• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Bad day-Hard drive died, now the motherboard died

Markfw

Markfw

Moderator Emeritus, Elite Member
OK, so I had a perfect setup, no problems at all. Then the hard drive died. So I put a new drive in, installed windows (XP 32 bit, SP2), installed the CPU client (no problem), then the GPU client.

Now, first, this card is not overclocked at all, but when I started running the client and got mdrun_gpu errors, I downclocked it from 620 down to 580, and memory from 1080 to 1000 just to eliminate that, but still to no avail. The hardware is virtually the same, so why now is it that I am getting these errors ???


Note: Please read the license agreement (Folding@home-Win32-GPU.exe -license). F
urther
use of this software requires that you have read and accepted this agreement.



--- Opening Log file [May 28 19:20:47 UTC]


# Windows GPU Console Edition #################################################
###############################################################################

Folding@Home Client Version 6.23

http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\folding_gpu1
Executable: C:\folding_gpu1\Folding@home-Win32-GPU.exe
Arguments: -gpu 1 -forcegpu nvidia_g80 -advmethods

[19:20:47] - Ask before connecting: No
[19:20:47] - User name: Mark_F_Williams (Team 198)
[19:20:47] - User ID: 35A964C929345CF7
[19:20:47] - Machine ID: 3
[19:20:47]
[19:20:48] Loaded queue successfully.
[19:20:48] - Preparing to get new work unit...
[19:20:48] + Attempting to get work packet
[19:20:48] - Connecting to assignment server
[19:20:48] - Successful: assigned to (171.67.108.11).
[19:20:48] + News From Folding@Home: Welcome to Folding@Home
[19:20:48] Loaded queue successfully.
[19:20:48] + Closed connections
[19:20:48]
[19:20:48] + Processing work unit
[19:20:48] Core required: FahCore_11.exe
[19:20:48] Core found.
[19:20:48] Working on queue slot 01 [May 28 19:20:48 UTC]
[19:20:48] + Working ...
[19:20:48]
[19:20:48] *------------------------------*
[19:20:48] Folding@Home GPU Core
[19:20:48] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[19:20:48]
[19:20:48] Compiler : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14
.00.50727.762 for 80x86
[19:20:48] Build host: amoeba
[19:20:48] Board Type: Nvidia
[19:20:48] Core :
[19:20:48] Preparing to commence simulation
[19:20:48] - Looking at optimizations...
[19:20:48] DeleteFrameFiles: successfully deleted file=work/wudata_01.ckp
[19:20:48] - Created dyn
[19:20:48] - Files status OK
[19:20:48] - Expanded 45435 -> 251112 (decompressed 552.6 percent)
[19:20:48] Called DecompressByteArray: compressed_data_size=45435 data_size=2511
12, decompressed_data_size=251112 diff=0
[19:20:48] - Digital signature verified
[19:20:48]
[19:20:48] Project: 5770 (Run 1, Clone 147, Gen 397)
[19:20:48]
[19:20:48] Assembly optimizations on if available.
[19:20:48] Entering M.D.
[19:20:54] Tpr hash work/wudata_01.tpr: 3142530096 2984723146 1955465856 389076
6175 1530411967
[19:20:54]
[19:20:54] Calling fah_main args: 14 usage=100
[19:20:54]
[19:20:54] mdrun_gpu returned
[19:20:54] Going to send back what have done -- stepsTotalG=0
[19:20:54] Work fraction=0.0000 steps=0.
[19:20:58] logfile size=4947 infoLength=4947 edr=0 trr=25
[19:20:58] + Opened results file
[19:20:58] - Writing 5485 bytes of core data to disk...
[19:20:58] Done: 4973 -> 1863 (compressed to 37.4 percent)
[19:20:58] ... Done.
[19:20:58] DeleteFrameFiles: successfully deleted file=work/wudata_01.ckp
[19:20:58]
[19:20:58] Folding@home Core Shutdown: UNSTABLE_MACHINE
[19:21:02] CoreStatus = 7A (122)
[19:21:02] Sending work to server
[19:21:02] Project: 5770 (Run 1, Clone 147, Gen 397)
[19:21:02] - Read packet limit of 540015616... Set to 524286976.


[19:21:02] + Attempting to send results [May 28 19:21:02 UTC]
[19:21:03] + Results successfully sent
[19:21:03] Thank you for your contribution to Folding@Home.
 
