Phoenix - Efficient, fast, modular miner - page 23.

finnthecelt

full member

Activity: 140

Merit: 100

Quote from: SchizophrenicX on June 14, 2011, 09:59:10 AM

constantly getting failed to connect: retrying message on Ubuntu 11.04. Followed the guide around here.
poclbm connects normally. so I'm guessing I'm missing something...

Yeah, windows.

SchizophrenicX

member

Activity: 112

Merit: 100

"I'm not psychic; I'm just damn good"

constantly getting failed to connect: retrying message on Ubuntu 11.04. Followed the guide around here.
poclbm connects normally. so I'm guessing I'm missing something...

jedi95

full member

Activity: 219

Merit: 120

AGGRESSION is used to set the number of nonces to run per kernel execution. Values above 16 won't do anything because nonces to run = 2^(16 + AGGRESSION) Setting a higher value for AGGRESSION will simply round down to the highest valid value (16)

The -a parameter is used for the hashrate display. With higher aggression settings kernels can take several seconds to run, so this makes taking 16 samples rather pointless. The reason multiple samples are used is that otherwise the displayed hashrate would fluctuate greatly at low aggression.

The -f option in poclbm is somewhat similar to aggression because it also controls kernel execution sizes. The main difference is that aggression sets the number of nonces to run per execution while -f sets the target time for each execution. (-f adjusts the number of nonces constantly to maintain the target execution time)

As for updates to Phoenix, I think I have a solid workaround for the idle problem now. I still don't know the exact reason for it, but the workaround should be sufficient to recover automatically. I will release this is 1.50 shortly if I don't find any issues in testing. These changes are included in the latest SVN revision (r101) if anyone wants to try it.

supa

copper member

Activity: 56

Merit: 0

I tried to skim through the thread but I still have a couple of questions regarding the kernel AGGRESSION parameter and -a (average samples).

For starters... In Windows, if I set my aggression higher than 15, it seems the whole thing explodes. 15 works OK with a laggy display (expected).

In Linux, I don't care about the display, so I put my aggression to 20. Linux gets a higher hash rate. I'm not sure anything over 15 actually does anything, but 20 seemed like a good number at the time....

What is this doing... ?

Is there a relation to this and the -f (frames) in poclbm? A windows machine with poclbm -f 0 is still mostly usable.

Finally, according to this -

Quote

-a (average samples) - Sets the number of samples to use for hashrate averaging. Default is 10. You might want to lower this for longer kernel execution times. (high aggression)

I should make a lower -a for higher aggression. I tried setting it to "-a 1" without much difference and even "-a 0". What does this do?

Using phatk kernel if it matters.

AngstHase

newbie

Activity: 39

Merit: 0

any chance to get a phoenix update soon?

SchizophrenicX

member

Activity: 112

Merit: 100

"I'm not psychic; I'm just damn good"

Would you guys recommend LinuxCoin?

I have been trying to get my Ubuntu up but to no avail. LiveCD installs on 4 GB drives but I think a full install requires > 4.4 GB

Mousepotato

hero member

Activity: 896

Merit: 1000

Seal Cub Clubbing Club

I just downclocked my memory to 300MHz (down from 1200MHz) and I picked up a good 22-23Mh/s. Sweet!

finnthecelt

full member

Activity: 140

Merit: 100

Quote from: Mato Kira on June 10, 2011, 09:33:59 PM

Quote from: 5tr1k3r on June 10, 2011, 08:36:35 PM

Quote from: Mato Kira on June 10, 2011, 07:58:27 PM

I'm running a 9800M GTS, only getting about 14.8Mh/s with AGGRESSION=3 and WORKSIZE=64. Anyone have any suggestions for upping this?

Unfortunately, there is only one suggestion — buy a decent amd card, and preferably not a mobile one. https://en.bitcoin.it/wiki/Mining_hardware_comparison

Kinda locked into a solution with it being a laptop and all.. are there any effective ways to overclock?

http://www.overclock.net/nvidia-drivers-overclocking-software/461651-9800m-gts-overclocking-possible.html

jedi95

full member

Activity: 219

Merit: 120

Quote from: MintCondition on June 11, 2011, 03:33:57 PM

Quote from: MintCondition on June 09, 2011, 11:56:38 PM

The immediate solution is to use Phoenix's undocumented lpaskrate command line parameter, which is the Long Poll equivalent of the askrate parameter. A more structural fix would be for the Phoenix client to change lpaskrate default value from 0 to something sensible in the range 10-119. Of course the expiry policy may also be changed by the pool operator. But the default Pushpool work expiry policy is not overly strict in my opinion; work that's over 2 minutes old misses out on a lot of recent transactions and therefore on the transaction fees these include.

An update on this bug: Upon further research into the Phoenix code it seems that specifying the lpaskrate argument does not actually solve anything, so please ignore the quoted solution. I originally assumed the queuesize was intended as both preferred and maximum size, so that the oldest work would be abandoned when overfilling the queue. Instead, current work will only be abandoned when a new block has been detected; any forced getwork responses will just be added to the workqueue, only making the problem worse.

