• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Todays Project

T

Tobit

[H]ard|DCer of the Month - March 2010/May 2011
Joined
Dec 12, 2005
Messages
5,145
So today, I should be able to get some decent, although not much, "D" back online for the first time in a couple months. I received the motherboard shown below late last week from someone here on [H]ard. Today, the mailman should be bringing me an E6400, 4GB of Memory, and a Scythe Ninja. On Friday, I have two 9600GSOs arriving from brokenarrow03 but, in the short term, I have an 8800GT I will be running. Come Friday, the 8800GT will likely go into my parents computer who are currently running a single instance of the classic CPU client on their 3.0 GHz P4.

On Nov. 3rd, I will be ordering moar "D" to the tune of around 25K PPD (rough estimate).

DFI LT P35-T2R - although you will note 3 x 16x slots, the third only runs at 1x speed.

2irvsd5.jpg
 
Edit: nevermind, false alarm. I thought I had the wrong backing plate for the HSF and I didn't have a manual for it but I figured out the mounting.
 
Last edited:
Ok, making progress. Now to shoe horn it into my small-ish chassis.

2ccnuqw.jpg
 
Motherboard installed in the case, BIOS configured, and now installing Windows 7 to do some overclocking and stability testing. I'm going to need to dig my Antec 300 case out of the closet though, this case is just a bit too small. Oh well.
 
Shinies! UPS is bringing me a bundle of boxen parts today as well. I am building a system for my dad. I'll probably drop a standard client in it but not gonna get too technical with it. He is fairly tech challenged so I'm not going to make it very complicated off the bat. Later on I might drop a vid card in it.
 
Thanks guys.

I put a mild overclock on the CPU, a roughly 500 MHz boost from 2.13GHz to 2.6GHz, and I've been folding on the 8800GT now for a few minutes. Currently 20% through my first 353 pernt WU. I have not decided yet how I am going to fold with the CPU.

Screenshot:

http://i37.tinypic.com/jfghe8.jpg

Very pleased so far with the 5K+ PPD out of this thing. :D :cool:
 
do all lan party boards look the same? that looks just like my 680i lol

if you run naked, that on board power button is amazing
 
Vaulter98c said:
do all lan party boards look the same? that looks just like my 680i lol
Click to expand...
For the most part. Thankfully this is NOT and nividia board. :D

PPD has now settled in at 5545.31 for the past few minutes. :cool:
 
with 2 9600 GSO's youll get about 7-7.5 , you could probly get 300-500 from a standard CPU client too
 
Vaulter98c said:
get 300-500 from a standard CPU client too
Click to expand...
yeah, I may just devote one of the CPU cores to a DC Vault project instead.. something like OGR-27 or RC5-72 -- leave the other one idle for my day to day tasks.
 
oops, things aren't going well after all. First WU just crashed.

Code: 
[20:01:19] Completed 53%
[20:01:19] mdrun_gpu returned 
[20:01:19] NANs detected on GPU
[20:01:19] 
[20:01:19] Folding@home Core Shutdown: UNSTABLE_MACHINE
[20:01:22] CoreStatus = 7A (122)

bummer
 
Tobit said:
yeah, I may just devote one of the CPU cores to a DC Vault project instead.. something like OGR-27 or RC5-72 -- leave the other one idle for my day to day tasks.
Click to expand...

Sounds like a good plan.

I look forward to the updates, you have a bunch coming.
 
Damnit, 2nd unit just crashed as well. Time to remove the CPU overclock and try again.
 
Do you have an OC on the GPU?

What PSU do you have?

What is Temp on the GPU?
 
Kendrak said:
Do you have an OC on the GPU?
What PSU do you have?
What is Temp on the GPU?
Click to expand...
Only overclock is the hard overclock done by XFX. PSU is a fairly new Antec 650W Signature -- very highly regarded by Jonny Guru. Temp on the GPU at full load is 67C.
 
Del your work folder and start it back up, see what happens.

Otherwise I'm going to tell ya to check the HSF on the GPU.
 
I removed the CPU overclock and just started the 3rd WU but this is a big and chewy 787 point one. Going to take 4h to complete, currently at 3%. Let's see how it goes.
 
Well, damnit. It happened again. Now what? 3 out of 3 unsuccessful attempts thus far. HSF problem despite it runs at 68C when folding? Windows7 problem?

Code: 
[20:20:38] Project: 5797 (Run 19, Clone 595, Gen 1)
[21:30:47] Completed 28%
[21:30:47] mdrun_gpu returned 
[21:30:47] NANs detected on GPU
[21:30:47] 
[21:30:47] Folding@home Core Shutdown: UNSTABLE_MACHINE
[21:30:51] CoreStatus = 7A (122)
[21:30:51] Sending work to server
 
Tobit said:
Well, damnit. It happened again. Now what? 3 out of 3 unsuccessful attempts thus far. HSF problem despite it runs at 68C when folding? Windows7 problem?

Code: 
[20:20:38] Project: 5797 (Run 19, Clone 595, Gen 1)
[21:30:47] Completed 28%
[21:30:47] mdrun_gpu returned 
[21:30:47] NANs detected on GPU
[21:30:47] 
[21:30:47] Folding@home Core Shutdown: UNSTABLE_MACHINE
[21:30:51] CoreStatus = 7A (122)
[21:30:51] Sending work to server
Click to expand...

