 1 second. The temp was about 72 celsius maximum as reported by CPUID. Should my despair begin by now??

.C.
 

hundredislandsboy

Distinguished
Well, it seems your fan has some room to try and keep your card cooler. For me, 72 C is on the warm side and I prefer below 70. You can always try the fan at 100% to see if the problem goes away. If it does, you can keep lowering the fan until the artifacts come back.

I haven't heard much about ECS videocards. I use an ECS mainboard one one system and it's been reliable for over a year so I guess the lack of bad news or rumors about ECS is a good thing. I'm guessing it has a warranty in case our ideas to fix those annoying artifacts don't work out. The card seems to be working fine as you've indicated the artifacts happen only after a couple of hours of intense gaming.

I'd be interested to know what kind of rigorous torture tests ECS conducts before sending the card in a nice colorful box for the retail shelves. Maybe the guy got lazy on his shift and tested your card for 1 hour and 30 minutes only instead of 2 hours?





 

hundredislandsboy

Distinguished


Your post wasn't visible yet when I posted for 100% fan. Great minds think alike!
 

CefVil

Distinguished
Jun 16, 2009
93
0
18,640
Greetings people :D

As mousemonkey and HundredIslandsBoy suggested, I increased fan speed to 100% while playing, still temp goes up to 75 Celsius, and so far I've had no luck with Kane & Lynch which still gives artifacts after only minutes of gaming and has a very poor overall performance, I don't think a game that asks only for a 128mb card should be causing these problems on my system, because Gears of War still plays smoothly and I think system requirements on both are similar... aren't they?

Last night I played Jericho and some weird noise came out of the case, like a rattling, I'll check if the GPU is well connected and if fans work properly.

In case that the card keeps on with the problems, should it be wise to return it and get a 4850 instead?

.C.
 

hundredislandsboy

Distinguished
Well, since you had the system built by a vendor they probably didn't pull out the removeable slot covers that cover the rear part of the case. Did you pull out two of those so that the exhaust openings of the card aren't blocked?
 

DataCabbitKSW

Distinguished
Jun 8, 2009
25
0
18,540
I'll echo what MouseMonkey and others have said: make sure you turn your system fans up to 100% and make sure the exhausts and intakes aren't blocked or overloaded with dust. This will help eliminate a heat problem. Beyond that, it sounds like a power problem, so you may just not have enough juice to keep stable flow to the video card and all the rest of the system when running that add-in card. Can you get Everest or others to output a log or report of voltages and temps while trying to run a game? That could see if something starts going awry in one of those departments. Also, make sure experimental stuff like Ambient occlusion is turned off, as I have seen that cause issues in some games (especially performance wise, sometimes even causing lock-ups).
 

CefVil

Distinguished
Jun 16, 2009
93
0
18,640
Hello to everyone and a big thanks for keeping up with the thread. :D

I just opened the case and checked up the fan, which is not obstructed by anything, the card was in it's place, I pushed it a little to see if it fitted well in the slot and it did.

Then I checked up the power cable... it seemed like well connected but I unplugged it to be sure, to my surprise it was half lose from the card's entry, so I pushed til I heard a nice and clean "click". After cursing the vendor for a while I went to play Kane & Lynch, which ran fine up to half an hour with no artifacts, EVGA Precision states that max temperature was 79 celsius with fan running at 80%, and I've heard no more rattles since reconnection.

I was only reporting temps I saw after crashing, now GPUZ is running now with log activated, but as "today's crash" just happened this morning I don't expect anymore lock ups for tonight.

So far I don't know how to make a log for PSU voltages, only of GPU temps... any ideas on how to do it?

@DataCabbitKS, I'm not sure if Kane & Lynch or GoW use Ambient Occlusion, the only heavy things I see is AAF and AntiAliasing, all set to max.

Will report if another lock up happens.

.C.
 

CefVil

Distinguished
Jun 16, 2009
93
0
18,640
Greetings

After about two days of working nicely the PC froze up again.

It happened while playing Left4Dead for one hour, here are the temps registered before the lock up:


