Windows won't load after AMD Catalyst 13.12 install

[Maverickbcp]

Changed the title as this obviously has nothing to do with the Catalyst drivers.

Updated 4/22

My second replacement card just arrived today while I'm stuck here at work. To be honest I'm kind of scared to put it in my PC. Now I work in IT(and by IT I don't mean Geek Squad) so know what I'm doing. I've tested my system each way I can think of and I can't find a fault. Now I know it's possible to get two bad cards from a manufacturer, but it doesn't happen every day. This leads me to think the issue is somewhere in my system but heck if I can find it. I verified each of the last two cards were bad by using parts that have been working on my main card before sending them back and XFX confirmed my findings. Issue is was it something in my system or bad QC by XFX?

Of the first two cards I bought the one is still running strong without any issues. System isn't overclocked and the 12v rail from multiple sources seems to be pretty solid. Part of me hoped they would send me one of their cards that didn't work with my water block so I could use it as an excuse to just sell it and be done with it. Alas that didn't happen. I'm kinda glad it didn't as this thing will bug me until I figure out the cause.


Updated 4/03 short story version:

System worked fine for a week on air. Broke out water cooling kit and system worked fine for a couple days until I reinstalled Windows. Black screen after splash screen with both cards enabled but will load fine with second card disabled. If I try to enable the second card in Windows my screen goes black/monitor goes into standby and I'm forced to reboot which gets my second card disabled automatically by Windows. There is no report in Event Viewer when Windows fails to load past the splash screen or when I'm forced to reboot after enabling the second card. Various uninstall/reinstalls of drivers and card placement/power supply connectors/water block swapping verified the second card wouldn't work even if used as the only card in the system. RMA'd and XFX verified it was defective so shipped me a replacement. This one actually allowed me to boot into Windows with both enabled but literally 60 seconds later my system locks up and I'm back where I began with the exact same issues. This new card now too won't work even if it's the only card in the system using the same slot/power connectors/water block as the known working card. 2nd RMA incoming.



Hey guys I could really use your help with this extremely frustrating situation. I'm going to provide you with the whole scenario, and it's a bit of information, but please take a second and any advice you can give would be greatly appreciated. Thank you.

First a quick summary of the problem. Was running fine when I had to reinstall Windows to resolve another issue. When it came time to reinstall the GPU drivers + reboot, Windows won't load past the splash screen and I'm left with a blank black screen. Rebooting in safe mode works and I can get back into Windows only after uninstalled the drivers.

Just built a new pc. i7 4770k, ASROCK Extreme4, 16gb G. Skill Sniper, 250gb Samsung SSD, EVGA 1300w PS, and 2 XFX R9 290 graphics cards. Now my system was running fine for over a month on an older graphics card while I awaited the arrival of the 290's. Once the 290's came in I ran them through some Furmark, MSI Kombuster, and other stress tests to ensure they were working fine which they appeared to be after several days.

This is when I broke out my new water cooling equipment and installed everything. Alphacool rads, Swiftech mcp355 pump, XSPC blocks/fittings. I installed and hooked up everything which was working great. Temps were fantastic. Then one morning I was just watching an .mp4 file and I received that bsod which forced me to do the reinstall of everything.

Now the system was working fine before the water cooling and again for a couple days once all the equipment was added. It was only after I received that bsod and had to reinstall everything that i started getting this problem. I've reinstalled Windows several times to no avail. Tried the once working 13.12 and even the provided disc's 13.11 to no avail. Tried uninstalling the drivers, running CCleaner/DriverFusion, editing all ATI/AMD entries from the register, rebooting and trying again to no avail. Made sure Windows was fully up to date and each time reinstalled the MB's provided drivers as I did once before when it worked...all to no avail.

Device manager sees the 2 AMD R9 200 series cards and reports they're working fine with up to date drivers. I have no other exclamation point issues to address, or event viewer errors to point me in the right direction. The only ones I can see still appearing are the "DCOM got error '1084' attempting to start the service WSearch/EventSystem/ShellHWDetection/wuauserv with arguments, and those only appear when loading safe mode.

I'm sure I'm missing more that I've tried but I have an appt. I must get to and your eyes are probably bleeding enough as it is. Mind you this whole time I've been plugged into the top GPU and am obviously getting video through it. It's with the generic drivers of course, it's just once I try to install the specific card ones and try to load past the Windows splash screen, that I run into an issue.

