Bitcoin Forum
June 25, 2024, 07:30:58 AM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: Cudaminer crashing  (Read 708 times)
laysdarren (OP)
Newbie
*
Offline Offline

Activity: 2
Merit: 0


View Profile
January 27, 2014, 12:06:09 AM
 #1

 Hello, I'm new to the forums and not sure how to post in the cudaminer thread, so I figured I'd post here and see if anyone could help me out.

I have SLI disabled, that's not the problem either.



I downloaded the newest version of cudaminer, and no matter what setting I use, it crashes either my display driver, or cudaminer itself. I've heard the newest version of cudaminer gives a bit more khash/s and I'd like to see how much I can get out of my 780's.  They get about 550-570 khash/s each right now overclocked.

 I'm not sure what to do, because the 11-20-2013 version works fine for me with settings:

color 0A
taskkill /f /im "cudaminer.exe"
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
 
 
CD "C:\Users\Darren\Desktop\Cudaminer\x64"
cudaminer -H 1 -m 1,1 -d 0,1 -i 0,0 -l T24x16,T24x16 -C 0,0 -o stratum+tcp://doge.netcodepool.org:4093 -u Lays.1 -p x

I can get about 1100 khash/s with both cards total



If I try and use those exact same settings with the new version of cudaminer, it instantly crashes my display driver and stops mining.

I've tried running my cards @ stock speed, and re-installing drivers, trying different settings. Nothing has worked, the only things that do work are settings that result in me getting like 200khash/s per card.


Temps aren't an issue either, everything's in a custom loop with 2 480mm 60mm thick radiators.

System specs:

i7 3930k @ 4.8ghz
2 Evga gtx 780 SC
32gb 1866mhz 4x8gb RAM
Rampage iv extreme
sjc4llc
Newbie
*
Offline Offline

Activity: 8
Merit: 0


View Profile
January 27, 2014, 04:15:19 AM
 #2

Make sure you have cuda 5.5 for the latest version https://developer.nvidia.com/cuda-toolkit. Otherwise stay with the old version if it does not crash. I had 1 or 2 versions that did not work for me.

cbuchner1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 502


View Profile
January 27, 2014, 07:51:02 AM
 #3

Make sure you have cuda 5.5 for the latest version https://developer.nvidia.com/cuda-toolkit. Otherwise stay with the old version if it does not crash. I had 1 or 2 versions that did not work for me.



don't run two devices with one cudaminer. This seems to be not working ever since I upgraded cudaminer to use CUDA 5.5. Run two cudaminer instances, one with -d 0, one with -d 1
laysdarren (OP)
Newbie
*
Offline Offline

Activity: 2
Merit: 0


View Profile
January 28, 2014, 01:51:50 AM
 #4

Make sure you have cuda 5.5 for the latest version https://developer.nvidia.com/cuda-toolkit. Otherwise stay with the old version if it does not crash. I had 1 or 2 versions that did not work for me.



don't run two devices with one cudaminer. This seems to be not working ever since I upgraded cudaminer to use CUDA 5.5. Run two cudaminer instances, one with -d 0, one with -d 1



Alright, I'll do that and repost what happens.

Smileyftw
Member
**
Offline Offline

Activity: 70
Merit: 10


View Profile
January 28, 2014, 02:04:43 PM
 #5

try lowering the blocks and warps values until its stable
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!