Random shut downs - Need help

Madrox

Distinguished
Sep 9, 2007
130
0
18,680
Hello everyone, Im having problems with my machine and i cant figure it out.

When in game my PC shuts down and its not blue screening, when it starts i dont receive any errors or "Start windows normally." It just acts like its all normal. What i have done in this order:

-Reinstalled OS
-Updated the BIOS
-Reinstalled OS
-Updated all drivers
-Bought a new PSU. The problem did not change. Did not get worse or better.
-Updated video to new AMD drivers released 7/27/11
-Swapped out the video card. I thought this was the problem after the PSU but apparently not.

Now its looking like CPU, Motherboard, Memory, or HDD.

Again its a random shut down, no warning. The system restarts and does not give any errors and does not ask to start in safe mode or start windows normally.

Any help is GREATLY appreciated. Thank you
 

puttsy

Distinguished
Aug 14, 2010
287
0
18,860
As stated, the CPU is probably over heating. Remove the cooler, clean off ALL the old thermalpaste, and apply new thermalpaste to the CPU and re-seat the cooler. If that doesn't solve the issue, try one stick of memory at a time and see where that gets you. Let us know if this solves the issue. If not, I can post more options...but they become less likely on down the line.
 

Madrox

Distinguished
Sep 9, 2007
130
0
18,680
I installed HWMonitor and had it running with speedfan and showed almost the same temp. steady 29-31c across all cores idle and 34-36c when in the BF3 aplha and GTA4 max graphics.

The funny thing is for some reason i thought about clearing the cmos, and when i did that my pc ran crash free in all games for the rest of the night and in the morning i bumped my desk and the pc shut off and it took some doing to get it to turn on again and now im back to square one.

Today i think im going to check the thermal paste and see what it looks like. i used a cheaper brand instead of arctic silver this last install. And if thats not it i will do the memory removal.

Thank you for the replies!
 

Madrox

Distinguished
Sep 9, 2007
130
0
18,680
Alrighty it looks like we are getting somewhere :pt1cable:

Ok im not sure if im allowed to post pictures on this forum but i need to show this to see i you know what this means.

Now one part i want to mention and im not sure if it matters but the right side column "CPU" went between 0,1,2,3,5 no 4

This is mid test:
midtest.jpg


It stopped and i had to power off:
haltedtest.jpg


EDIT: i did this with all sticks of memory. i just wanted to see what happened. if you think i should still do each individually let me know.
 
If you are overclocking, put it back to stock speeds and try again. I still recommend testing one stick at a time. Make sure the ram is in a slot your motherboard manual recommends for using just one stick. As of now, it appears your ram is configured wrong or you've got a bad stick of ram. Fortunately most ram warranties are lifetime.
 

Madrox

Distinguished
Sep 9, 2007
130
0
18,680
I ran the test on all 4 sticks and all ended with the same result as the second screenshot showing a halt.

I used the asus core unlocker for a day one time but thats about as far as i have ever went in overclocking.
 

Madrox

Distinguished
Sep 9, 2007
130
0
18,680
I tried the manual setting for the memory and it still came up with the same error.

il have to rma them all i spose.

Mobo - Asus m4a87td evo
Mem - G.Skill Ripjaw DDR3-1600
PSU - Antec High Current Gamer Series HCG-900 900W
GPU - XFX Radeon HD 5870

as for cpu oc i have used the asus core unlocker if that counts, nothing else.
 
G'Skill is good about their rma's. I use G'Skill and have had no problems exchanging bad sticks of ram. You'll need to return the whole kit, not just one stick, regardless if it's just one bad stick.

I still recommend testing one stick at a time though as some memory controller have problem running more than two sticks without a voltage bump.
 

Madrox

Distinguished
Sep 9, 2007
130
0
18,680
im on the third stick at the moment. this round of tests has also been individual and at the manually adjusted 1600. all are receiving the same errors.

Is there a cpu test i could run? i find it odd they all end in the exact same error at the exact same time.
 
Q - What exact set of 'G.Skill Ripjaw DDR3-1600'? I prefer a link

You already have a full 6-core, so there's nothing to unlock, if you used a dip-switch on the MOBO then place it in it's Default position or if in the BIOS Disable/Off.

Q - What other BIOS Changes?

