Heat Issue

[Lyxtwing]

I'm having what I think are overheating issues with my graphics cards but can't figure out how to check them. I use two 7600gt cards at the moment that are only a year old and I have done no manual overclocking. As of late my computer has started to shut off randomly with no warning or slowdown before hand and I believe it is either my power supply or the cards.

Anyway, with my old XP drivers I could check the temperature of my cards, but with the new driver I can't figure out how to track it. Is there a way that I'm just missing? For that matter, could this problem even be caused by my graphics cards?

[vertex112]

I would say that it is most likely the power supply for random system crashes like the ones you are experiencing. If you want to check temps use Speedfan:

SpeedFan - Access temperature sensor in your computer


otherwise you can download a tool off the nvidia website to monitor video card temps:

NVIDIA nTune

[Lyxtwing]

Many thanks, both should come very much in handy.

[CasheKicker]

Riva Tuner is also a great utility to read the Graphics card temps and fan speeds.

Let us know what the temps are when you are get them going.


RivaTuner v2.08 download from Guru3D.com

[Lyxtwing]

It would seem not to be my graphics cards as the highest one of them goes is 55* under load.

[sswilson]

Originally Posted by Lyxtwing

It would seem not to be my graphics cards as the highest one of them goes is 55* under load.

Could still be the memory or VRMs on the vid card.

Are you overclocking your board at all? What have you got for a PSU?

What you're describing does sound heat related, but if you're overclocking the board it could be virtually anything.

[Moneyless]

If his graphics cards are overheating his PC wouldn't just randomly shut-off like that. If it's the graphics cards you'd usually see deformation/discoloration/corruption of the image, then a restart. TBH I think it sounds more like a PSU or maybe RAM issue. A couple things you could do would be: make sure the cards are seated well, you never know (or the PCI-E connectors could be corroded, it's really rare but it could happen, happened to the AGP card in my mom's Dell...); you could also download memtest+ to check your RAM; and grab a cheap/borrow a multimeter to test the PSU rails (mainly 12V).

[CasheKicker]

OK I am going to suggest something that may sound "DUH" but will ask you to physically check anyway.

What sort of cooling is on your GPU? Stock? Is it the style that has the exhust port for the fan that exhausts out the rear slot of your case? If so is the metal tab removed from the back of the case in the second slot below the PCIe slot to allow venting?

I only ask because I have seen a case where a reputable computer store actually installed an X850XT which uses 2 slots, one for the card to plug in and the next one down to vent the cooler, with the tab still in place, there by creating back pressure on the cooler and not allowing the heated air to excape, causing GPU over temp conditions. See my story Case Cooling and Airflow - How much does it matter

[Lyxtwing]

Nothing is overclocked shy of the factory overclock that came on the cards. I'm thinking it is the PSU as it's just a 450watt that came with the case. Was planning on upgrading it in the near future but looks like that may be bumped ahead. I'll check the other ideas as well, thanks for all the help.