Bitcoin Forum
June 23, 2024, 07:05:39 PM *
News: Voting for pizza day contest
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: GUIMiner no longer working - Argh! - any ideas?  (Read 5734 times)
Zithras (OP)
Newbie
*
Offline Offline

Activity: 12
Merit: 0


View Profile
February 10, 2013, 11:25:46 PM
 #1

I'm getting a strange issue with GUIMiner, and hope someone has fixed this already!

As of yesterday, I was running GUIMiner just fine on 3x5870s

Last night, I tried to install AMD Catalyst 13.1, the the computer went nuts.
Uninstalled 13.1, reinstalled 12.10, reinstalled SDK 2.8x64

Catalyst is working fine, games are working fine, other programs are working fine, GUIMiner now will not start, and instad gives me a 'guiminer.exe has stopped working' error in Windows 7 (Ultimate, x64).
Nothing about the computer has changed, I just uninstalled the new driver and reinstalled the old one.

Clicking 'view problem details' gives me:

Problem signature:
  Problem Event Name:   BEX
  Application Name:   guiminer.exe
  Application Version:   0.0.0.0
  Application Timestamp:   49180193
  Fault Module Name:   StackHash_0a9e
  Fault Module Version:   0.0.0.0
  Fault Module Timestamp:   00000000
  Exception Offset:   00000000
  Exception Code:   c0000005
  Exception Data:   00000008
  OS Version:   6.1.7601.2.1.0.256.1
  Locale ID:   1033
  Additional Information 1:   0a9e
  Additional Information 2:   0a9e372d3b4ad19135b953a78882e789
  Additional Information 3:   0a9e
  Additional Information 4:   0a9e372d3b4ad19135b953a78882e789

Well, BEX is a data execution prevention thing, right.  I didn't have to add it into a DEP-ignore list before, but maybe something changed...so then I specifically disabled Data Execution Prevention for GUIMiner.exe...which gets me a slightly different, but similar error message:

Problem signature:
  Problem Event Name:   APPCRASH
  Application Name:   guiminer.exe
  Application Version:   0.0.0.0
  Application Timestamp:   49180193
  Fault Module Name:   StackHash_0a9e
  Fault Module Version:   0.0.0.0
  Fault Module Timestamp:   00000000
  Exception Code:   c0000005
  Exception Offset:   00000000
  OS Version:   6.1.7601.2.1.0.256.1
  Locale ID:   1033
  Additional Information 1:   0a9e
  Additional Information 2:   0a9e372d3b4ad19135b953a78882e789
  Additional Information 3:   0a9e
  Additional Information 4:   0a9e372d3b4ad19135b953a78882e789
Has anyone run into that? How do I fix it?  It was running fine yesterday: this is oh so very frustrating!

Thanks for the help,
Zithras
jerye
Newbie
*
Offline Offline

Activity: 57
Merit: 0



View Profile
February 11, 2013, 06:07:34 AM
 #2

Try cgminer and see if it works Smiley
Beyondo
Member
**
Offline Offline

Activity: 104
Merit: 10



View Profile
February 11, 2013, 07:41:07 AM
 #3

Its worth spending the time to learn how to use CGMiner
Zithras (OP)
Newbie
*
Offline Offline

Activity: 12
Merit: 0


View Profile
February 11, 2013, 08:24:59 AM
 #4

After much grief, I figured it out.

When uninstalling AMD drivers and moving to an earlier version, the included uninstaller does not, in fact (even after a reboot!), uninstall all drivers, devices, and software properly, leading to corrupted drivers, malfunctioning software, and endless BSOD loops upon restart (aaarghhh!!!)  The BSOD loops, of course, will also cause Windows filesystem corruption after awhile, leading to even more driver errors, loss of internet access, data corruption, etc., and each time you run the Windows 'repair' or 'filecheck' utilities, you pretty much guaranteed to break something new.

In this case, among other issues, my OpenCL driver was not functioning correctly, even though the installer(s) (both Catalyst and SDK) swore up and down it loaded just fine, and repairing it multiple times all ensured me I had a functional install each time...

Guiminer, GPU Caps Viewer, and poclbm all gave me that unhelpful crash message, but the newest GPU-Z was more helpful, telling me I had a bad OpenCL driver (just before crashing, of course).  Of course, GPU-Z confusingly claimed it was an INTEL OpenCL driver that was broken (wrong), despite the fact that I have no integrated graphics, and an ATI video card...but oh well.

If anyone else wants to downgrade from 13.1, run the AMD uninstall utility, restart in safe mode, run the uninstall a second time (only partially worked the first time), delete all folders beginning with ATI or AMD in the Program files and program files (x64) directories.  Then restart again in safe mode, and clean up regis0rty keys, and C:\Windows\assembly files, and uninstall your video cards via device manager.  Restart into normal mode, and you should be able to install an older version of Catalyst.  If you get a BSOD loop, something's only partially uninstalled.  Except that didn't work for me either, so I ended up installing Catalyst software suite 12.8, then upgrading everything except the Control Center to 12.10 (the 12.10 control center still refuses to start for me...)  I'm probably missing a step or two, but you get the general idea >.< Chances are your video card manufacturer only has the 13.1 drivers, so youll have to get older ones from the AMD website.  And no, according to AMD, there is no way to install *just* the OpenCL drivers.   You either have to go through the Catalyst suite, or the SDK installer.  Don't use Driver Sweeper to uninstall anything, it will only cause pain.  You should probably avoid CCleaner too.

If this sounds like a huge pain, it definitely is.  In other words, Don't upgrade AMD Catalyst to 13.1.  It seems to be extremely buggy, and downgrading is a huge pain, and certainly not guaranteed to work well either.  I read plenty of forum threads where the only solution to an attempted downgrade was reformatting and reinstalling Windows Sad  My computer's definitely buggier than what it was 2 days ago, but at least I'm (mostly) back to a working video card driver.  And GUIMiner/poclbm run once again.

The most annoying part of all this is that it took me almost all day just to get a HELPFUL error message to figure out exactly what was happening, rather than random stack overflow warnings and APPCRASH hashes.

Frustrated, and going to sleep,
Zithras
Beyondo
Member
**
Offline Offline

Activity: 104
Merit: 10



View Profile
February 11, 2013, 08:31:47 AM
 #5

Yeah newer Catalyst drivers are always a crapshoot.  I always wait a month and let other be the guinea pig.  ASIC will end all that anyways...
opn
Jr. Member
*
Offline Offline

Activity: 54
Merit: 1



View Profile
February 26, 2013, 11:29:10 PM
 #6

Zithras, thank you for your long and useful post and idea with intel video driver. Driver Fusion utility is useable for it, plus driver 13.1 working good after many many attempts and reboots.
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!