Having some issues and need some avice

[jhowell]

Basically I just purchased a new processor
http://www.newegg.com/Product/Product.aspx?Item=N82E16819103772

new graphics http://www.newegg.com/Product/Product.aspx?Item=N82E16814102730

and a 500w psu saying 2 12v rails @ 17 amps

I first purchased the cpu and gpu, then upon loading games like borderlands UT3 ect, the game would run kindy sh#%@ then smooth out and look great, then comp would freeze and i would have to hard restart. It never blue screened, never shut itself down.

So I thought lack of power for ma baby, i had an older 400w powersupply.

Went to fry's, landed the new PSU, now it seems that the computer runs faster no doubt, i'll load games in no time, but 100% of the time, once the game comes to the normal environment of the game... i.e. not the loading cinematics ect, it'll shut itself down. Seems typical of a PSU that is overpower.

Soon as it powers down, i dont get a light from the mobo, and have to unplug the psu and wait a few minutes, plug it in... switch it on. As you can see by this short novel i wrote, it runs w.o a graphics load just fine. I want to believe its the GPU however, the thing was running games earlier today until the new psu.

If anyone is familiar with any issues, please let me know. Im hella frustrated =/

 

[skora]

What mobo do you have. Sounds like what youre experiencing is the CPU overheating and you're not putting a real load on it till trying to load a game. To confirm this, download prime95 and run it, see if it locks up with just a cpu load, no gpu load.

 

[jhowell]

What mobo do you have. Sounds like what youre experiencing is the CPU overheating and you're not putting a real load on it till trying to load a game. To confirm this, download prime95 and run it, see if it locks up with just a cpu load, no gpu load.

I thought overheating too, just added the stick amd cooler back to it to the same avail. Ptime95 did shut it down, however it shuts down within minutes of being turned on anyway so it's pretty well a mute test.

Mobo is a msi-7312. When I flash the mobo to newest, i am not 100% it's fully completing. CPUid is saying it's Still the old bios , however now it recognizes both cores. I usually can fix this kind of stuff but this is unlike any issue I have seen, has symptoms of overheating but is running 32c.

Could the processor just be bad? Seems like it wouldn't even boot. What about thermal grease? Can to much be used? I use the same amount each time and never had an issue b4.

Anyway any help is huge.

 

[jhowell]

And srry for spelling I am having yo respond via iPhone lol

 

[jhowell]

Also I am going to try to run prime without the CPU in it off mobo VGA, think it well make a difference?

 

[jhowell]

The mobo gpu prime test lasted longer, got to test three but shutdown again, do I know it's either powersupply or CPU, I'd there a program I can use to watch CPU temp? Right Now ac in apartment is on65 with the case open lol, I am running out of options

 

[skora]

That CPU isn't supported by that board.
http://www.msi.com/index.php?func=prodcpu2&prod_no=1104&maincat_no=1&orderby=cpu_name%20ASC#menu

That cpu model is ADX6000IAA6CZ. Its the older 125w flavor. The board will only support the newer ADA6000IAA6CZ. One letter can make all the difference. You're trying to pull too much wattage through the board and its probably what's shutting down to protect itself from overvolting.

Either a new CPU or new Mobo would be the way to go. If you can send that AGP GPU back, I'd get a new mobo and a more powerful GPU for less than what you spent on the 3850 AGP. If you can send the CPU back too, the Athlon II x2 240 is the same price and much more powerful.

 

[jhowell]

That CPU isn't supported by that board.
http://www.msi.com/index.php?func=prodcpu2&prod_no=1104&maincat_no=1&orderby=cpu_name%20ASC#menu

That cpu model is ADX6000IAA6CZ. Its the older 125w flavor. The board will only support the newer ADA6000IAA6CZ. One letter can make all the difference. You're trying to pull too much wattage through the board and its probably what's shutting down to protect itself from overvolting.

Either a new CPU or new Mobo would be the way to go. If you can send that AGP GPU back, I'd get a new mobo and a more powerful GPU for less than what you spent on the 3850 AGP. If you can send the CPU back too, the Athlon II x2 240 is the same price and much more powerful.

I wanna preface this by saying I think what you are saying is legit, and has a sound logic behind it, however it doesn't explain the random shutdowns. =/ Which is really what i am after..

I just put my old processor back into the machine, which I have had in there for I would say a good 2 years, and i get the same result. It HAS to be the mobo the powersupply or a flippin freak of nature kind of thing.

It is shutting down in about the same time frame as with the new processor, and I know the current old one i have is compatible.

