Author

Topic: Help locating bottleneck on new mining rig (Read 1496 times)

-ck
legendary
Activity: 4088
Merit: 1631
Ruu \o/
January 19, 2012, 09:53:17 PM
#7
We're quite the cgminer evangelists, huh DAT?

I'll just add this:
An ancient sysadmin technique is to evade a non-trivial bug rather than tackle it head-on.
Cgminer allows you to solve quite a few problems gracefully, using just one piece of code.
Grin
full member
Activity: 210
Merit: 100
January 13, 2012, 07:08:22 PM
#6
That's PRECISELY why I'm always bashing Windows on these forums.
No, I don't think Microsoft is the devil, I am using Windows 7 on my laptop right now.
I've put in a lot of work hacking and customizing it and I'm very happy with how it works.

The issue is, Windows is a general-purpose OS, written so that your grandmother is able to use it, the temp office girls are able to use it, and a 8-year-old is able to use it.
It is just horribly out of place in a heavily task-oriented mining rig.

DAT, a couple of days ago I helped out a guy here who thought his mining rig was crashing and powering off.
As it turned out, it was going into freaking STANDBY after 30 minutes.
That poor soul was ripping his mining rig apart and rebuilding it for a whole darned day and definitely NOT because he's an idiot.
He's non-technical, to be sure, but his mistake was choosing a bad tool for the job.
donator
Activity: 1218
Merit: 1079
Gerald Davis
January 13, 2012, 06:53:50 PM
#5
Yeah the alternate I have seen when people run into weird bugs like this (and cgminer may not fix it) is a re-install of os & drivers.

You may "work" on it for days without accomplishing anything.  The more direct answer is there is no hardware limitation that would cause that.  It is screwed up OS/drivers/OpenCL SDK.  My guess it is a windows box and when they get "hosed" they tend not to fix themselves.  He can try uninstalling the drivers and reinstalling, and trying to copy dll and remove cards, and do reboot, and use driver cleaning software and a couple weeks later give up and reinstall OS and it will work perfectly right away.  Smiley  I recall someone else having similar issue recently and indicated problem went away after a re-install.
full member
Activity: 210
Merit: 100
January 13, 2012, 06:50:09 PM
#4
We're quite the cgminer evangelists, huh DAT?

I'll just add this:
An ancient sysadmin technique is to evade a non-trivial bug rather than tackle it head-on.
Cgminer allows you to solve quite a few problems gracefully, using just one piece of code.
donator
Activity: 1218
Merit: 1079
Gerald Davis
January 13, 2012, 06:39:26 PM
#3
use cgminer.  I still don't understand why people mess will multiple instances.

To answer your direct quesiton there is no bottleneck.  You likely got a driver, sdk, OS "glitch". 

full member
Activity: 210
Merit: 100
January 13, 2012, 06:34:54 PM
#2
It might be some weird spin-lock problem in AMD's software.
Which driver and SDK version are you using, BTW?

My suggestion is, download cgminer and give it a try.
Just one instance of the miner will support those cards, each of them can be individually overclocked and temperature-controlled.

EDIT :: I posted my reply on SE as well.
sr. member
Activity: 444
Merit: 313
January 13, 2012, 02:40:08 PM
#1
Originally asked on Bitcoin SE by Jon Garvin:

http://bitcoin.stackexchange.com/questions/2562/help-locating-bottleneck-on-new-mining-rig
Quote
I built a mining rig with 3 Radeon 6850 GPUs. When I run one miner (either Phoenix or m0mchil's poclbm) it reports a hash rate of a little over 200 Mhash/sec. When I fire up a second miner pointed at a different GPU, the hash rate of both miners is about half what the one running alone would get. When I fire up a third pointed at the last GPU, all three miners report an even lower hash rate, such that the total is around the 200 Mhash/sec that a solo miner would get.

Initially, I thought that the miners were ignoring the DEVICE=# argument and all using the same GPU. I posted a question seeking help in figuring out why they would do that. @Lodewijk's answer there clued me in to the fact that the miners are using separate GPUs and that there's a bottleneck elsewhere in the system.

So, my question is, what's the likely bottleneck?

I used the recommendations at the bitcoin.it wiki entry on Mining Rigs to pick out my hardware. I'm running on a Ubuntu 11.10 system, with an msi 890FXA-GD70 motherboard, 2GB Kingston 1333MHz DDR3 Non-ECC RAM, 80GB WD IDE Drive, and a 2.8Ghz AMD Semperon 145 processor. It's on a cable Internet connection with tons of bandwidth and I'm connecting to the Arsbitcoin mining pool.
Jump to: