Video problems (I think)

Status
Not open for further replies.

scentmental

Solid State Member
Messages
7
Okay, I've registered here in hopes that the smarter folks can figure this out for me.

I've built three or four PCs in the past, and I've recently upgraded my own PC (well I pretty much bought new everything). I've been troubleshooting to the best of my ability over the last few months, but haven't been able to solve this.

THE SYMPTOMS:

1. my screen will freeze when playing fullscreen stuff like youtube videos, farmville, and HD vimeo vids. I already tried disabling hardware acceleration and it didn't help.

2. (the worst one) my PC will just freeze, no response. If there's audio playing it will lock up that too, making a godawful racket until I push reset. The freeze usually happens when I'm playing video with quicktime or Windows Media Player, both under heavy and light system usage.

3. Crysis will run, but not for long...if my PC doesn't freeze, it will eventually give me a black screen and an error message. I'm at work right now so I don't have that specific message to post here, but it's something related to video failure.

PC SPECS:

OS: Windows XP sp3

PSU:

Chiefmax 750W 20/24 Pin PS-CM-750W-24PIN Power Supply

MAINBOARD:

GIGABYTE GA-MA785G-UD3H(rev. 1.0) Socket AM3/ AMD 785G/ DDR3/ A&V&GbE/ ATX Motherboard

CPU:

AMD Phenom II x4 965 processor

GRAPHICS CARD:

Sapphire ATI Radeon HD4870 VAPOR-X 2GB DDR5 VGA/DVI/HDMI PCI-Express Video Card

RAM:

Patriot Viper II Sector 5 4GB (2 x 2GB) 240-Pin DDR3 SDRAM DDR3 1800 (PC3 14400) Desktop Memory w/ 3DMark Vantage Model PVV34G1800LLKB

DUAL MONITORS: (my troubles preceded my second monitor accrual by several months, so I doubt it's because of the dual-mon)

Westinghouse set at 1680 x 1050

Another newer monitor (not home right now, don't know the model) set at 1920 x 1080 using an HDMI cable.



Earlier today I took the video card out and put it back in, didn't seem like I'd done it wrong the first time. If it's a compatibility issue, I'd be mildly surprised...I research compatibility before I buy parts. However, I looked over a list of compatible RAM with this particular motherboard, and did NOT see my model number (PVV34G1800LLKB). I hope that's not actually the problem though because I paid $200 for 4GB. My RAM is recognized by Windows XP, and it seems to be working. All of my drivers are current as well.

I'm about ready to give up on trying to figure this out because I've exhausted my own limited knowledge on the subject of computer hardware. I'm no novice but I'm not an expert either. I'm using my PC for 3D animation and video editing (Maya and After Effects) so I need an answer to this badly.

Good fortune times infinity to anyone that can/will attempt to help me...
 
Welcome to TF!

Your "upgrade" with all new parts would fall under a "new build" :lol:

Make sure to test temps on your CPU as well as your GPU. Use CPUID

Also, make sure your memory is in par. Run Memtest
 
I really wouldn't be surprised if you're pushing that Chiefmax PSU to it's limits. I'm not too informed on the brand, so I can't be positive of it. They're so cheap to buy, which usually means they're built cheap. I would suspect that the voltages are dropping dramatically, and causing the errors. Especially with the power hungry 4870.
 
hmm thanks for the replies. I'll try the two testing progs and consider the PSU...I haven't been too suspicious about that.

I doubt the PSU would affect the video hangs, like when a fullscreen youtube video will freeze, though the audio continues, and ESC will still take you back to normal. It COULD affect the full-on system hangs though.

I'll get back soon with the results of my tests, thanks!
 
Welcome to TF!

Your "upgrade" with all new parts would fall under a "new build" :lol:

Make sure to test temps on your CPU as well as your GPU. Use CPUID

Also, make sure your memory is in par. Run Memtest

Well here's the results of CPUID Hardware Monitor:

Hardware monitor ITE IT87
Voltage 0 1.47 Volts [0x5C] (CPU VCORE)
Voltage 1 1.60 Volts [0x64] (VIN1)
Voltage 2 3.31 Volts [0xCF] (+3.3V)
Voltage 3 5.08 Volts [0xBD] (+5V)
Voltage 4 11.84 Volts [0xB9] (+12V)
Voltage 5 -12.80 Volts [0xC8] (-12V)
Voltage 7 4.84 Volts [0xB4] (+5V VCCH)
Voltage 8 3.25 Volts [0xCB] (VBAT)
Temperature 0 51°C (123°F) [0x33] (TMPIN0)
Temperature 1 63°C (145°F) [0x3F] (TMPIN1)
Temperature 2 61°C (141°F) [0x3D] (TMPIN2)
Fan 0 5533 RPM [0x7A] (FANIN0)
Fan 2 1795 RPM [0x178] (FANIN2)