I've read through endless amounts of people having similar issues but none of their "fixes" worked for me. Especially 9/10 it turned out they also had an HDMI cable plugged in(mine's DVI), or forgot to unplug from the onboard and plug into the GPU. Thank you for your time.

 

[TeknoBug]

This happening after installing a water cooling kit gives me a sign that your power supply might be at fault. Or the water cooling kit wasn't properly installed.

Also try an older AMD driver like 13.9 and see if it keeps doing it.

 

[Z15CAM]

Could be the PSU. Ensure you are running the latest BIOS for the MB.

I have read there are issues with the 13.12 driver for some combinations and reverting back to the 13.9 driver fixed it.

What version of Win?

 

[alexruiz]

...Was running fine when I received a BSOD asahci64.sys error...

When you installed catalyst 13.12, you installed only the cat 13.12 for graphics, or did you install also the cat 13.12 SB drivers? asahci64.sys is related to storage, sata in specific in ahci mode for the asmedia as106x. There is no way catalyst 13.12 would have broken it, even if you tried installing the catalyst 13.12 SB drivers.

I would do a clean reinstall of windows using only one video card, remover whatever you have connected to the asmedia sata ports, and go from there.

 

[Z15CAM]

asahci64.sys is related to storage, sata in specific in ahci mode for the asmedia

True - Wonder if he is not using the asmedia controller that he should try disabling it in BIOS to see if Win boots.

Personally I disable any unused controller in BIOS to alleviate possibilities of IRQ and Driver conflicts.

You may have to update the ASMedia Driver.

 

[Maverickbcp]

Thanks For The ReplieS And I'll Look Into Those When I Get Home. Posting From My Phone And Am Not Sure Why It's capitalizing Everything. Sorry I Forgot I'm Running Windows 7 64Bit. Not Sure How The Water Cooling Could Be Causing The Issue As It Only Happens After I Install The Gpu Drivers And Originally Was Working With The Drivers For A Couple Days With WaTer Cooling Installed Plus A Week Without It. No Changes Were Originally Made From The Working Setup To The Now Not Working. My Bios Is Up To Date. I Only Mentioned The Asahci64 Error For Timeline Purposes.

After I Was Forced To Reinstall Windows I've Been Installing The Complete Catalyst Package And Not Just The Drivers. Had An Issue Where My System Hung During The First Attempt To Install The Software And Because Of That Everything But The Drivers Would Install Leaving Exclamation Points On The Graphics Cards And Video Controller In The Device Manager. Had To Wipe The System Again.

I'm No Longer Receiving The Bsod Error And Haven't Changed Anything As Far As Drivers/Sata Connections That I HaD Working For Over A Month. I Can Try Reinstalling Them But Again I'm Not Receiving The Error Anymore And Only SeeM To RuN Into Issues Once I Try Installing ThE Gpu Drivers.

This Really Screams A Software Issue So Have Been Putting OfF Disconnecting Any Cards As I'll Have To Tear DowN/Drain The System. Missing Traditional Cooling Right Now. Much Easier To Test/Swap Out Equipment Then. I'll Do That If I Can't Find Anything Else Software-Wise That Could Be To Blame. Just So Frustrating. Everything Originally Installed/Worked Fine.

 

[Maverickbcp]

Also Water Cooling Was SEt Up Correctly. Temperature's On All Components Are Very Good. Regarding The Power Supply It Worked FoR A Week Straight Of Stress Testing Both CardS Without Issue And For Over A Month Before That Using MY Old Geforce 275.

 

[Maverickbcp]

No one else has any ideas eh? I'm at work for another 4 hours so must wait before more troubleshooting. System works fine until I install those drivers and attempt to log into Windows. Worked fine before so not sure what's causing the issue now. Guess I'll google some more.

 

[Phynaz]

Pull both video cards, get a functional system using the on board graphics and nothing else installed in the system. Then add in one video card and drivers. Then swap the cards and see if you are still good. Finally put both cads in.

Trouble shooting is a process of eliminating the variables one at a time.

 

[Maverickbcp]

Pull both video cards, get a functional system using the on board graphics and nothing else installed in the system. Then add in one video card and drivers. Then swap the cards and see if you are still good. Finally put both cads in.

