Home Built System Crashed and Freeze constantly.

daidamtac

Honorable
Dec 18, 2012
16
0
10,510
Hello tom'shardware,

I am having problem with my old/new built computer and it is driving me crazy. So hopefully some experts can help me out here.

I built this system last year and it kept freezing. The whole system freeze, no BSOD, it usually freeze when it tried to at particular point where it tried to load a screen in my game. And so i have to do hard reboot.

I suspect it is my hardware problem, so recently i bought a new motherboard thinking that it might solve the problem but apparently it can not:

Here is my built:

CPU: AMD Phenom IIX4 955
Mobo: Asus M5A97 LE R2
Graphic card: Radeon HD 5770
RAM: 4GB no brand name.
PSU: Thermaltake 430 watt
Cosair Liquid Cooling
HD: Seagate 500 GB

CPU Temp ranging between 3X Idle state - to 5X-6X on workload
GPU temp ranging from 29-30 Idle state - 5x on gaming.

Here are some methods that I used but none seem to be working:

I ran Memtest 86+ for 9 passes, 10 hour straight, 0 error => RAM can't be the issue
I ran SeaTools for all the test, all passed => HD can't be the issue
Mobo can't be the issue because it happens to the old Mobo and happen on this one too, I don't think I am that unlucky to get 2 defective units :(
The last guy is GPU, this seems to be the culprit but i am not really sure.
As I play game there is a scene where is just black out my screen and reboot. No mini_dump file, no BSOD, just crash.
So I increase my CoreEngine and Memory of my GPU using CCC and SmartDoctor. It passed that screen without issue. But that only fix the in-game problem

I also experience many crashes and freeze when computer stay idle too. Sometimes, when i just log in, it boot up to the window logo then crash, sometimes it successfully load the icons on Desktop but then still crash.

This is a clean installing window, because since i switch new mobo, i have to reinstall a fresh window, fully reformatted the HD too

I don't know if window is the issue?
I am really frustrated, it has been a week and i can not fully have my computer functioning. Someone please helps!
 
Solution
hmmm, now i think we're getting somewhere...

OK, temps are definitely not an issue here (Good to know). So using the in-place FTT and the blend test, where RAM is teste the most, make the PC crash. Try changing the RAM sticks, but first, check if the one you have are correctly placed (see if the pins on both sides of each stick are all the way towards the stick to keep it in place).

I'm not 100% sure but judging by what you reported back, this might be a faulty RAM issue, just not a common problem that memtest can detect probably. If it's not the sticks, i would think it could be the slots on the mobo, but you mentioned it's a new mobo and the same crashes happened with the old one, so that's unlikely.

Check the sticks, if they're...
Since you don't have on board graphics, isolating a graphics problem may be an issue. I am going to look at your PSU as a possible culprit here though. First, AMD recommends 450W PSU for the HD5770 you're a little low there but if the unit is good, that shouldn't be a problem. The real problem seems to be that even ThermalTake's best 430W "Smart SP-430P" is not a very good unit - reviewed here http://www.anandtech.com/show/6013/350450w-roundup-11-cheap-psus/12 I'm going to venture here that you will get far replacing your power supply with a quality 500W one.
Just my $0.02 worth
 

ajchavez91

Honorable
Dec 16, 2012
70
0
10,640
First of all, lemme tell you i'm not an expert, but from what you wrote, 5x-6xC temps on liquid cooling makes me think it could actually be a temperature problem... your processor is an AMD, and those are not so good for high temps.

Doing some research, the thermal specification for your processor is 62C, meaning core temps shouldn't reach that temp or the processor will start throttling to avoid damage, so you have to keep it at 60 or lower under load to avoid long-term damage.

Check your cooling system for any problems(i'm not familiar with water cooling, so i can't give you a better advice on this). If your CPU is overclocked, try lowering clock speeds and voltages to more stable ones and try gaming or doing whatever you do before the PC crashes. If it still does then it's not a temp problem (but still you should check if your cooling system is working as it should, temps are still high) and maybe try with a friend's graphics card and start testing, if it doesn't crash or anything then it's a graphics card issue and you should replace it (or make use of the guarantee if you still can).

Sorry i can't think of anything else, good luck.
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510


I do agree that my PSU is very low for the card. But to me, i have been running this computer for a year without much problem. The freezing and crashing just occurred recently. If it can't handle the much burden, it should die in the first place already. Don't you think.

But what you are saying might be the case too.
I will go to Frys tomorrow, and pick up a temporary one and run a test to see if crash still happen.
But if that is not the case, can you think of any thing else beside the PSU ?

Thanks.
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510


I don't think the low temp might be an issue, Cuz be4 it runs up to 80C and shut down instantly because of overheat protection, so i bought the liquid to cool it down, and on the label of the liquid it said lower temp by 30C so it makes sense.