2009-06-20 20:32:08 , 740.0 , 1000.0 , 75.0 , 80 ,

2009-06-20 20:32:09 , 740.0 , 1000.0 , 75.0 , 80 ,

2009-06-20 20:32:10 , 740.0 , 1000.0 , 75.0 , 80 ,

2009-06-20 20:32:11 , 740.0 , 1000.0 , 75.0 , 80 ,

2009-06-20 20:32:12 , 740.0 , 1000.0 , 75.0 , 80 ,

2009-06-20 20:32:14 , 740.0 , 1000.0 , 75.0 , 80 ,

2009-06-20 20:32:15 , 740.0 , 1000.0 , 75.0 , 80 ,

2009-06-20 20:32:16 , 740.0 , 1000.0 , 75.0 , 80 ,

2009-06-20 20:32:17 , 740.0 , 1000.0 , 76.0 , 80 ,

2009-06-20 20:32:18 , 740.0 , 1000.0 , 76.0 , 80 ,

2009-06-20 20:32:19 , 740.0 , 1000.0 , 76.0 , 80 ,

2009-06-20 20:32:20 , 740.0 , 1000.0 , 76.0 , 80 ,

2009-06-20 20:32:21 , 740.0 , 1000.0 , 76.0 , 80 ,

2009-06-20 20:32:22 , 740.0 , 1000.0 , 76.0 , 80 ,

2009-06-20 20:32:23 , 740.0 , 1000.0 , 76.0 , 80 ,

2009-06-20 20:32:24 , 740.0 , 1000.0 , 76.0 , 80 ,

2009-06-20 20:32:25 , 740.0 , 1000.0 , 76.0 , 80 ,

2009-06-20 20:32:26 , 740.0 , 1000.0 , 73.0 , 80 ,

2009-06-20 20:32:27 , 740.0 , 1000.0 , 70.0 , 80 ,

2009-06-20 20:32:28 , 740.0 , 1000.0 , 69.0 , 80 ,

2009-06-20 20:32:29 , 740.0 , 1000.0 , 68.0 , 80 ,

2009-06-20 20:32:30 , 740.0 , 1000.0 , 68.0 , 80 ,

2009-06-20 20:32:31 , 740.0 , 1000.0 , 67.0 , 80 ,

2009-06-20 20:32:32 , 740.0 , 1000.0 , 67.0 , 80 ,

2009-06-20 20:32:33 , 740.0 , 1000.0 , 66.0 , 80 ,

2009-06-20 20:32:34 , 740.0 , 1000.0 , 66.0 , 80 ,

2009-06-20 20:32:35 , 740.0 , 1000.0 , 66.0 , 80 ,

2009-06-20 20:32:36 , 740.0 , 1000.0 , 66.0 , 80 ,

2009-06-20 20:32:37 , 740.0 , 1000.0 , 65.0 , 80 ,

2009-06-20 20:32:38 , 740.0 , 1000.0 , 65.0 , 80 ,

2009-06-20 20:32:39 , 740.0 , 1000.0 , 65.0 , 80 ,

2009-06-20 20:32:40 , 740.0 , 1000.0 , 65.0 , 80 ,

2009-06-20 20:32:41 , 740.0 , 1000.0 , 65.0 , 80 ,

2009-06-20 20:32:42 , 740.0 , 1000.0 , 64.0 , 80 ,

2009-06-20 20:32:43 , 740.0 , 1000.0 , 64.0 , 80 ,

2009-06-20 20:32:44 , 740.0 , 1000.0 , 64.0 , 80 ,

2009-06-20 20:32:45 , 740.0 , 1000.0 , 64.0 , 80 ,

2009-06-20 20:32:46 , 740.0 , 1000.0 , 63.0 , 80 ,

2009-06-20 20:32:47 , 740.0 , 1000.0 , 63.0 , 80 ,

2009-06-20 20:32:48 , 740.0 , 1000.0 , 63.0 , 80 ,

2009-06-20 20:32:49 , 740.0 , 1000.0 , 63.0 , 80 ,

2009-06-20 20:32:50 , 740.0 , 1000.0 , 63.0 , 80 ,

2009-06-20 20:32:51 , 740.0 , 1000.0 , 63.0 , 80 ,

2009-06-20 20:32:52 , 740.0 , 1000.0 , 62.0 , 80 ,

2009-06-20 20:32:53 , 740.0 , 1000.0 , 62.0 , 80 ,

2009-06-20 20:32:54 , 740.0 , 1000.0 , 62.0 , 80 ,

2009-06-20 20:32:55 , 740.0 , 1000.0 , 62.0 , 80 ,

2009-06-20 20:32:56 , 740.0 , 1000.0 , 62.0 , 80 ,

2009-06-20 20:32:57 , 740.0 , 1000.0 , 62.0 , 80 ,

2009-06-20 20:32:58 , 740.0 , 1000.0 , 61.0 , 80 ,

2009-06-20 20:32:59 , 740.0 , 1000.0 , 61.0 , 80 ,

2009-06-20 20:33:00 , 740.0 , 1000.0 , 61.0 , 80 ,

2009-06-20 20:33:01 , 740.0 , 1000.0 , 61.0 , 80 ,

2009-06-20 20:33:02 , 740.0 , 1000.0 , 61.0 , 80 ,

2009-06-20 20:33:03 , 740.0 , 1000.0 , 61.0 , 80 ,

2009-06-20 20:33:04 , 740.0 , 1000.0 , 61.0 , 80 ,

*Last two values to the right are temp and fan speed


Max was 77 degrees with fan working at 80%, however max temp was registered way before the crash.

I must mention that to benchmark I left the PC working all night yesterday, don't know if that has to do with the crash. But I think it shouldn't since I did that all time with my old Dell and never crashed.

While checking CPUID I saw a power variation of 11.97 volts maximum to 11.52 minimum, does that mean the PSU is the guilty of all this mess?

Or did I just put too much pressure onto the machine by letting it on all night and day and then playing? I'm about to write to those damn vendors...

Well, thanks to everyone again, I await for your advise.

.C.
 
Status
Not open for further replies.