Trouble shooting is a process of eliminating the variables one at a time.

Thanks for your input. I work in IT so have way too much experience troubleshooting lol. I was looking for ideas, most likely software related as this appears to be, before I start ripping apart my water cooling setup/reattaching heatsinks. Already got the spare system ready to go if I can't think of anything else to try software-wise. Still haven't given up on that possibility as too many things are pointing to it.

 

[smithkt]

I had a similar problem with both a 7970 and 290. It would boot properly into safe mode, but as soon as I tried to boot up normally, it would fail. I was able to work around the problem on with the 7970 by disabling the onboard video in the device manager. Once I did that, it worked fine.

Later, when I upgraded to a 290 I also upgraded to Windows 8. Had the same problem. Again disabling the onboard video in the device manager worked, but I was not happy with this solution. I checked and found ASUS had supplied an updated bios for my motherboard. Flashed it and enabled the onboard video. The problem was resolved. I am now using the onboard to drive a secondary monitor and xfire 290s to drive my main.

 

[Maverickbcp]

Ya I've read about a few people who had that issue with the onboard video and it resolved their issue. Unfortunately didn't fit my case.

HOWEVER by sheer luck I resolved the issue last night. I logged into safe mode, uninstalled the cards, and went to uninstall the Catalyst software so I could try a different driver. For whatever reason I received an error and it wouldn't let me uninstall the software so I said whatever and rebooted anyways. I was able to log into Windows where it recognized the cards and finished the installation. Also at the same time I received an error stating Microsoft .net 4.039xxx failed to start. Some people whom have had this issue before had it because of the .net installing wrong. So either uninstalling the cards but leaving the drivers installed and/or not having Microsoft .net 4.039xxx load got me past the black screen of death.

The IT guy in me wants to fix the .net issue and see if I can recreate the issue but the guy who just spent many thousands of dollars on this system broke the IT guys kneecaps as he wants to enjoy the computer for a bit lol.

 

[Z15CAM]

MB BIOS Vs Win IRQ re-assignment for Hardware & Driver's is the RUNT to getting any Platform up and running without BSOD's.

 

[Maverickbcp]

Well it appears my "lucky fix" was just a temporary fluke as I ended up running into the issue again. Only this time I determined and verified it's due to one of the cards going bad. Now I gotta set up an RMA with XFX and hope it doesn't take forever to get my replacement.

 

[ocre]

wow. read the thread and i can tell you were put through the ringer.

lavaheadache is having crazy issues out of his 780ti's too.

makes me think they dont make cards like they used to........

just kidding............i think.
lol

anyway, my condolences.

keep us posted.

 

[tweakboy]

Told you don't go red imo. jk I think. gl

 

[Maverickbcp]

Updated as you requested. After having to ship it back out of my own pocket and waiting two weeks, I received my replacement card. My card was the basic reference model but they sent me their top tier Double Dissipation overclocked model that has a lifetime warranty. After confirming with XSPC the pcb is reference I was like sweet!

So I hooked everything up and was actually able to log into Windows. Man it was a beautiful site. After confirming everything looked good in the device manager and various temperature monitoring software, I decided to let it run through a couple low stress tests using MSI Kombustor to make sure I didn't receive another faulty card. Not even 10 seconds into the test the temp gauge jumps up to 90 degrees...back down to 39 degrees...and the system freezes.

Now I think the gauge was reporting for the replacement card but this all literally happened with about 5 seconds so I didn't have time to really study anything. I powered off the system and was wondering wth happened but wasn't too concerned as the temps were fine/steady before the test and it was only 5 seconds which isn't going to hurt anything.

So anyways I try to boot the system back up and sure enough the same problem as before rears it's ugly head. This time I know what to test for so I try enabling the replacement card while in Windows...black screen. Try booting up with only the replacement card using the same slot/waterblock/cables as the working card...won't load past the splash screen.

So here I am contacting XFX for yet another RMA. I still can't say for 100% certainty whether they keep sending me duds or if something in my system is the culprit. My voltages look good and nothings overclocked. The temps were steady before the test and even if the test was reporting accurately, 5 seconds with it wouldn't have caused this. I'm just at a loss and completely frustrated with this whole experience.

