Ubuntu Natty Narwhal 11.04 Mining Guide / HOWTO

[Nivco]

Great HowTo Inaba, thanks a lot.
I followed step by step, had almost no trouble but a major pain in the a** with my extender (Ubuntu won't load properly Xorg, very weird stuff).
What's your adress for donation ? (why not in signature btw ?)

[Inaba]

Irishmick:  As far as other problems, people have had miriad of driver issues and library issues with the live installs.  I think most of those were ironed out last month, but new problems always seem to crop up and they are all solved with a normal install.  Live installs are great for CD's but horrible for anything else.

Nivco: Huh, I have never really thought about it ... I have signatures turned off as most people have obnoxious sigs with pictures and other crap in them that nobody but the person posting cares about.  I suppose I will add it to my signature though, for those that have them turned on.  I'm of the philosophy that if your signature is more than 4 lines long, it's 3 lines too long.  There's no reason to have a signature that's bigger than most posts, other than to look like a dork.

Donation/tip address is in the first post, but here it is: 168Bgg6HL6bLiUxQDBbdmHRbXtSsMixYCT

Thanks!

[Christian Pezza]

Hello

there is a way I can have one console and manage multiple GPU in the same machine or/and more machine(GPU'S) from may LAN?

ps Inaba thank you so much for your work and all of you partecipating of this forum!!! Just open up new opportunity for me!

asap I will donate but let me get some BTC first

[Inaba]

Not really sure what you are asking specifically?

There was/is a program floating around that does what you want I think, though I've never used it myself and I couldn't tell you what the name is off hand.  It's in the forums somewhere though.

[irishmick]

Thanks. Adding noprompt worked. Gratuity on its way to you.

[m0w3r]

The json-rpc.org site appears to be down.  Is there an alternative?  Thanks.

edit - this really sucks that I'm unable to continue due to this repository being down, now on 2nd day

[stellan0r]

Hello

there is a way I can have one console and manage multiple GPU in the same machine or/and more machine(GPU'S) from may LAN?

ps Inaba thank you so much for your work and all of you partecipating of this forum!!! Just open up new opportunity for me!

asap I will donate but let me get some BTC first

cp you can use the command "screen" in your terminal window / ssh session.
just connect to your miner via LAN (works also via internet if you have a static ip or a dynamic dns service ), enter "screen" and a new prompt appears. start your miner or whatever, and then hit ctrl+A (together, hold both) and then hit C. then a new prompt appears, you get a new tab in that terminal window. like tabed browsing in firefox (just without the tabs to click at). enter your commands here. open a new "window" with ctrl+A +C again and so on.

cycle through windows with ctrl+A (again hold both keys) +N (for next) or +P (for previous) to go forward and backwards through your tabs.

you can close the terminal window with the screen ssh session and resume it even after restarting the computer you used to ssh into the miner or from another machine, just connect to your miner again and type "screen -r" into your terminal. et voila, you can flip through your screen tabs again.

With this method you will need only 1 window per miner with as many gpus inside as you want.

[Flappy]

The json-rpc.org site appears to be down.  Is there an alternative?  Thanks.

edit - this really sucks that I'm unable to continue due to this repository being down, now on 2nd day

Me too.

[Inaba]

Ok, you can download the file from my site: https://eclipsemc.com/files/python-jsonrpc.tar.gz

You can then untar it in your directory and continue on instead of the svn checkout.

[CYPER]

Ok, you can download the file from my site: https://eclipsemc.com/files/python-jsonrpc.tar.gz

You can then untar it in your directory and continue on instead of the svn checkout.

Thanks you are a saviour 

[Inaba]

No problem... USE MY POOL!

[ThiagoCMC]

Guys!

 Why this guide does not use the "ati-driver-installer-11-6-x86.x86_64.run"?

 Sorry but, my 6990 arrives today and I'm already mining but, the system is pretty weird, the GPU0 reaches 395MH but the gpu1 reaches only 300MH...

 The temperature is around 89~83, is that normal?

 The cooler is running at high speed.

Thanks!
Thiago

[ThiagoCMC]

Hi!

 I'm running:

 ./phoenix.py -u http://X:Y@uswest.btcguild.com:8332/ -k phatk VECTORS AGGRESSION=14 BFI_INT WORKSIZE=128 FASTLOOP=false DEVICE=0

MHash: 395
Temp: 87.0 C

 ./phoenix.py -u http://X:Y@uswest.btcguild.com:8332/ -k phatk VECTORS AGGRESSION=14 BFI_INT WORKSIZE=128 FASTLOOP=false DEVICE=1

MHash: 290
Temp: 79

 What am I doing wrong?!