Plus it is extremely cold here, i am freezing with heater on :fou: and i also have my case open so 50-60C is expected
 

ajchavez91

Honorable
Dec 16, 2012
70
0
10,640


Actually it diesn't make that much sense. Your processor might shut down at 80C but it doesn't go that high and stay there when you're gaming, even with stock cooler (unless bad installation or not enough thermal paste).

Well, 80 might be the actual limit at which the processor definitely shuts down, but even so, you should go by what the manufacturer says is the max temp. at 80C it might shut down, but on 62-63+ it might start damaging the processor (still not shutting it down) and if you don't solve that quick, it can end up dying on you.

You say weather is cold where you are and you're actually freezing even with the heating on, so if ambient temp is really low and you keep your case open, temps should be A LOT lower with a liquid cooling system, as ambient temperature plays an important role on the overall temps of your components. Something might be wrong with either your processor that makes it run hotter or with your cooling system that is not cooling very well. It might not actually be the 'cause of your crashing issues but i'd definitely look into it to avoid damaging the processor and end up with an definite crash...

good luck and sorry for my insistence :wahoo:
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510
Okay, i did not know low temperature can damage my cpu.

So, i did another testing game/ movie today. At boot up and idle the CPU ranging from 29-33C.
On gaming and movie at the same time it went up to 45C maximum.
Room temperature is 11C now, :( cold cold cold.

Does this temperature reasonable, alot of people said 60 is standard, but my CPU can't reach there lol, what is wrong please help.
 

ajchavez91

Honorable
Dec 16, 2012
70
0
10,640
No no, don't get me wrong, low temps don't damage your cpu, high temps do. it's just that it depends on the processor and the max temp specified by the manufacturer on how hot your processor can be without getting damaged.

For example, intel processors usually have a Tjmax temps of 105C which is a lot, and they can actually work pretty good up to 80C, but that doesn't mean it's ok to keep them at that temp and do nothing about it, 'cause their lifetime keeps decreasing the longer you let them get to high temps.

Amd processors on the other hand, are not that good when it comes to handling high temps like intel ones. that's why i told you to check your cooling system. A lot of people say 60C is standard because most of us have intel processor that reach 60-70C when under load. But for AMD, that's is a little on the edge.

Now, 45C is now more reasonable, you can even get to 55C and don't worry at all about it; the cooler, the better. Try getting a graphics card from a friend and use that one for a while, to see if it still crashes.
 

ajchavez91

Honorable
Dec 16, 2012
70
0
10,640


Good guess, but he ran all SeaTools tests and the HDD passed fine. so it can't be the hard drive. Also his PC isn't running slower, it just freezes.
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510
yeah most of the time it crash black screen then auto-boot back into window. Sometimes, in the middle of gaming it freeze completely ( the whole system ).
I just bought a PSU 500W and a cheap GPU 5450. :D
Let me test the GPU for a while, if the problem still persist then ill replace the PSU.
It got to be one of the two.
If both fail, well i don't know what to do then, built a new system LMAO :lol: :lol:

Thanks for your support, will report back soon.
 

ajchavez91

Honorable
Dec 16, 2012
70
0
10,640
Excellent, best way to find out is getting some cheaper stuff and test it lol. It's always a little complicated when there's no error code or anything lol.

Hope you find the problem. Good luck!

I'll be waiting for your report.
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510
Thanks alots for your time ajchavez.

So I installed the 5450. After I install the driver from the CD. It demands a restart.
Upon restarting it freeze with black screen, doesn't even boot into window. Okay so I hard reset, when i log in and check, it the same old Event ID 41 with no minidump.

So i let it sit there for 2 hours from last reply, play some HD Video from YT, and start to download some game
boom crash again. LoL

I am waiting on the next crash, then replace my PSU.
So the GPU isn't the problem right?

Thanks
 

ajchavez91

Honorable
Dec 16, 2012
70
0
10,640
Yeah you can discard the graphics card being the problem. The "Event ID 41" will not generate a minidump as it is only indicating that the PC was not shut down corretly or the system restarted from a critical error.

Also, you never answered my question: Is your processor/RAM overclocked?

I don't really think the PSU could be the problem... usually when the PSU fails, the system won't just freeze, it will actually just shut down on it's own (maybe reset, but that's not the case here). But if you still want to try another PSU, ask a friend to let you use his/hers; and IF you really wanna buy one, just don't go buy a $100 one to test, okay? lmao, buy a cheap Cooler Master/Thermaltake/Corsair/etc one, like $35-$40 i guess? MAYBE $50, but don't spend more than that... If that's not the problem either, then you can save it for when you need a temporary PSU.