IMO - Load Defaults (F5) and set the RAM in the BIOS manually. Once you link the RAM I'll post the settings.
 
I see F3-12800CL9D-4GBRL on AMD all of the time - edit: and they're rarely 'bad'; it could be the RAM but start with the BIOS.

RAM:
1600 9-9-9-24-2N @ 1.50v
Use Blue Slots only

BIOS:
Load Defaults
SATA Port 1-4 -> AHCI ; particularly if used before **
SATA Port 5-6 -> AHCI
DRAM Frequency -> DDR3-1600MHz
DRAM CAS -> 9
DRAM RAS to CAS -> 9
DRAM RAS PRE -> 9
DRAM RAS ACT -> 24
-
DRAM Command Rate -> 2T
-
DRAM Voltage -> 1.50v

Save & Exit = Yes

Verify DRAM LED is OFF; see manual pg 1-22

Verify Start Values = 0 ; Use Fix It or manually run REGEDIT -> http://support.microsoft.com/kb/922976
 

Madrox

Distinguished
Sep 9, 2007
130
0
18,680
I see F3-12800CL9D-4GBRL on AMD all of the time - edit: and they're rarely 'bad'; it could be the RAM but start with the BIOS.

RAM:
1600 9-9-9-24-2N @ 1.50v
Use Blue Slots only

BIOS:
Load Defaults
SATA Port 1-4 -> AHCI ; particularly if used before **
SATA Port 5-6 -> AHCI
DRAM Frequency -> DDR3-1600MHz
DRAM CAS -> 9
DRAM RAS to CAS -> 9
DRAM RAS PRE -> 9
DRAM RAS ACT -> 24
-
DRAM Command Rate -> 2T
-
DRAM Voltage -> 1.50v

Save & Exit = Yes

Verify DRAM LED is OFF; see manual pg 1-22

Verify Start Values = 0 ; Use Fix It or manually run REGEDIT -> http://support.microsoft.com/kb/922976

:( i did everything you said and still the same error on two separate sticks.

As for the core unlocker i had that on for only one day a while back. the switch is in the default position.

Again thank you guys for helping me with this problem!
 

Madrox

Distinguished
Sep 9, 2007
130
0
18,680
Alrighty i requested the RMA.

Again thank you for your help.


[strike]Im gonna try some old OCZ reaper memory and see what the test says.[/strike]
EDIT: strike that its ddr2 lol
 
CPU -> Phenom II X6 1100T BE 3.30GHz
3300 / 200 = 16.5
3300 / 240 = 13.75 -> 14 ; CPU OC 3.36GHz

MOBO -> ASUS M4A87TD EVO - http://usa.asus.com/Motherboards/AMD_AM3/M4A87TD_EVO/#specifications
4 x DIMM, Max. 16 GB, DDR3 2000(O.C.)/1600/1333/1066 ECC,Non-ECC,Un-buffered Memory

BIOS minimum = 1005 ; recommended = BIOS 2001
Phenom IIX6 1100T(HDE00ZFBK6DGR),3.3GHz,125W,rev.E0,SocketAM3,6-Core ALL 1005


RAM:
1600 9-9-9-24-2N @ 1.50v
Use Blue Slots only

Next, there are 2 ways to approach this 1. DOCP or 1. FSB OC. Rarely with a 6-core is either generally necessary, but this should 'fix' the problem. IF the long-term plan is to OC then choose method #2 Manual - FSB OC.

BIOS:
1. DOCP
AI Overclock Tuner -> DOCP
DRAM OC Profile -> DDR3-1600MHz
DRAM CAS -> 9
DRAM RAS to CAS -> 9
DRAM RAS PRE -> 9
DRAM RAS ACT -> 24
-
DRAM Command Rate -> 2T
-
DRAM Voltage -> 1.50v

2. Manual - FSB OC
AI Overclock Tuner -> Manual
CPU Ratio -> 14
CPU Bus Frequency -> 240 MHz
DRAM Frequency -> DDR3-1600MHz
DRAM CAS -> 9
DRAM RAS to CAS -> 9
DRAM RAS PRE -> 9
DRAM RAS ACT -> 24
-
DRAM Command Rate -> 2T
-
DRAM Voltage -> 1.50v