Thanks!!
Thiago

[Christian Pezza]

Hello

there is a way I can have one console and manage multiple GPU in the same machine or/and more machine(GPU'S) from may LAN?

ps Inaba thank you so much for your work and all of you partecipating of this forum!!! Just open up new opportunity for me!

asap I will donate but let me get some BTC first

cp you can use the command "screen" in your terminal window / ssh session.
just connect to your miner via LAN (works also via internet if you have a static ip or a dynamic dns service ), enter "screen" and a new prompt appears. start your miner or whatever, and then hit ctrl+A (together, hold both) and then hit C. then a new prompt appears, you get a new tab in that terminal window. like tabed browsing in firefox (just without the tabs to click at). enter your commands here. open a new "window" with ctrl+A +C again and so on.

cycle through windows with ctrl+A (again hold both keys) +N (for next) or +P (for previous) to go forward and backwards through your tabs.

you can close the terminal window with the screen ssh session and resume it even after restarting the computer you used to ssh into the miner or from another machine, just connect to your miner again and type "screen -r" into your terminal. et voila, you can flip through your screen tabs again.

With this method you will need only 1 window per miner with as many gpus inside as you want.

stellan0r, thank you for the tip I just was wishing to have a more friendly way of something cn regroup up to 15 GPU. It could become too busy. I got someone told me was possible but I have to just look in the net how... so just trying to get that way to do it make it practical

In the mean time I will practice what you suggest me!

[Christian Pezza]

help, I try to install another system and after all installing etc I am getting

poclbm.py: error: no such option: --SERVER

any advise please?

[ThiagoCMC]

help, I try to install another system and after all installing etc I am getting

poclbm.py: error: no such option: --SERVER

any advise please?

Just remove "--server"

./poclbm.py -d0 -f0 -w128 http://X:Y@uswest.btcguild.com:8332

[ThiagoCMC]

Hi!

 I'm running:

 ./phoenix.py -u http://X:Y@uswest.btcguild.com:8332/ -k phatk VECTORS AGGRESSION=14 BFI_INT WORKSIZE=128 FASTLOOP=false DEVICE=0

MHash: 395
Temp: 87.0 C

 ./phoenix.py -u http://X:Y@uswest.btcguild.com:8332/ -k phatk VECTORS AGGRESSION=14 BFI_INT WORKSIZE=128 FASTLOOP=false DEVICE=1

MHash: 290
Temp: 79

 What am I doing wrong?!

Thanks!!
Thiago

Hi! I figure out the following to stabilize:

 "Do not run the gnome / X applications while mining!"  \o/

1- reboot
2- login via SSH with root
3- run: service gdm stop
4- run: screen -aAdR
5- run: X

6- Press: control+a c
7- run: ./poclbm.py -d0 -f0 -w128 http://X:Y@uswest.btcguild.com:8332 - getting [357.657 MH/s (~400 MH/s)] - temp 89.50 C

8- Press: control+a c
9- run: ./poclbm.py -d1 -f0 -w128 http://X:Y@uswest.btcguild.com:8332 - getting [357.701 MH/s (~250 MH/s)] - temp 89.50 C

 But if I change the poclbm to phoenix, I got:

7- run: ./phoenix.py -u http://X:Y@uswest.btcguild.com:8332/ -k phatk VECTORS AGGRESSION=12 BFI_INT WORKSIZE=128 FASTLOOP=false DEVICE=0 - [395.46 Mhash/sec] - temp 98.00 C


9- run: ./phoenix.py -u http://X:Y@uswest.btcguild.com:8332/ -k phatk VECTORS AGGRESSION=12 BFI_INT WORKSIZE=128 FASTLOOP=false DEVICE=1 - [395.34 Mhash/sec] - temp 97.10 C

All coolers are running at 100%.

No overclock, I just bought the 6990 and install, all defaults from factory.


My question is: It is safe to run close to 100 ºC ??!

Code:

root@garimpeiro-2:~# aticonfig --odgc --adapter=all

Adapter 0 - AMD Radeon HD 6990
                            Core (MHz)    Memory (MHz)
           Current Clocks :    880           1250
             Current Peak :    880           1250
  Configurable Peak Range : [500-1200]     [1250-1500]
                 GPU load :    99%

Adapter 1 - AMD Radeon HD 6990
                            Core (MHz)    Memory (MHz)
           Current Clocks :    880           1250
             Current Peak :    880           1250
  Configurable Peak Range : [500-1200]     [1250-1500]
                 GPU load :    99%

 Second question: anybody running water-cooled 6990?
 I'm thinking in get one of these:

 http://www.koolance.com/water-cooling/product_info.php?product_id=1117
 http://www.swiftech.com/epsilonhd-6990fullcovervgawaterblock.aspx

 What do you guys think?!

Cheers,
Thiago

[Christian Pezza]

works thank you

about 100c from my reading and some users asked XFX customer service they don't suggest to go over 94c!

I already burn one 6950 because of it :/

[ThiagoCMC]

Now I chaged poclbm:

./poclbm.py -v -d0 -f0 http://X:Y@uswest.btcguild.com:8332 - [392.275 MH/s (~113 MH/s)] - 92.50 ºC

./poclbm.py -v -d1 -f0 http://X:Y@uswest.btcguild.com:8332 - [392.422 MH/s (~112 MH/s)] - 93.50 ºC

 I'm very concerned about the temperature... Please, some advice?! Bought a watercooler?

 I'll disable "-v" for instance...

Thanks!
Thiago

[ThiagoCMC]

Shit! $#@%#$$#$%@#$%@F   

Now even with no "-v", the temperature is 94.00 ºC for GPU0 and 95.50 ºC for GPU1...

Damn it