Hardware monitor AMD Phenom II X4 965
Temperature 0 62°C (142°F) [0x1ED] (Core #0)
Temperature 1 62°C (142°F) [0x1ED] (Core #1)
Temperature 3 62°C (142°F) [0x1ED] (Core #3)
Temperature 2 62°C (142°F) [0x1ED] (Core #2)

Hardware monitor Radeon HD 4870
Temperature 0 64°C (147°F) (GPU Core)

Hardware monitor ATI Radeon HD 4800 Series
Temperature 0 64°C (146°F) (GPU Core)

Hardware monitor WDC WD10EADS-00M2B0
Temperature 0 37°C (98°F) [0x25] (Assembly)

Hardware monitor WDC WD10EADS-00M2B0
Temperature 0 39°C (102°F) [0x27] (Assembly)

I'm not sure if the temperature levels are too high.

Apparently I have to burn an .ISO to use the memtest86+ utility, will get back on that one...

the help so far is appreciated!
 
Everything looks "ok". Are these idle temps? If so, the CPU seems to be running quite hot, and could be your problem. The GPU is a little warm, too warm for my taste, but safe. I believe that can get up to 100C and survive, albeit not the best thing in the world for it.

Can you run some stress tests? Prime95, Intel Burn Test, OCCT, FurMark, ect. ect.

Mainly the Prime95 and OCCT. Leave HWMonitor open, and record the highest temperatures for :

Temperature 0 51°C (123°F) [0x33] (TMPIN0)
Temperature 1 63°C (145°F) [0x3F] (TMPIN1)
Temperature 2 61°C (141°F) [0x3D] (TMPIN2)

and

Temperature 0 62°C (142°F) [0x1ED] (Core #0)
Temperature 1 62°C (142°F) [0x1ED] (Core #1)
Temperature 3 62°C (142°F) [0x1ED] (Core #3)
Temperature 2 62°C (142°F) [0x1ED] (Core #2)

Also, check the voltages, I want to know the lowest voltage reading for the +12V

Voltage 4 11.84 Volts [0xB9] (+12V)

If these are idle voltage (no programs running, just the desktop), I would be very worried. It would cause some of the heat problems too. It would cause the system to lockup like you're seeing too.
 
thank you 95BlackGA for your suggestions. I'll be back soon with results from those stress test programs.

I have a decent cooling arrangement, stock as it may be: 2 case fans (lower front pulling in and back blowing out) and of course the stock heatsink that came with my CPU. I put the back of my hand against the door side of the case often, and it's never more than warm. The Radeon card of course has a fan as well, so I wouldn't expect it to not hold its own. Regardless of the issue I should probably maybe one-up that cooling setup. N-up, maybe.

The way I see things, it's either 1) Windows-based settings, 2) faulty components, 3) BIOS settings, or 4) cooling setup.

appreciate the help!
 
Prime95 ran with no problems/errors with four bots at 8k...

Having run OCCT for an hour test, the 12v is no lower than 11.58 and no higher than 11.84, which is unstable apparently. I don't know much about electricity. Should it be lower?

What should I look for in an appropriate PSU for my system specs?

If this is part of the problem, I'll not hesitate to get a better PSU.
 
Your voltages, if they are what you say they are, are acceptable. Anything below 11.4 is dangerous. The heat is most likely the issue. What are the ambient temps in the room the computer is in? It's got a constant supply of fresh air around the case? (I.e. it's not closed up in a computer cabinet?)

EDIT : Do you have a DMM or DVM? HWMonitor doesn't seem to be very accurate at the moment. At least definitely not for me...

It said my -12V was -6.6V, measured with a DMM, and it's -12V. My +12V is reading at 12.61V, but tested at 12.1V.

At the moment I suggest a reading outside of BIOS/Windows. If you're not very comfortable poking around with the probes, read this.

http://pcsupport.about.com/od/toolsofthetrade/f/powersupplytest.htm

This could just be my motherboard not sending information properly.
 
The ATX case is standing on a hardwood floor under my desk and is always well ventilated. I had a nice multimeter for probably a decade and just gave it up because I never did anything with it. Didn't really know what I was doing. I'd need to get another one if I wanted to manually test my stuff.

I used this as a guide when I picked my PSU, and I chose it mainly based on the 750W. It seems that the Radeon HD4870 might cap out around 300W, and the CPU would only draw ~100W give or take from what I've been reading.

Well I'll be moving into a new place in a week and will have to set up my PC differently.

Tell me if this is crazy: heavy-duty power cord going from the wall socket to a power strip. Six of the eight slots are taken on that strip, one of which is ANOTHER power strip with 3/6 slots taken. That's my PC and TV setup, because this room sucks and there's only one outlet.

thanks 95BlackGA for your help. Will continue to diagnose this.
 
Status
Not open for further replies.
Back
Top Bottom