Just something I haven’t been able to figure out. Is it possible to overwork my chipset to the point of shut down? Because ever since I upgraded my GPU to an ATI Radeon HD 4800, and my computer’s been shutting off randomly. I then realized my chipset (GeForce 8200m) was getting extremely hot, so I put a fan on it. My computer’s only shut off once ever since, but this time trying to type something in the YouTube search box. I have noticed that the new GPU has been thrashing my chipset, and now I have unbearable coil whine that worsens under any graphics operation, to the point where there’s clicking sounds, and I have noted that my computer would always shut off on one of those clicks, so I just take them as a warning sign. Another strange thing, it only cuts out on Blender and Cry of Fear. It used to also cut out on Super Wolfenstein HD, but stopped after the fan mod. My thought: I may have heat damaged my motherboard, since my case has crappy cooling (the front fan is 80mm and used to be covered by a piece of plastic until I cut it off). But what do you guys think? Is there such thing as too much work for a chipset, and is my PC boned? I need this issue resolved if I am going to be a game developer, because I am NOT getting rid of this expensive and amazing machine.
Here is my setup, if that’s any important:
M3N78-VM motherboard
ATi Radeon HD 4800
6GB 800Mhz DDR2 RAM (only 3.3 GB usable)
Maxtor DiamondMax 21 320GB SATA HDD
Seagate Barracuda 1TB SATA HDD
450Watt PSU (already confirmed not the problem)
AMD Athlon X2 2.5 GHz CPU (overclocked to 2.81 GHz)
Windows 7 32bit
And can someone explain to me why this thing dies at Blender, but can emulate a GameCube without an issue?
If I had to venture a guess. It sounds like the onboard gpu is not properly turned off in the bios and is fighting with the ATI card. The heat of both gpus + the overclock is causing the pc to overheat and turn its self off. It also might just be a bad overclock.
Currently i would declock the cpu take out A stick of ram as it is a waste but still draws power and generates heat. Make sure your on-board gpu is turned off, and get ride of the maxtor hdd. Mechanical harddrives create alot of heat also.
A quick mock up on a power simulator estimates 381w pulled from your 480w. Maybe you need a bios update for that card. make sure bios and drivers are up to date. Lastly the card might be bad, can you try that card in another system >?
personal recommendation:
build something based around a GTX 750 currently 89usd off newegg and a amd-6300 also 89usd from amazon
quick build (as of 8/9/15 total price is 341.50 usd)
6 cores 8 gig of ram 2g gpu (give your self breathing room if you want to be a game developer)
You may notice I left out an OS and storage. put 64bit Ubuntu on it and use both sata drives you have now. http://pcpartpicker.com/p/4vt3D3
This may not be your problem but it seems small to me. I’m not sure what your ‘clicking’ sound is… but if it’s the PSU switching then you might be overdriving it.
Just did a quick google, and I don’t know which of the 4800 series you have but there is this page:
…which also tends to hint at being underpowered.
Personally, for a workstation I wouldn’t put anything smaller than 600 watts in these days. You just never know.
And as another word of advice, a cheap PSU can over time eat the other components of your system so it’s worth paying for a decent one. I’ve lost stacks of hardware due to off-brand cheap PSUs.
If the “whining” noise you are talking about is related to high FPS… then turn vsynch on. You are rendering frames that never get displayed and the noise is electrical interference generated by your GPU and coming your your speakers.
The thing is, My system would still shut off even after I declocked the CPU. That is, until my little fan hack. My hard drives are cool to the touch, same with everything else except the chipset. My BIOS is up to date, but not sure about the GCard’s BIOS. And last time I checked, the onboard GPU is disabled, but I’ll check again. My drivers are up to date. And I’m planning on upgrading to Windows 10 x64 this weekend, but I’ll take the extra RAM out until then just to be safe. I was an absolute train wreck with my first time using Ubuntu, I am a Windows person. And I kind of can’t get rid of my Maxtor drive, because the Seagate one is in a dynamic format for some reason and I don’t know how to fix it. It may perhaps be a bad card, since it was given to me by a friend’s room mate. But I don’t have any other system to test it on as of now, maybe I will later.
Already did that, which is the reason why I put that extra fan over the chipset.
I don’t know which version either, no tool will tell me. But the speakers have been making noise ever since the system’s first boot. But the noise is coming from the motherboard, not the speakers. I’m planning on getting a 600W PSU eventually. And VSYNC does quiet down the coil whine by a lot, but doesn’t help the speaker interference.
My PC kept doing the exact same thing (the turning off), and only when I played Skyrim. So I thought it was overheating too, tried the desk fan and everything. The PSU broke about 2 weeks later, replaced it and everything worked fine.
My machine is actually coming up on two years old, so I highly doubt it’s because my caps are getting old. But the PSU has been a little sketchy, so I’ve been trying to replace it. Hopefully it’s just that instead of a broken motherboard. And when I first tried the HD4800, the PSU would get very very hot, until I removed a plastic shield covering half the fan. Doing that helped reduce the shut offs, but not by a lot.
That’s true. I’ll replace the PSU and hope for the best because I REALLY don’t want to replace the motherboard. The video card I’m not too worried about because I can probably find something more powerful that I can afford.
Not directly related to this case, but it is generally curious to see the marjority of people buying the cheapest psu since it does not directly increase the performance.
Yep. To a one, the thing all of my cheap PSUs have in common is that they are dead on a shelf now. I think the one that took out two GPUs and a motherboard I may have taken a sledge hammer to.
Always worth spending a little more on an Antec or something.
On the effects on performance: when the PSUs died, the system didn’t perform very well anymore. lol
P.S.: you’d be surprised how good it feels to flatten a bad component with a two handed sledge hammer.
Well I have the rule 10W/1€ minimum, worked out fine so far with a little research before buying.
So while I might have spend 60€ on a 500W supply, it is by now around 6 years old, and still working perfectly.
Btw I once had a board, that died with capaciators after 2,1 years. (Never buy noname 30€ motherboards)
I just pick parts that aren’t ungodly expensive. I also look at the reviews to see if it’s any good. And my motherboard is by ASUS, so I’m trusting it not to die prematurely.
I am getting a new graphics card this week, let’s see if the issue still persists after I try it out. I don’t know what the model is since it was supposed to be a Christmas present, but ended up at the other side of town. They just now decide to return it, and now it’s on it’s way here.
Have you checked you still have a thin layer of gel between your cpu and ventilator? My rig suffers from that problem and can stop suddenly if all ventilators aren’t working at full speed. Yup, I should correct it :D.
I know some people who use Antec PSUs in their office and only one of the four PSUs had to be replaced after 4-5 years. IMO that’s not a bad statistic.
