Weird readings on i5 760

Ceekayed

Honorable
Sep 10, 2013
2
0
10,510
Bear with me, I'm not a computer wiz by any means. Exactly why I'm asking here.

First the specs:

Gigabyte P55a-UD3
Intel i5 760 @ 2.8ghz
Stock cooler. I have an EVO 212 lying around but didn't feel the need to overclock so I never installed it.
12gb of some Kingston DDR3 (2x4 and 2x2, all same model)
Nvidia Geforce 460 gtx 1gb
TI firewire pci-e card with a Presonus Firebox plugged into it.
Antec Sonata III 500 (original PSU intact)

So, I've had this setup for casual gaming and more serious composing for around two years now. It's worked flawlessly, no crashing or other weirdness. However, I recently started getting some random crashes. It's not BSOD'ing, it just freezes without any warning. Windows logs have been of no help as apparently it doesnt have time to log anything prior to crashing. So I started investigating into it, reinstalled windows, made sure all hardware drivers are up-to-date and working properly and the freezes continued. So I updated bios. Still freezing. That made me suspect a hardware fault, so I ran prime95 overnight, memtests and seagate tools. No faults whatsoever, temps stay below 70c during p95, and below 40c while idle. I've yet to crash a single time when actually under heavy load like during gaming, the freezes happen when I'm just surfing on the net or just listening to music with everything else on idle.

So here are some questions I'd like an answer to:

CPU-Z is showing me clock speeds of 3.3ghz while Easytune 6 is showing me clock speeds of 2.4ghz even during prime95. As the cpu is on factory settings, shouldn't it be at ~2.8ghz? Bios indeed is showing me 2.8ghz on boot and P95 is reporting 2.9ghz. I'd understand slight variances but I'm pretty sure those readings are just waaay out there. Any clue?

My cpu core voltage is set to 1.3V, yet it's mostly hovering around 0.8-0.9V, occasionally jumping to 1.1V or so. Is this normal droop and isn't 1.3V a really high core voltage anyway? Again, on boot bios gives me readings of 1.2V.

Another thing to note, a few days before the freezes started happening, my Steelseries diablo 3 mouse started disconnecting itself randomly, as if the mouse was plugged out and back in again. Event logs were of no help there either, so I reinstalled the drivers and update firmware, yet the disconnects continued. So I gave the mouse over to my friend to test it, but he hasn't gotten around to it yet so I'm not sure if it's related at all or not. I just found out that a lot of people have been having the same kind of issues with SS mice, so I assumed it's just a faulty mouse.

So based on all of above, what piece of the hardware should I suspect, if any? The cpu readings makes me suspect that I'm in need of a new one, but I don't want to start replacing parts until I know a bit more. Can it just be the PSU?
 
Solution
A PSU beginning to go bad may cause those issues
The voltage regulation (VRM) on the motherboard may be going bad or possibly overheating, either of which may cause that
Neither the VRM nor the PSU are easily isolated to determine where the problem may lie (if in fact it is either).
I will suggest; 1) resetting your CMOS clock and 2) inspecting the motherboard for swollen or leaky capacitors or dust build up under the motherboard's heatsinks as a first course of action.

xaira

Distinguished
might be memory or board, check for shorts in any of your ports and clean them out, run memtest to verify your memory stability, run a hdd test and clean it out, its not guaranteed, but it shpould help, if not then the problem is most likely your board
 
A PSU beginning to go bad may cause those issues
The voltage regulation (VRM) on the motherboard may be going bad or possibly overheating, either of which may cause that
Neither the VRM nor the PSU are easily isolated to determine where the problem may lie (if in fact it is either).
I will suggest; 1) resetting your CMOS clock and 2) inspecting the motherboard for swollen or leaky capacitors or dust build up under the motherboard's heatsinks as a first course of action.
 
Solution

Ceekayed

Honorable
Sep 10, 2013
2
0
10,510


Thanks for the tips. I clean the dust from inside the case fairly often, but never figured I should be checking under the heatsinks aswell. There was a surprise there. Also dusted the gpu and checked all the sockets, reset cmos clock via the jumper.
I could not find any signs of overheating though.
I also removed the firewire card as I noticed it was creating some weird cpu spikes whenever I had audio playing. Moved back to my old usb audio interface for the moment.

No crashing for 2 days, so something probably helped. I'll try plugging the firewire card back once I'm sure I'm getting no crashes and see if they return, it could be just that what was faulty.

The voltage readings still jump around between 0.8 and 1.3V, so that is slightly worrysome. Other voltages report exactly what they're set to though.
Also, once I updated Easytune6, it too has started showing me idle core frequency of 3.3ghz. Bios is still claiming that it should be 2.8ghz so something is still sketchy. I'm not sure if I should be worried about that or not. System temps stay below 40c and cpu stays now well below 70c under heavy load and under 40c idle.