Last edited:
Googling, I found this:

ardichoke said:
You probably don't need to reinstall, just go into the working directory and delete queue.dat, FAHlog.txt, any FahCore executables, unitinfo.txt and the work directory. Then relaunch the client. If it still goes straight into an EUE loop, then you might be experiencing hardware problems. I know my card started doing that shortly before it started flaking out on me.
Click to expand...
 
I did all that, and I am sure the video card is good. Thanks anyway.

OK, I figured this out maybe, at least its now working. After the zillionth try, it finally came up with a prompt about unblocking fah_11.exe, I said yes, and now its crunching...

Damn windows security !
 
Last edited:
Markfw900 said:
I did all that, and I am sure the video card is good. Thanks anyway.

OK, I figured this out maybe, at least its now working. After the zillionth try, it finally came up with a prompt about unblocking fah_11.exe, I said yes, and now its crunching...

Damn windows security !
Click to expand...

I have horror stories of Windows Security and ghosting (using norton ghost) at work. Ugh.... It's a bloody nightmare. That's why all of my servers are Linux. If something isn't right, I know how to find AN EASILY CONFIGURATION FILE to adjust it. Now if I can jsut get DPAD working on my linux boxes. 🙂

PS--Need a hard drive?
 
Fullmetal Chocobo said:
I have horror stories of Windows Security and ghosting (using norton ghost) at work. Ugh.... It's a bloody nightmare. That's why all of my servers are Linux. If something isn't right, I know how to find AN EASILY CONFIGURATION FILE to adjust it. Now if I can jsut get DPAD working on my linux boxes. 🙂

PS--Need a hard drive?
Click to expand...

newegg refurb, $31, and on the way....
 
Markfw900 said:
newegg refurb, $31, and on the way....
Click to expand...

Okay. I've got quite a few spare hds piling up. And they are still usable and under warranty, so I'd (almost) hate to use them for target practice with the .308.

🙂

So for what it is worth, you can trust me that a hard drive in my previous possession has paid for your hd's transgressions. 🙂
 
Fullmetal Chocobo said:
Okay. I've got quite a few spare hds piling up. And they are still usable and under warranty, so I'd (almost) hate to use them for target practice with the .308.

🙂

So for what it is worth, you can trust me that a hard drive in my previous possession has paid for your hd's transgressions. 🙂
Click to expand...

Well, actually, I still have need of a couple. I have a 15 gig thats about ready to die, and my external got used up to fill in for now. Should I PM you my address ? Need shipping ?
 
Bad day...I am pretty sure I burned out the mobo. I put my 550 BE in there, and reset the bios. All I get is a quick flicker on the power light, and the fans move 1/8th of a revolution, just enough to know there is power applied. I put the X6 in my other motherboard, and it powers up, but no post (N68-S, doesn't support Thuban)

So I ordered a GIGABYTE GA-790XTA-UD4 AM3 AMD 790X SATA 6Gb/s USB 3.0 ATX AMD Motherboard . What do you think ? will it last longer ? Remember, I run mine overclocked, and 100% load 24/7/365.
__________________
 
Markfw900 said:
Bad day... reset the bios. All I get is a quick flicker on the power light, and the fans move 1/8th of a revolution, just enough to know there is power applied.
Click to expand...

1) If you removed and reconnected the front panel power switch cable. reverse it. If it's the wrong polarity, it will exhibit those symptoms. (It behaves as if you are holding the switch in for power off.)

2) Check the 12V cable to the mobo socket near the cpu. If that is dead or loose, you will get those symptoms. If it requires more power than you a supplying, (i.e., 4x connected, 6x needed) you will get those symptoms.

3) Some PSUs have a delay to wait for CPU fan to spin up before sending power to the CPU. Try connecting the CPU fan to a regular Molex so it is always spinning. This might allow you to get to the BIOS to disable the no-fan warning and to disable C&Q to always have the fan spinning.
 
Last edited:
Thanks for the advice ... I just built a comp for one of my sons and it had similar problems ... advice #1 solved it right away.
I did not know that polarity of the power switch matters - I have tried it in both ways some years ago and it did not matter than - but it obviously does now.
 
Rudy
Are you sure about the power switch thing? I think ATX case power switches merely give a brief connection when pressed rather than staying on.
Also when powering a mbrd without a power switch you only need to quickly bridge the pins to fire it up.
[edit] OK I've just seen Petrus's post & I'm confused now ...........

