Jump to content
Posted 25 March 2014 - 10:03 PM
Posted 26 March 2014 - 12:00 AM
My first thought is that you may have an extra post screw under your mobo shorting it out (the screws you put into your case to screw your mobo down). Failing that it may be your cpu not having its heat sink properly placed causing it to over heat and shut down. I'd check the mobo first then if that is not it or doesn't fix it check my cpu fan. But it does sound like either a short or over heating problem.
Posted 26 March 2014 - 12:09 AM
Posted 26 March 2014 - 07:45 AM
"I tried doing memtest86 as well but can't even complete one pass on all 4 sticks trying one stick at a time in different channels lasted even less."
EDIT: disconnect your cases reset and power button etc. You can power the computer on by shorting the power button pins (it will not damage the board this is exactly what the power button does) just in case it is related to a faulty switch.
Lets start here, what version of memtest were you using? Version 5.01 has only Added preliminary support for Intel "Haswell" CPUs (Core 4th Gen). I don't recall if I have tried memtest on my current haswell (I think I have), I may do that tonight to check compatibility for you.
What exactly happens when you run memtest? Does it freeze right away? Do you get errors? We can be certain if your restarts are happening at a bios level (when you are in the bios) there is a hardware problem. The first thing I would do is disconnect every bit of hardware that is not needed, add in video card, hard drives, disc drives, etc.
Try to run a memtest with just the cpu, cpu cooler, use the onboard video, 1 stick of ram in 1 slot, and the power supply only plugged into the 24 pin and 4/8 pin connectors on the main board. If 1 stick passes, add another stick, if that passes add another, etc. The memory controller is built into the cpu. If you can never get memtest to run stable with even 1 stick it could be either a problem with memtest compatibility or cpu/mainboard/ram. You might want to double check that no cpu pins on the board were bent, (it is very easy to do accidentally). The other thing you can do if you have your old board/cpu is run a memtest on all the ram with that, if it passes there you can at least remove the ram from the equation and that leaves the cpu or mainboard since you have already swapped the psu with no success.
One last question: did you use the standouts to mount your mainboard (the standout raises the board away from the case about 1/4inch)? I know that is a silly question but I have seen people forget to do this countltess times.
I have read a few threads about people running MT on Haswell chips, it seems to work. I am not sure which version they are using but I would assume 5.01
Edited by zingo156, 27 March 2014 - 07:46 AM.
Posted 26 March 2014 - 09:03 PM
Posted 27 March 2014 - 07:48 AM
Errors in the windows event viewer are actually completely normal (critical errors and other certain errors can idicate a problem). Most are of no concern. I would actually be suprised to see a computer that had no errors in the event viewer. I am glad to hear it is up and running.
Edited by zingo156, 27 March 2014 - 07:53 AM.
0 members, 0 guests, 0 anonymous users