After the random shutdowns, I go straight into the BIOS and check temp, each time im getting 35c so I think heat is out of the issue. What sounds more likely? The new powersupply being a lemon or the mobo suddenly going out once I added the hardware?

Anything helps, and I thank you for your help thus far. I think once i get this issue resolved, i will still have issues with the new processor, but what can ya do, maybe i can just at least get the old processor to work

 

[jhowell]

Oh and if it were in fact the mobo that is the issue here, isn't it safe to say that it would even boot up? Comp goes into windows and I can do some things, then it crashes, when it does crash, I have to unplug the power cable from the PSU and flip it to off, and leave it like that for a good 2-3 minutes before... If i just try to turn it on, nothing happens.

I.e. after this crash issue happens, there is no power in the mobo. If that helps any

 

[skora]

To monitor temps, I'd use HWmonitor. It gives more than just the CPU temps and maybe something else will show that can point to one thing or another.

Since the board has integrated graphics and the old CPU in it:

1 pull the 3850 and see if it reproduces the same issue using the IGP. Take that out of the equation.
2. If it still does the same thing, use the old PSU with the integrated graphics and see how that responds.
3. If all these combinations fail in the same manner, the mobo is at fault. I'd try to reflash the bios to the former version, and test the system with the original parts. If it goes away, then the board is fine, but the bios update failed. Try to update again and see if the new parts work that time.

 

[jhowell]

To monitor temps, I'd use HWmonitor. It gives more than just the CPU temps and maybe something else will show that can point to one thing or another.

Since the board has integrated graphics and the old CPU in it:

1 pull the 3850 and see if it reproduces the same issue using the IGP. Take that out of the equation.
2. If it still does the same thing, use the old PSU with the integrated graphics and see how that responds.
3. If all these combinations fail in the same manner, the mobo is at fault. I'd try to reflash the bios to the former version, and test the system with the original parts. If it goes away, then the board is fine, but the bios update failed. Try to update again and see if the new parts work that time.

A quick question for you....

Is it possible to have a GPU and a CPU that aren't compatible, also is it possible that my MOBO cant handle an 8x agp card. It says it can, but maybe the card isn't getting the volts it needs? I am not sure how to look up such information, if you think its worth a shot let me know and i'll Bios the slot to 4x and see what up... Currently im on step 1 about to run prime95 with old cpu new psu and no gpu and see what happens..

Temps from hwmonitor are excepetionally low

 

[jhowell]

So old cpu in, with original bios version and i am still getting crashes =/ I can do more on the computer, but a session of wow for example lasted maybe 7-10 minutes then it died.

I want so badly to believe its the mobo, but its acting just like overheating, but isn't overheating =p

Just an update

 

[jhowell]

K one last update, then i swear I think i'll have my answer... Ima put in the voltages being read off my machine from bios and cpuid, i know these are merely estimates, but i think there might be something profound in here... If you think this is cause to replace the powersupply, please recommend one that is overkill imo, i dont wanna do this anymore lol

From BIOS:
vcore: 1.4v
+12v: 12.3v
+3.3v: 3.36v
+5v: 5.12v
vbat: 3.24v
5vsb: 4.99v

From CPUID:
CPU VCORE: 1.44v
3.3v: 3.41v
And thats all it gives as far as relevent information

I am not very familiar with what exactly those values are supposed to be, but it looks like S@%$ smells like S@%$ and now im starting to wonder if my psupply really is s@#%

Thanks yo =)

 

[jhowell]

jesus i cant stop, so my gpu calls for 30 amps on a 12v rial... someone explained dthis wrong saying that add up the amperage on the rails, if i am understanding this correctly:

the cables from the cpu are on a specific rail, my psu provides 17 amps per rail on 2 total rails... meaning this issue has arose from a lack of amperage to the gpu?

 

[skora]

The easy way to rule out the power and GPU issue is use the onboard video. Pull the GPU out. (might have to enable the IGP in bios.) If you crash under prime95, then the power draw isn't a problem and if you're using the old CPU, the only thing left is the mobo.

 

[jhowell]

well it looks like now, i went out and bought a top of the line psu, singel raid 52 amp 650w, big motrucka. Thing screams now, even with the new cpu in it, the only thing that is killing me now, is randomly the monitor says disconnected and the machine will restart. It happens very infrequently, its just more of a pain then anything.

The other issue is when i make videos fullscreen like on hulu or what not, they play hella jumpy.

If anyone comes across this post in a google search who just bought this gpu, send it back, there is no flippin support for this thing, and it shows in how effectively it runs.

 

[jhowell]

lemme know if you know of a fix