Power supply still not powerful enough?

Toggle sidebar Toggle sidebar
U

udprod

Distinguished
Jun 2, 2009
67
0
18,630
Hi,

I recently created a thread here, explaining how I got frequent computer crashes after being in graphic mode for 10 minutes with my ATI 4850. I was advised to switch out my power supply (I went from 250 watt to 450 watt) and stuff started working again. I was also told "Some people will say that it does not have enough amps, but it will run fine) Now, I've ran into some new problems:

1) I've played WoW for 8 hours straight. After about 7 hours of play, I started to see graphical artifacts, one hour later I got the same crash (monitor turns black, speakers go all FLAKFLAKFLAKFLAK and I have to hold the power button where it previously took me 10 minutes before crashing).

2) After playing Oblivion on Ultra High for 1 minute, I got the same crash.



Is this still a power supply issue or is it overheating? Is it possible to fix this without having to buy new hardware (I'm open for any solution)?

Thanks LOADS in advance.
 
U

udprod

Distinguished
Jun 2, 2009
67
0
18,630
By the way, this is my specs:

System tab
Operating system: Microsoft Windows XP Proffessional (5.1, build 2600)
Processor: Intel (R) Core(TM)2 Quad CPU Q9300 @ 2.50 GHz (4 CPUs)
Memory: 3070 MB RAM
DirectX Version: Direct X 9.0c (4.09.0000.0904)

Display tab
Device Name: ATI Radeon HD 4800 Series
Driver Version: 6.14.0010.6936 (Engelsk)
Driver Date: 16-03-2009 22:26:14
Driver DDI Version: ati2mtag.sys
DirectDraw Acceleration: Activated
Direct3D Acceleration: Activated
AGP Texture Acceleration: Activated
 
D

Dustpuppy

Distinguished
May 29, 2009
102
0
18,690
Was that picture taken under load or at idle? Also on the PSU does it say how many watts are under the +5/+3.3 or +12v? or did you list everything?

Regardless, artifacts generally indicate something other than the power supply though they don't rule it completely out. It does sound like overheating so if you've a desktop fan maybe point it into your computer with the case open and see what happens.


eta: a bad psu generally causes a reboot or decreased performance. Though it can also fry components if the voltage starts jumping around.
 
E

efeat

Distinguished
Jul 13, 2008
272
0
18,790


I was thinking the same thing...the OP's CPU cores are sitting anywhere from 50C-62C. If those are his/her idle temps, the load temps must be off the charts. OP, what does it say underneath the +12v1 and +12v2 ratings? Is there some kind of box or text that gives additional information? It would look similar to this http://hi-techreviews.com/index.php?option=com_content&view=article&catid=14:current-reviews&id=3710 (scroll to the bottom of the page.)
 
U

udprod

Distinguished
Jun 2, 2009
67
0
18,630
urr, I'm unsure on what load or idle means, but I'll assume that load means I was doing something, yes, I've had WoW open for 20 minutes before taking the piccies.

Under the +12v1 it says 15A, under +12v2 it says 16A.
 
E

efeat

Distinguished
Jul 13, 2008
272
0
18,790
I mean underneath those numbers even.

+12v1----+12v2
--15A-------16A--
Is there anything down here?

Also, try closing everything down (including wow) and making a new image of your CPU core temps. The only way to get usable data is to check when your CPU is at 0% use (idle) or 100% use (load.)
 
U

udprod

Distinguished
Jun 2, 2009
67
0
18,630
Idle CPU temperatures:
CPU 0: 57 degrees celcius
CPU 1: 54 degrees celcius
CPU 2: 50 degrees celcius
CPU 3: 46 degrees celcius
 
D

Dustpuppy

Distinguished
May 29, 2009
102
0
18,690


Yeap :) , but only if you got one don't run out and buy one just for this. I mentioned it because it's an easy quick fix/diagnoses if you have one sitting around.
 
G

Guest

Guest
Your Cpu does seem a tad warm for idle, but it should clock throttle when it gets warm, not crash the system, another thing, have you tried memtest on the Ram, you could have a faulty stick in there, also, how long have you had the OS installed, lets not forget it is windows after all, and a buggy os will cause these problems as well,

1) reseat your heatsink with new thermal paste 9 I take it its the stock intel cooler) and retest temps at idle and load (Use prime95 to get load whilst it is in desktop)
2) If the temps are still to great remove side panel and use a normal deskfan aimed at the side to add extra cooling and check temps, if they drop considerably then the case may be to warm, check exhaust fans on case and check Psu fan is running correctly and clean of dust
3) remove ram and test one stick at a time, this will conclude if there is a faulty stick of ram causing the probles
4) If you have a spare HDD use that to install a fresh copy of windows and re-test to see if this resolves the problem.

Unless its a really dodgy Psu which is bad luck, systematic testing is the only way forward really to get to the bottom of this.

Download Prime 95 and run it to see if the mcahine is stable running with the cpu and Ram runnng at max, run th test with the most amount of heat produced (Small ftu I think) and then run large Ftu test for max ram.

What temps is the graphics card running at, HWmonitor from the same guys that make cpuz is good for checking cpu and gpu + HDD temps whilst running machine.
 
G

Guest

Guest
just re-read the posts again, was it a trust psu, not sure of the quality of the trust psu, they do not have great efficiancy, and your quad does need stable juice to run effectivly, I think after testing you may well have to get a better psu, like corsair 650w
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom