Custom pc build, BSOD 0x00000124

[cloud32187]

Hi guys, friend of mine put together his own build a couple weeks ago he tried to figure it out on his own but now its in my shop.

Part list:
GPU: GIGABYTE GV-R929WF3-4GD Radeon R9 290 4GB 512-Bit
CPU: AMD FX-8350 Black Edition Vishera 8-Core 4.0GHz
Case: Antec TruePower Classic series TP-750C 750W 80 PLU
MOBO: ASRock 980DE3/U3S3 AM3+ AMD RX881/760G SATA 6Gb/s
RAM: Team Dark Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM
HDD: Seagate Barracuda ST1000DM003 1TB 64MB Cache SATA
OS: windows 7 home premium 64-bit

So heres the deal, he started getting bsod's from day one immediately after installing the gpu drivers. He tossed in a spare card (it was a peice of junk nvidia cant remember what model off the top of my head) and it worked fine, he could game for hours no problems, he played like this for about a month. He chalked it up to a bad GPU and rma'd the card.

Yesterday, got the new card in the mail. It was an "HIS R9 290" this time instead of GIGABYTE. Put the new card in and installed the drivers, bam, same issue as before.

I used bluescreenview and eventvwr to get some info and here's the 2 errors i found, which show up every crash:

Error 1:
"A fatal hardware error has occurred.

Component: AMD Northbridge
Error Source: Machine Check Exception
Error Type: HyperTransport Watchdog Timeout Error
Processor ID: 0"

Error 2:
==================================================
Dump File : 120814-19344-01.dmp
Crash Time : 12/8/2014 12:17:39 AM
Bug Check String :
Bug Check Code : 0x00000124
Parameter 1 : 00000000`00000000
Parameter 2 : fffffa80`085558f8
Parameter 3 : 00000000`00000000
Parameter 4 : 00000000`00000000
Caused By Driver : ntoskrnl.exe
Caused By Address : ntoskrnl.exe+4a5ebc
File Description : NT Kernel & System
Product Name : Microsoft® Windows® Operating System
Company : Microsoft Corporation
File Version : 6.1.7600.16792 (win7_gdr.110408-1633)
Processor : x64
Crash Address : ntoskrnl.exe+4a5ebc
Stack Address 1 :
Stack Address 2 :
Stack Address 3 :
Computer Name :
Full Path : C:\Windows\Minidump\120814-19344-01.dmp
Processors Count : 8
Major Version : 15
Minor Version : 7600
Dump File Size : 262,144
Dump File Time : 12/8/2014 12:17:46 AM
==================================================

Things I've tried:
-increasing and decreasing voltages to try and get it stable
-Ran stress tests with the card disabled(ran for 6 hours, temperatures all in normal range)
-Tried every version of driver out there for the card
-Memtest
-Re-seating all cables

Any help at all is welcome, if you need anymore information just let me know and ill be happy to provide.

 

[Pants]

What voltages did you change? When I built my system I had to increase the ram voltage to get it stable, but I didn't mess with the voltage of any other components. I would recommend testing one component at a time, if you are playing with voltages.

Looking at the support website for the board, it doesn't look like there's much in the way of updates. I was gonna suggest an updated chipset driver, for one thing.

Other than that, for starters, try disconnecting as many things as you can from the motherboard... Get the system running with the bare essentials plugged into the motherboard.. Once that's done if it's still not stable and you haven't isolated the problem, try plugging some of the external devices into different usb ports and testing to see if the problem reoccurs, to help isolate potential stability problems.

Might want to check if a live linux cd will run stable...that'l help determine whether the problem is hardware or software.

 

[Ed_Zipper]

I would try a $65 Gigabyte AMD Mobo. Sounds like you possibly have one that shipped bad.

 

[cypress]

I would try a $65 Gigabyte AMD Mobo. Sounds like you possibly have one that shipped bad.

It's possible, but did you also run Memtest on the memory as well?

 

[ohio_grad_06]

And if it's something about those series of cards, have him return the 290 and get one of these.

http://www.newegg.com/Product/Product.aspx?Item=N82E16814500362

GTX 970. These are the hot new cards all the gamers seem to want, I think they are supposed to be more power efficient too. What are the temps like in his case?

 

[jbartlett323]

Error 1:
"A fatal hardware error has occurred.

Component: AMD Northbridge
Error Source: Machine Check Exception
Error Type: HyperTransport Watchdog Timeout Error
Processor ID: 0"

I think you might want to look in this direction, as this error has nothing to do with the Card. This is telling you that the northbridge driver is whats going toes up. When you load the Catalyst package, is it only loading video drivers? or is it doing the whole AMD chipset too? Have you loaded the chipset drivers? Where are you getting said drivers, Manu or AMD?
Have you stress tested this system (with any card) to make sure the board and proc are stable? You say it works with an old card, maybe when you try to make the system use the throughput of a new, better card, it hiccups? Have yo tried bad card in another system? Preferably an Intel based system?

 

[ohio_grad_06]

He said he can game on an old nvidia card. But the northbridge is the issue? That 290 consumes like 300 watts. I wonder if the 290 is getting hot and it making the northbridge too hot which is causing his errors? That northbridge is right above where the card is, and just below the cpu according to the manual.

If you insist on the 290, maybe try one of these positioned over the north bridge?

http://www.amazon.com/Antec-Spot-Co...=2025&creative=165953&creativeASIN=B00JYE5I1Y

 

[bitznpcz]

I wonder if the 290 is getting hot and it making the northbridge too hot which is causing his errors?

STOP 124 are usually processor related - under/over voltage or heat issues are the main problems - I think you have hit the nail on the head and I would guess an overheated Northbridge is to blame...

 

[MikeLierman]

STOP 124 are usually processor related - under/over voltage or heat issues are the main problems - I think you have hit the nail on the head and I would guess an overheated Northbridge is to blame...

This is my experience as well. Had a customer's machine that was BSODing instead of shutting off. Turned out the fan was stopping at random. The stock OEM fans are garbage. Do yourself a favor and get a decent aftermarket CPU fan.

 

[ohio_grad_06]

If you are talking cooling for the FX 8350, the Hyper 212 evo is good if you are on a budget. Will run 30 bucks maybe a little more, but install that with some Arctic Silver 5 and cpu temps should drop down. As far as the video card, maybe you can add a case fan nearby to give supplemental cooling to that and the northbridge. They also make a PCI slot cooling fan. I know these may look ghetto, but look interesting as well.

http://www.ebay.com/itm/New-PC-Comp...US_Video_Card_GPU_Cooling&hash=item27d5ec7159

Of course there are other smaller models.

Or this setup, clearly from China, but the way they show in pictures could be useful.

http://www.ebay.com/itm/Coolertec-S...US_Video_Card_GPU_Cooling&hash=item4ac9f9c29e


Does that northbridge have any kind of active cooling? Some of them I know have heatpipes etc.

 

[ohio_grad_06]

Looked up that board, is this the board he's got?

http://www.newegg.com/Product/Product.aspx?Item=N82E16813157366

If so return that thing. Look at the VRM's on it. Does not look like there is any cooling on the VRM's and that only looks to be a 4+1 power phase board, which I think is a little low for that FX 8350. I would suggest chucking that board and grabbing a better one.

I run the older version of this board in my personal system with an FX 8120 overclocked to 4.2 ghz stable for 24/7 use.

http://www.newegg.com/Product/Product.aspx?Item=N82E16813128651

8+2 power phasing, see how much more cooling there is on the various board parts? If budget is an issue, maybe this one. Of course you may want to check their site for compatibility.

http://www.newegg.com/Product/Product.aspx?Item=N82E16813128565

But see how much better cooling is on the VRM's and what not on the lower end board than on the AsRock? The lower end gigabyte I think has the 4+1 power phasing, but looks like parts are a little better cooled. I've used various variations of that board in cheaper builds and have rarely had issues with any of them.

 

[Pants]

I wouldn't argue against any of these suggestions, but I wouldn't buy anything until you've tried no-cost solutions first. From my experience and what's noted in "Upgrading And Repairing Windows 2nd Edition", Stop errors, even though they display data about what happened, the data does not always mean the specified components/drivers are the culprit; The specified culprits could be indirect results of some other problem.

I still do recommend seeing if a live linux cd will run stable, to help isolate the culprit, and tearing the system down to the bare essentials to see if you can get the system stable....PSU, motherboard and cpu, keyboard (if feasible, don't plug in a mouse), monitor, HDD (or live cd if ruling out hdd/software problem).

 

[ohio_grad_06]

+1 for the above. If you do have to get new parts, hopefully you can rma them to exchange.

 

[Pants]

I would also suggest resetting the voltages to the defaults when running a live cd or stripping down the system, to avoid compounding the problem. Do voltage changes only after you've tried the most basic stuff, and change the voltage of only one component at a time to...Changing the voltage setting of the cpu while testing the voltage changes of the ram could lead to inaccurate test results.