A better avenue for improvement, although more involved, might be to add an expiration mechanism to the Phoenix code. Until then I'd recommend never mining at less than 36MH/s to avoid rejected shares caused by this bug.

This is not accurate concerning the queue. The queue itself uses the deque class which has a maximum size set to the size the user specifies. Adding additional work to queue will automatically purge the oldest work if the queue is full. See the documentation for deque here:

Quote

If maxlen is not specified or is None, deques may grow to an arbitrary length. Otherwise, the deque is bounded to the specified maximum length. Once a bounded length deque is full, when new items are added, a corresponding number of items are discarded from the opposite end. Bounded length deques provide functionality similar to the tail filter in Unix. They are also useful for tracking transactions and other pools of data where only the most recent activity is of interest.

http://docs.python.org/library/collections.html#collections.deque

jondecker76

full member

Activity: 238

Merit: 100

For a little bit while deepbit was down today, I decided to try solo mining... DUring the short period that I tried it (About 10 minutes), phoenix kept reporting "Result didn't meet full difficulty, not sending"

what does this mean? I was showing a normal hash rate, and showing 0 accepter, 0 rejected. I've never seen this before mining in any pool.

thanks!

MintCondition

legendary

Activity: 1147

Merit: 1007

Quote from: MintCondition on June 09, 2011, 11:56:38 PM

The immediate solution is to use Phoenix's undocumented lpaskrate command line parameter, which is the Long Poll equivalent of the askrate parameter. A more structural fix would be for the Phoenix client to change lpaskrate default value from 0 to something sensible in the range 10-119. Of course the expiry policy may also be changed by the pool operator. But the default Pushpool work expiry policy is not overly strict in my opinion; work that's over 2 minutes old misses out on a lot of recent transactions and therefore on the transaction fees these include.

An update on this bug: Upon further research into the Phoenix code it seems that specifying the lpaskrate argument does not actually solve anything, so please ignore the quoted solution. I originally assumed the queuesize was intended as both preferred and maximum size, so that the oldest work would be abandoned when overfilling the queue. Instead, current work will only be abandoned when a new block has been detected; any forced getwork responses will just be added to the workqueue, only making the problem worse.

A better avenue for improvement, although more involved, might be to add an expiration mechanism to the Phoenix code. Until then I'd recommend never mining at less than 36MH/s to avoid rejected shares caused by this bug.

CubedRoot

sr. member

Activity: 291

Merit: 250

CAn someone tell my why a LP push like the ones below locks my GPU down, and forces me to restart?

I am using Ubuntu 64, newest phoenix and phatk kernel

Code:

[11/06/2011 14:42:10] Result: ac8fb0cd accepted
[11/06/2011 14:42:18] Result: a42fd6a2 accepted
[11/06/2011 14:42:35] Result: 652fb629 accepted
[11/06/2011 14:49:17] LP: New work pushed
[11/06/2011 14:52:32] LP: New work pushed
[11/06/2011 14:56:03] LP: New work pushed
[11/06/2011 14:57:19] LP: New work pushed
[309.70 Mhash/sec] [10 Accepted] [0 Rejected] [RPC (+LP)]

asdfg

newbie

Activity: 2

Merit: 0

Quote from: Etheric on June 11, 2011, 01:56:10 PM

Quote from: asdfg on June 11, 2011, 12:20:34 PM

Traceback (most recent call last):
File "phoenix.py", line 123, in
File "Miner.pyc", line 75, in start
File "phoenix.py", line 111, in makeKernel
File "kernels\poclbm\__init__.py", line 22, in
import pyopencl as cl
File "zipextimporter.pyc", line 82, in load_module
File "pyopencl\__init__.pyc", line 3, in
File "zipextimporter.pyc", line 98, in load_module
ImportError: MemoryLoadLibrary failed loading pyopencl\_cl.pyd

Huh

I had this problem too initially. I think it is related to the SDK driver not being properly installed.

I fixed it by uninstalling my current drivers, booting into safe mode, and using Driver Sweeper to remove all traces of old drivers and then installing the full 11.5 Catalyst suite.

Hope that helps you too!

Updated my drivers and it works, thank you.

Etheric

newbie

Activity: 46

Merit: 0

Quote from: asdfg on June 11, 2011, 12:20:34 PM

Traceback (most recent call last):
File "phoenix.py", line 123, in
File "Miner.pyc", line 75, in start
File "phoenix.py", line 111, in makeKernel
File "kernels\poclbm\__init__.py", line 22, in
import pyopencl as cl
File "zipextimporter.pyc", line 82, in load_module
File "pyopencl\__init__.pyc", line 3, in
File "zipextimporter.pyc", line 98, in load_module
ImportError: MemoryLoadLibrary failed loading pyopencl\_cl.pyd

Huh

I had this problem too initially. I think it is related to the SDK driver not being properly installed.

I fixed it by uninstalling my current drivers, booting into safe mode, and using Driver Sweeper to remove all traces of old drivers and then installing the full 11.5 Catalyst suite.

Hope that helps you too!

asdfg

newbie

Activity: 2

Merit: 0

Traceback (most recent call last):
File "phoenix.py", line 123, in
File "Miner.pyc", line 75, in start
File "phoenix.py", line 111, in makeKernel
File "kernels\poclbm\__init__.py", line 22, in
import pyopencl as cl
File "zipextimporter.pyc", line 82, in load_module
File "pyopencl\__init__.pyc", line 3, in
File "zipextimporter.pyc", line 98, in load_module
ImportError: MemoryLoadLibrary failed loading pyopencl\_cl.pyd

Huh

Mato Kira

newbie

Activity: 2

Merit: 0

Quote from: 5tr1k3r on June 10, 2011, 08:36:35 PM

Quote from: Mato Kira on June 10, 2011, 07:58:27 PM

I'm running a 9800M GTS, only getting about 14.8Mh/s with AGGRESSION=3 and WORKSIZE=64. Anyone have any suggestions for upping this?

Unfortunately, there is only one suggestion — buy a decent amd card, and preferably not a mobile one. https://en.bitcoin.it/wiki/Mining_hardware_comparison

Kinda locked into a solution with it being a laptop and all.. are there any effective ways to overclock?

memvola

hero member

Activity: 938

Merit: 1002

Quote from: MintCondition on June 09, 2011, 11:56:38 PM

I've been playing around with Pushpool (software to run a mining pool) in combination with Phoenix, and found that the default Phoenix settings may cause some unnecessary rejected shares for slow miners on a Pushpool server that offers Long Polling. Since several Pushpool-based mining pools have recently come into existence this probably affects some users already.

...

The immediate solution is to use Phoenix's undocumented lpaskrate command line parameter, which is the Long Poll equivalent of the askrate parameter. A more structural fix would be for the Phoenix client to change lpaskrate default value from 0 to something sensible in the range 10-119. Of course the expiry policy may also be changed by the pool operator. But the default Pushpool work expiry policy is not overly strict in my opinion; work that's over 2 minutes old misses out on a lot of recent transactions and therefore on the transaction fees these include.

Thanks for the hint. I've been getting a very high rejected rate (50%-75%) on slower miners using phoenix. Doesn't happen while using poclbm for instance...

Looks like lpaskrate was introduced in 1.48, but it didn't solve my problems. Options I use are roughly:

Code:

./phoenix.py AGGRESSION=9 WORKSIZE=256 -u http://1XXX0pps:[email protected]:8332/;lpaskrate=20 DEVICE=0

The same with swepool. I tried reducing askrate to 5 as well. Any ideas why this didn't work?

5tr1k3r

newbie

Activity: 39

Merit: 0

Quote from: Mato Kira on June 10, 2011, 07:58:27 PM

I'm running a 9800M GTS, only getting about 14.8Mh/s with AGGRESSION=3 and WORKSIZE=64. Anyone have any suggestions for upping this?

Unfortunately, there is only one suggestion — buy a decent amd card, and preferably not a mobile one. https://en.bitcoin.it/wiki/Mining_hardware_comparison

Mato Kira

newbie

Activity: 2

Merit: 0

I'm running a 9800M GTS, only getting about 14.8Mh/s with AGGRESSION=3 and WORKSIZE=64. Anyone have any suggestions for upping this?

mosie

newbie

Activity: 15

Merit: 0

Quote from: keybaud on June 10, 2011, 08:54:20 AM

Quote from: derbe on June 10, 2011, 05:56:42 AM

Quote from: maxcorrads on June 10, 2011, 05:49:29 AM

Quote from: derbe on June 10, 2011, 05:39:31 AM

Hey guys,

i got a big problem with my 5850 Cards, i installed a fresh win7 64x system with 2x of ATI 5850 card.
Ok, i download the lastest driver catalyst driver for ati und SDK. Than the system installed this cards correct.
Up to here everything is fine.

Now i start Phoenix and it says that i only have one card (id = 0 ->GPU , and id = 1 ->CPU)
it says that at starting the programm in dos windows, before processing.

But where is my 2nd GPU? (It is full installed and also shown in msi afterbruner e.g.).
Or do i have to use a crossfireX hardware brigde? (I do not have any installed, and iam not able to active this feature at the drivers)?!

Sorry for my bad english

It is not my mother speech.

You need to connect a monitor on the second card, or a dummy plug

Dam, ok now i know why i used nvida all of my life Cheesy

Ok i found some nice:
http://www.overclock.net/folding-home-guides-tutorials/384733-30-second-dummy-plug.html

THX for respone Wink

Nothing to do with ATI or NVidia, it's a Windows limitation.

Sure !

Topic: Phoenix - Efficient, fast, modular miner - page 23. (Read 760701 times)