This crashing issue is a little weird... if replacing the PSU doesn't solve it, then we must consider a driver issue or a damaged windows copy and might need to reinstall windows from scratch with a different DVD (maybe even another version like Home Premium?) and test afterwards.

P.S: An idea just came to my mind. Have you ever tried checking the Event Viewer on Windows? if you haven't, then just click Start and type: Event viewer, once you've opened it, check the first folder: Personalized views (my windows is in spanish so i don't know what it says in english) and then double-click Administrative events. There you'll find all error messages that have been generated by the system. Just ignore the ID 41 ones and check for others related to anything else that could be the cause.
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510
I haven't tried the new PSU yet, i am waiting on another crash then switch PSU.

the PSU i got is LEPA 500W brand new. 40 buck

i did alot of research and many people mention about the driver.
So i test by booting into a Live Linux (Ubuntu). It ran fine for 2 hours and crash.
So we can ignore the driver problem here.

The window also cause my attention so I tried 2 different version Win(ult) and Win(home)
same symtom.

I did take a look at Administrator events, and also Window logs. There are a few errors and warnings but they are from my game so that is no a big problem.

I am waiting on another crash but have been waiting for two hour, let me do some gaming and video to see what happen, will report back soon.

Thanks for all your time.
 

ajchavez91

Honorable
Dec 16, 2012
70
0
10,640
Lol, that made me laugh... xD! Ok, hope it's the PSU then... if not... this might have proven to be a badass problem with one of your components...

For how long did you test every component? CPU, RAM, HDD, etc.?
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510
Ok both freeze and crash with new PSU.
crash in game, and freeze on boot up.
So what is the plan now lolzzz.
Imma go to fry now to get a new HDD and 4GB RAM.
I will try anything to make this works hahahaha

Oh i did 9 hour of memtest, 0 Error.
I ran all the test in SeaTools all pass.
How do i test my CPU?

To me it runs fine.

BTW should i flash my mobo? Well i did flash to latest version be4 an it happens,
so i switch back to stock firmware.
 

ajchavez91

Honorable
Dec 16, 2012
70
0
10,640
well if you haven't tested the cpu, that might be the problem. Maybe the CPU is faulty...

You need to test the processor stability by using one of those 2 softwares. You never answered my question about your CPU being overclocked or not... is it overclocked?

Try aida64 stress test, or prime95, try leaving it over night if it doesn't crash before you go to sleep lol. don't start spending money on a new HDD and RAM, you won't solve anything by replacing components that are working just fine, be a little patient and eventually we'll find the issue.
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510
no it is not overclock,

I do not know how to set up bios so just leave everything as default.
ok, let see me try the CPU test tonight to see what happen.

will report back, thanks.
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510
something interesting happen.

I ran prime 95 stress test.
i ran the blend mode, it said test RAM alot , crash immediately.
so i ran the first test, where it said not much RAM tested.
Got couple passes, still running.
How long should i run this test?

Here is the info, 4core at 100% load, 53C
 

daidamtac

Honorable
Dec 18, 2012
16
0
10,510
Okie my report lol

Fully test prime95 for 1 hour, only small test works
MAx temp 56C. Full load on 4 core.

Immediately crash when i start blend test and middle test(forgot what it called)
So the test where it actually uses the RAM is where it fails.

I will do Memtest tonight to see what happen.
Oh forgot to mention, there is a day where it freeze during memtest. Not sure what is going on
but memtest never gives me Errors. all passes. So wierd

Let me know if you can think of something

Thanks.
 

ajchavez91

Honorable
Dec 16, 2012
70
0
10,640
hmmm, now i think we're getting somewhere...

OK, temps are definitely not an issue here (Good to know). So using the in-place FTT and the blend test, where RAM is teste the most, make the PC crash. Try changing the RAM sticks, but first, check if the one you have are correctly placed (see if the pins on both sides of each stick are all the way towards the stick to keep it in place).

I'm not 100% sure but judging by what you reported back, this might be a faulty RAM issue, just not a common problem that memtest can detect probably. If it's not the sticks, i would think it could be the slots on the mobo, but you mentioned it's a new mobo and the same crashes happened with the old one, so that's unlikely.

Check the sticks, if they're already installed correctly then change them for new or borrowed ones, test and report back.

good luck!
 
Solution

daidamtac

Honorable
Dec 18, 2012
16
0
10,510


Does this mean we are getting closer to the solution, or are we getting something else lolzz, don't scare me.

I just go back from work, and i put in the new stick. Let me test prime 95 then ill report back.

BTW this is brand new PAtriod 4GB DDR3 1600 MHZ.

Hopefully everything will go well\

Thanks