I'd still check, maybe a RAM module is not getting cooled.

At this point I'm at a bit of a loss.
 
Just took the HSF off the card, applied new Ceramique, and reinstalled HSF. Now rebooting and will try again. If this doesn't work, I may go back to WinXP.
 
I think I'm going to WinXP as it just crashed again. I don't get it and can't think of anything else.
 
ya try XP. XP I find from experience is loads more stable for folding than vista.

case in point I would need to restart my vista machine once or twice a week to recover the video the driver and make everything shiny again. With my XP machine I actually haven't hit a stability limit, it can go months without ever asking for so much as a break.
 
XP now installed, trying another WU now.
 
son of a #@#&!! Still having the same problem. I must have a lemon of a card. I dunno what else to do at this point. Very disappointed.

Code: 
[00:25:23] Project: 5766 (Run 8, Clone 165, Gen 435)
[01:31:51] Completed 70%
[01:31:51] mdrun_gpu returned 
[01:31:51] NANs detected on GPU
[01:31:51] 
[01:31:51] Folding@home Core Shutdown: UNSTABLE_MACHINE
[01:31:54] CoreStatus = 7A (122)
[01:31:54] Sending work to server
 
68C is what to expect out of a slimline system (trust me...) at light loads.

Your case, I'd check:
Stock TIM (probably some weird, hard biscut by now [based on my experiences with XFX's lower end cards])

Folding in another slot

Checking northbridge temps.
 
jeremyshaw said:
Stock TIM (probably some weird, hard biscut by now [based on my experiences with XFX's lower end cards])
Click to expand...
Already re-did the TIM, see post #22. I used AS Ceramique.
 
Tobit said:
son of a #@#&!! Still having the same problem. I must have a lemon of a card. I dunno what else to do at this point. Very disappointed.
Click to expand...
It doesn't sound like it's the card and if you are at CPU default and northbridge auto settings then I would suggest looking up your memory specs and and be sure the real voltage/not the bios setting voltage is enough for the memory at those speeds.
It may be that the info that is being given to the video card may be getting corrupt in the memory.

After that be sure the motherboard has an updated bios.
-After that I would try the card in a different slot and/or try a different video card to narrow it down.

Good luck bud.
 
Razor_FX_II said:
I would suggest looking up your memory specs and and be sure the real voltage/not the bios setting voltage is enough for the memory at those speeds. It may be that the info that is being given to the video card may be getting corrupt in the memory.
Click to expand...
Memory spec is 1.8V and motherboard has 1.88V set.

I'll check the BIOS version next and then try another slot.
 
BIOS was three revisions behind. Updated and now trying again. Now working on a 787 pointer.
 
Welp, after updating the BIOS, it still failed. I'm going to try the card in a different PCIE slot and, if that doesn't work, I'm going to try again with just one stick of memory.
 
Did you try just running a benchmarking tool / burn tool to see if it's just messed up WU getting through?
(improbable, but I have gotten 2 bad lightbulbs in a row before).
 
jeremyshaw said:
Did you try just running a benchmarking tool / burn tool to see if it's just messed up WU getting through?
(improbable, but I have gotten 2 bad lightbulbs in a row before).
Click to expand...
Yeah, but only about an hour of stress testing the CPU followed by an hour of testing the memory with memtest86. So far, I've had 8 failed WUs in a row with no successes so something is wrong on my end, not a bad WU.
 
I'm beginning to think that this card just can't fold. I pulled one of the DIMMs and now running single channel on just one 2GB stick and I still can't fold. I'm going to try the cuda client for RC5-72 for a bit and see what that does.
 
RC5-72 Cuda testing is working out well. I have 2 packets done already. Going to let it run to 25 and see how it does. I have 2 x 9600GSOs on the way, cards that folded well for their PO, and they will be here either today or tomorrow. We'll see if they have the same problem with FAH in this boxen.
 
yeah RC5-72 doesnt put as much of a load on the GPU.. the only thing i would recommend to really test the gpu is to run furmark.. if it crashes in furmark then you know the card is a lemon.. if it doesnt. then you know the problem is with either the drivers or the FAH client..

my guess is that its the drivers..


btw a cpu overclock is going to have almost no effect on the performance of the gfx cards since they are not related to each other with the PCI-E clock being locked at 100mhz so stress testing the cpu wouldnt do anything to tell you if the cards bad..
 
Yup, RC5-72 is still running nicely so I will stress test the card itself next after this batch of packets is finished. I have 16 left in the bucket.

FedEx is showing my two other cards as "on vehicle for delivery" so at least I will have other cards later today to test as well.
 
Welp, I stress tested with FurMark for an hour with no issues. I upgraded to the latest nvidia drivers, tried to fold again and it crashed once more. I don't get it.

Code: 
[14:14:55] Completed 5%
[14:14:55] mdrun_gpu returned
[14:14:55] NANs detected on GPU
[14:14:55]
[14:14:55] Folding@home Core Shutdown: UNSTABLE_MACHINE
[14:14:58] CoreStatus = 7A (122)
[14:14:58] Sending work to server
[14:14:58] Project: 5798 (Run 20, Clone 581, Gen 10)
 
You must log in or register to reply here.
Back
Top