Bitcoin Forum
November 14, 2024, 05:32:49 PM *
News: Check out the artwork 1Dq created to commemorate this forum's 15th anniversary
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: 5970 dead?  (Read 1650 times)
Tesla71 (OP)
Sr. Member
****
Offline Offline

Activity: 302
Merit: 252



View Profile
March 05, 2013, 09:00:31 AM
 #1

Hi, I am running atm a mainboard with a 5870 and win 7 without problems.

Now I got a used 5970 and wanted to test it. Win7 and catalyst 13.1 regognized it correctly but if I try to run cgminer,
the screen went shortly dark and comes back up with the notification "Display Driver Stopped Responding and has Recovered"
I read that this may be a common win 7 error, but as my 5870 is running OK, I don't know if I should try maybee with Linux or if the card is surely dead.. (PSU is 700W, so that should be OK)

-ck
Legendary
*
Offline Offline

Activity: 4284
Merit: 1645


Ruu \o/


View Profile WWW
March 05, 2013, 09:07:19 AM
 #2

13.1 is broken shit. Try 12.8

Developer/maintainer for cgminer, ckpool/ckproxy, and the -ck kernel
2% Fee Solo mining at solo.ckpool.org
-ck
r3demon
Newbie
*
Offline Offline

Activity: 45
Merit: 0



View Profile
March 05, 2013, 12:00:20 PM
 #3

Try underclocking it, and it's always good to check if heatsink pads are installed properly.
Tesla71 (OP)
Sr. Member
****
Offline Offline

Activity: 302
Merit: 252



View Profile
March 05, 2013, 12:12:09 PM
 #4

Try underclocking it, and it's always good to check if heatsink pads are installed properly.

it is not OC'd now but will try to underclock, also will try Catalyst 12.8.
Heatsink should be at least no matter for the test because the failure happend immediately, with card still at around 40 deg C.

Aseras
Hero Member
*****
Offline Offline

Activity: 658
Merit: 500


View Profile
March 05, 2013, 02:08:41 PM
 #5

The 1st GPU VRM are toast or dying.  Run cgminer with -d 1 -c conf.conf 

if it runs, then the second gpu is fine. You can try dropping the voltage/clocks to the first gpu or keeping it disabled and only using 1/2 the card.
Tesla71 (OP)
Sr. Member
****
Offline Offline

Activity: 302
Merit: 252



View Profile
March 06, 2013, 07:57:09 AM
 #6

The 1st GPU VRM are toast or dying.  Run cgminer with -d 1 -c conf.conf 

if it runs, then the second gpu is fine. You can try dropping the voltage/clocks to the first gpu or keeping it disabled and only using 1/2 the card.

I think you are right! if I run cgminner -d 1 the second GPU works perfectly.
If I try cgminer -d 0 I have to add -I 10, then it works very slowly, but only for 5 or 10 seconds, then got that error again (regardless of which clock settings)

Any chances to fix the 5970 or is the first GPU-chip itself the Problem?


r3demon
Newbie
*
Offline Offline

Activity: 45
Merit: 0



View Profile
March 06, 2013, 11:05:31 AM
 #7

That's pretty common problem for 5970's, the GPU is probably fine, but VRM chip/chips are dead. You can try to replace VRMs, if you can find some spare cards lying around.
Tesla71 (OP)
Sr. Member
****
Offline Offline

Activity: 302
Merit: 252



View Profile
March 14, 2013, 02:43:55 PM
 #8

So, card seems fixed, don't laugh I put it in the oven for 30 min@160 deg C.

Now its running for over a week continously without error  Grin
Gator-hex
Hero Member
*****
Offline Offline

Activity: 490
Merit: 500


View Profile
March 14, 2013, 04:40:18 PM
 #9

Quote
don't laugh I put it in the oven for 30 min@160 deg C.

I give it 3 months, hot sunny weather, and it'll be dead again. Baking never lasts.

Tesla71 (OP)
Sr. Member
****
Offline Offline

Activity: 302
Merit: 252



View Profile
March 15, 2013, 06:10:56 AM
 #10

yeah maybee, but if it lasts until the diff is unprofitable at least (LTC)....
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!