Bitcoin Forum
May 30, 2024, 01:23:08 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: AMD Driver crash on 4x6 GPUs  (Read 1238 times)
GZ (OP)
Newbie
*
Offline Offline

Activity: 1
Merit: 0


View Profile
December 28, 2013, 11:47:08 PM
 #1

I have 4 mobos powering 18-24 280x
The first 2 machines are running 6 GPU on EVGA Z68 FTW
One of these has 3 GPU installed one is crashed on boot saying "not enough resources" the next crashes on cgminer launch and then crashes cgminer moments later in the attempt to restart it, currently that machine can only mine with -d 0 command at about 500 kh
The next machine has 6 plugged in with 5 in the devices manager and crashes mid video driver install windows reboots several times with no display until it reverts to known working config from pre-install of video drivers
The other 2 machines are MSI Z77A-G45 with 3 x GPU I have yet successfully boot these as I have been focusing on the EVGAs which can run more, they will be 6x when I get more risers, yet i would like these running 3x tell the risers arrive.
All of my problems that i can’t quickly get past have to do with drivers crashing.

Specs:
celeron G1610 x4
8gb RAM x4
4x 860w corsair PSU for MOBO and riser power
8x 1000w Antec PSU for GPUs
2x EVGA z68 FTW mobos
2x MSI Z77A-G45
4x 64gb SSDs
4x Windows 8.1
24x R9 280x various manufacturers (Asus, Sapphire, Gigabyte, XFX, MSI, Club, Powercolor)
enough 16x to 16x PCI-E powered risers (all powered risers on all GPUs)
1x USB mouse/keyboard 1x DVI-D/D-Sub Monitor
splawik21
Legendary
*
Offline Offline

Activity: 1372
Merit: 1005


DASH is the future of crypto payments!


View Profile
January 30, 2014, 02:48:00 PM
 #2

I think that the mouse cause the problems Wink

Have you tried to put intensity down and clocks with memory? Do you use powerder risers? If not use them!


BE SMART, USE DASH ( ͡° ͜ʖ ͡°)
pontiacg5
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250



View Profile
January 30, 2014, 02:56:28 PM
 #3

I think you using windows is part of the problem. I've never seen 6x r9 cards working in any windows, but 5x is possible.

I have 6x sapphire 280x working on z77-gd45 no problem, but in linux. Had 6 gigabyte 280 cards for some time, as well as asus 280's too. Still have plenty of 290's working on gd45's though.

Ribbon riser cables are total crap, and x16-x16 versions are the absolute worst of the worst. Why did you do that? You don't need x16, now you've got cards trying to run full bandwidth through shitty unshielded riser cables with 15x more lanes than necessary. If a single solder connection is broken, it's likely that it would kill the entire pci-e bus.

That is your problem, almost with a certainty. I'd bet if you go through and install x1 presence detect jumpers on most all pci-e slots you'd have it working in no time. I'd bet if you had bought USB riser cables from the start you would be mining right now.

Please DO NOT send me private messages asking for help setting up GPU miners. I will not respond!!!
Smileyftw
Member
**
Offline Offline

Activity: 70
Merit: 10


View Profile
January 31, 2014, 06:34:30 AM
 #4

Check for any .dmp files in the C:/ Windows/Minidump folder.
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!