GPU or PSU

DChuv

Honorable
Jan 9, 2014
26
0
10,530
Hey, guys. I put together a budget desktop about a month ago (AMD FX - 6300, BioStar A960D+, 16 GB DDR3 Ram, NVIDIA GeForce GTX560, PSU Corsair 600W) At first everything was fine, but then it started to reboot or freeze after about an hour or so at first, and then every five minutes.

Seems like the overheating issue to me, but the CPU and GPU temps are fine; the PSU gets pretty hot when gaming, though. So, I replaced the gtx560 with my old Radeon HD 5700 and that took care of the problem (except the old video card is glitchy as hell and about to die, so gaming is not an option). I guess the GTX 560 needs a lot more power than the Radeon since it requires two 6-pin pci-e connectors, while HD 5700 needs only one.

So, what do you guys think? Is my new video card bad or is it the power supply? Thanks!
 
Gtx 560 doesn't consume that much power, and a faulty gpu will not roboot your pc and temps are fine as you said, do you have another psu you can try from your old pc? To me it sounds like a faulty psu, i've tried something similar where my pc shutdown while gaming and it was the psu dying, if the psu fail to give enough power the pc will shutdown/reboot and a new fixed the problem for me
 


If Gtx 560 doesn't need much more power than my old Radeon, why would the system run fine with the old gpu then? In other words, if the psu is the problem, wouldn't the pc reboot even with the old graphics card?

Unfortunately, I don't have another psu.

Do you know if there's a way to tell for sure without trying a different psu? Like some diagnostic software I could run maybe?

Thanks for the reply, btw!
 


It seems so, but the Radeon needs only 75W, while the GTX 560 needs 150W. Is there a way to tell for sure by running some software maybe?
 


Sorry, I only have one.

Btw, at first it didn't reboot; sometimes it froze, but more often it just stopped sending the signal to monitor and responding to keyboard commands or mouse movement. It looked to me like it goes into a sleep mode, so I turned the sleep mode off in windows and set "suspend mode" in BIOS to S3 (it was on S1 before) and to "repost video on S3". I don't know if it's important or not.
 
Go into the bios and tell us, the following:

+12 Volts Actual Voltage Here
+5 Volts Actual Voltage Here
+3.3 Volts Actual Voltage Here

Tell us which card is in the computer, or if you want, do it with both cards.
 


CPU Voltage: 1.212V
DDR Voltage: 1.608V
+12.0V: 11.610V
+5.00V: 5.032V
Chip Voltage: 1.224V

I tried it with both cards and the readings are identical, except that with the gtx I noticed that 12.0V ranges from 11.558V to 12.661V (I did not notice that with Radeon but it doesn't necessarily mean that the voltage didn't fluctuate; I'll check again if necessary).

Thanks a lot for your reply.
 
My local computer shop charges, $55 to check a PC and tell you which part is bad, if it is worth it to you.

The free option, if both the video card and power supply are under warranty, call both manufacturers and they will help you troubleshoot with the goal of returning the one bad product to the manufacturer under warranty.
 


It's $75 where I live. And yes, I did it. They couldn't tell me anything specific. According to the technician, everything ran just fine at his shop. When I took it back home the problem persisted...
 
Update: It happened again with the old GPU. The system rebooted after about 13 hours of proper functioning. So, does it mean that it's the psu for sure? Is there a chance that it could be the mobo, cpu or any other parts?
 
The last time it rebooted when I was gaming, after I had my old Radeon card in for 12-13 hours. However, when I have the gtx 560 in, it reboots spontaneously regardless whether I'm gaming or not.
 
I am shocked the tech could not tell you which part, I suggest taking it back and talking nicely to the manager and ask him to test the power supply under load. For the money you already paid him. Something is out of spec and it was their job to find which one.

The only other thing it could be is the motherboard, since the first 75 watts of power to the video card come though the motherboard. But normally that is not a problem unless the motherboard and the caps are old.

What you described is shockingly unstable (11.558V to 12.661V) and should be easy for a tech to troubleshoot because they have spare parts.
 


It's a small shop; I assume the tech is either the owner or related to him. He didn't charge me anything, so no hard feelings there.

I'll just wait till I get the new video card next week.Then, if the system keeps acting up, I'll return the psu and get a new one. Then, if it still doesn't help (which is highly unlikely), it's the motherboard.

 
I had a similar issue where everything was fine for 4 years but started randomly rebooting. Turns out a capacitor on the PSU wore out and when in a game, the power draw grew till it could no longer keep up and would cause a reboot. Antec did an advanced RAM and it has been fine ever since.

Please let us know if the new card works. However if it does not, I would call Corsair and try an rma. I hear Corsair's RMA department is very nice and quick.
 
Corsair 600 watt power supply is a good solid supply (I'm guessing it's a CX600 ????) Even that, their cheapest model is More than up for the job.....

Load (both cards actually draw SAME power)
- 5770 ?? draws 150watts (there is no "5700 card" its' either a 5750 or 5770
- 560 draws 150watts
NOTE: I run CX600 to run 5870's which draw 210watts & PSU's are "cold as ice..."

Condition, "Hot PSU"....
- is the FAN in the PSU running at ALL times? (if fail fails, PSU will get hot....)
- what system case to you have
- I'm guessing it MIGHT be one with a top mount power supply arrangement?
- Heat from the GPU rises and blows thru the PSU giving the impression PSU is hot but, it's not...
- what make and model of Each GPU
- I'll guess one dumps hot air into the case and the other blows it out the back of the case....