Mark
Not having much luck with that PC atm are you?!, have you confirmed that the PSU is OK? (I've had bad PSUs give the same symptoms).
 
Assimilator1 said:
Rudy
Are you sure about the power switch thing? I think ATX case power switches merely give a brief connection when pressed rather than staying on.
Also when powering a mbrd without a power switch you only need to quickly bridge the pins to fire it up.
[edit] OK I've just seen Petrus's post & I'm confused now ...........

(edited by petrusbroder)
Click to expand...

Well, it worked. May have something to do with grounding ... I usually test the mobos and CPUs with minimum RAM and simplest GC on the desktop and then the shorting of the powerswitch pins is sufficient to start it all ... but in a case there may be some other potentials ...
 
Assimilator1 said:
Rudy
Are you sure about the power switch thing? I think ATX case power switches merely give a brief connection when pressed rather than staying on.
Also when powering a mbrd without a power switch you only need to quickly bridge the pins to fire it up.
[edit] OK I've just seen Petrus's post & I'm confused now ...........

Mark
Not having much luck with that PC atm are you?!, have you confirmed that the PSU is OK? (I've had bad PSUs give the same symptoms).
Click to expand...

I also find it hard to believe (reversing the power on pins), but I will try again.

As for the PSU's both are only a week old, one is 700 watt >85% efficiently, the other is 1010 watt, and same efficiency. I find it hard to believe that both died, especially if one wasn;t even hooked up (but was tested after I got it)
 
Yes, it is a grounding issue. In the old days the front panels were not grounded and so the polarity was a non-issue. Now they are grounded and the wrong polarity will cause the BIOS to recognize a grounded connection instead of an intermittent connection.
 
Last edited:
Rudy Toody said:
Yes, it is a grounding issue. In the old days the front panels were not grounded and so the polarity was a non-issue. Now they are grounded and the wrong polarity will cause the BIOS to recognize a grounded connection instead of an intermittent connection.
Click to expand...

Interesting. I don't believe that I've run across a case that grounds the power switch connection. Do you recall any offhand? That's good to know, though, in case I run across one.

Edit: I have, however, come across several motherboards recently that didn't list any polarity for the power switch. Uh-oh!
 
VirtualLarry said:
Interesting. I don't believe that I've run across a case that grounds the power switch connection. Do you recall any offhand? That's good to know, though, in case I run across one.

Edit: I have, however, come across several motherboards recently that didn't list any polarity for the power switch. Uh-oh!
Click to expand...

I just tore the lid off my Antec 900 and saw that the ground wire goes into the center of the panel. This take care of 2 USB, 1 firewire, and audio. The power switch and reset switch are immediately adjacent to those guys and I cannot tell if there is an internal ground buss that connects them. I don't want to remove the 4 screws to see more because the power led is glued in a little carelessly and would need to be removed to get that panel off.

I have found this polarity problem with Gigabyte mobos. I have 2 Biostar mobos in use, but I haven't hooked up the front panels yet. (Biostar has these nifty little switches to power on and reset mounted on the mobos.) I will be completing a build for a neighbor this weekend and will play around with it then.
 
Rudy
So you're saying that for some cases that one of the wires from the power switch goes via earth instead of just going straight to the mbrd?
 
Well, I checked.
One of the leads goes directly to the mobo. That lead is marked "+".
The other goes also to the mobo, but is also connected - where the switch is - to the chassis (case) which in this case is - I think - electrical ground. That lead is marked "-".
I checked with my multimeter: the "-"-marked lead shown no resistance at all compared to chassis; the "+".marked lead has infinite resistance to the chassis.
From the "reset"-switch both leads go the mobo and there is no connection to the chassis.
 
Last edited:
petrusbroder said:
Well, I checked.
One of the leads goes directly to the mobo. That lead is marked "+".
The other goes also to the mobo, but is also connected - where the switch is - to the chassis (case) which in this case is - I think - electrical ground. That lead is marked "-".
I checked with my multimeter: the "-"-marked lead shown no resistance at all compared to chassis; the "+".marked lead has infinite resistance to the chassis.
From the "reset"-switch both leads go the mobo and there is no connection to the chassis.
Click to expand...

So Rudy was right. I never would have known.

What case is that, for future reference?
 
I checked the BioStar mobos. They don't have polarity marked. The power switch works in both postitions.

So, this appears to be mobo dependent.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top