 

[thesmokingman]

Not even 10 seconds into the test the temp gauge jumps up to 90 degrees

That doesn't sound good. I would check your loop/block right away. You should look into getting quick disconnects so that you can remove your gpu block at will without draining your loop.

For the driver install, did you disable auto driver updates? If auto updating is enabled, it will fubar your catalyst driver install and cause problems.


http://superuser.com/questions/665158/can-i-prevent-a-driver-from-being-installed-with-windows-8-1

 

[Maverickbcp]

That doesn't sound good. I would check your loop/block right away. You should look into getting quick disconnects so that you can remove your gpu block at will without draining your loop.

For the driver install, did you disable auto driver updates? If auto updating is enabled, it will fubar your catalyst driver install and cause problems.


http://superuser.com/questions/665158/can-i-prevent-a-driver-from-being-installed-with-windows-8-1

Ya after the amount of tear downs I've had to do I really wish I went with quick disconnects. I may just pick some up. I checked both blocks and they look identical. As I mentioned earlier I tried the replacement card in the block I've been using successfully for weeks with no success. My loop is good and all temps have been great. The test jumping between high 30's and low 90's for that few seconds I'm thinking was just a glitch. I ran that test before and my cards didn't really go above 60 degrees. This was before the replacement card. Otherwise my loop has been keeping my cpu/single working gpu in the low 30's.

As for the driver install no I didn't disable anything. However it won't even post and my motherboard throws out a '62' error code immediately so driver's aren't even in play yet. This is when the card is ran by itself. When ran together I'm able to boot into Windows but only with the second card disabled. If I try enabling it the screen goes black on me. If it wasn't for the system automatically disabling the card and the fact I can't even boot with it, I'd too think it was a driver issue. I wish it were a driver issue.

 

[thesmokingman]

Error code 62? It depends on your board. On some it refers to the southbridge or PCH, and others like Asus it's actually b2 and refers to memory.

Driver wise you need to get back to safe mode, disable auto driver updating, then run DDU at this point since your driver subsystem is a wildcard. Ya need to clean off the drivers and works well enough for this purpose. *After DDU wipes the old drivers, upon reboot windoze will install the default base drivers. You should be able to log in now.

 

[Maverickbcp]

Well I mentioned in my first post but I have a ASROCK Extreme 4 z87 board so error code 62 points to the GPU. The drivers work fine when I'm using the known good card by itself. Either way driver's wouldn't cause the system to boot/log into Windows fine with the working card but show no post/throw up an error code 62 with the replacement card. Also I'm running Windows 7 64bit. The link you provided talked about Windows 8.1. All my previous troubleshooting which included various driver installs/uninstalls is in the first couple of posts.

 

[Maverickbcp]

I do know how to go about disabling the updates in Windows 7. Just making sure you read my first post that included my system specs and everything I've done to this point.

 

[Arkadrel]

When you installed catalyst 13.12, you installed only the cat 13.12 for graphics, or did you install also the cat 13.12 SB drivers? asahci64.sys is related to storage, sata in specific in ahci mode for the asmedia as106x. There is no way catalyst 13.12 would have broken it, even if you tried installing the catalyst 13.12 SB drivers.
I would do a clean reinstall of windows using only one video card, remover whatever you have connected to the asmedia sata ports, and go from there.

That is true.

Doesnt sound like its a gpu issue.

Just out of curiousity how high do you have your CPU/Memory overclocked?

 

[Maverickbcp]

That is true.

Doesnt sound like its a gpu issue.

Just out of curiousity how high do you have your CPU/Memory overclocked?

None at all. That error seemed to be a fluke as I haven't had anything close to it since. The card I sent back to them was tested and found to be broke. Now whether it broke due to my power supply, motherboard, etc or just was defective I cannot say 100% one way or the other.

I've tried running the card in the same slot, using the same waterblock/power cables that work with my other gpu, and no post. If it were the slot on my motherboard, or the power cables, than when I move the card to the primary slot and run it by itself it should work but it doesn't. Quite a frustrating situation.

 

[Maverickbcp]

Part of me kinda hopes they replace my card with one of their models that doesn't work for water cooling so I can just sell it and be done with this. However another part of me doesn't want to give up and let this thing beat me. I'd also like to know if I have a defective piece in my system that may cause more issues later on.