OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.1

[Joshwaa]

You will be better off just buying the coins. The Radeon HD4xxx series is not worth it. You will pay more for power.

[os2sam]

Q: Which ATI SDK is the best for cgminer?
A: At the moment, versions 2.4 and 2.5 work the best. If you are forced to use
the 2.6 SDK, the phatk kernel will perform poorly, while the diablo or my
custom modified poclbm kernel are optimised for it.

I can't find any version of cgminer which would allow me to mine using Radeon HD4670 on WinXP SP3. I have Catalyst
version 11.12 installed, which is the last version supporting SDK on WinXP. SDK version is 2.5, OpenCL version is 1.1

I would like to setup both solo and pooled mining, with long-polling, as explained here:

https://bitcointalk.org/index.php?topic=111950.msg1215083#msg1215083

I am mining with an AGP Radeon 4670 in my print server.  I had a very hard time getting it to work at all.  I am currently getting about 34Mhs/s in a configuration that still lets me game at the same time.

Here is what I use to make it work and it is a bit off topic for this forum and I apologize in advance.

The ONLY Catalyst version I could get to work is the 10.7
The ONLY SDK I could get to work is 2.2
The miner I could get to work best is Phoenix 1.5 and 1.75
CGMiner only gets 20Mhs/s or so since Catalyst 10.7 does not have ADL support

The pool that works the best with this setup is Ozco, all other pools give me about 20-25% rejects.

If you need any other suggestions please PM me or take it the Phoenix 1.x thread.
Hope this helps,
Sam

Edit: Or we can work through it on this thread

https://bitcointalk.org/index.php?topic=23486.msg294832#msg294832

[slush]

Code:

 [2012-10-10 16:38:44] Stratum from pool 0 requested work restart
 [2012-10-10 16:38:44] Accepted 96dd686b Diff 1 BFL 0
 [2012-10-10 16:38:50] Accepted 45f658d1 Diff 1 BFL 0
 [2012-10-10 16:38:50] Accepted a2c9ffeb Diff 1 BFL 0
 [2012-10-10 16:38:55] Pool 0 http://192.168.3.9:3333 not responding!

It is already over 5 minutes and no reconnect so far :-(. It happen during pool restart (no client.reconnect() has been sent). Using latest git.

It was fault in my code, I'm sorry for the report. That happen when all pieces of used software are bleeding edge, one never know what exactly might be broken *right now*.

[eleuthria]

There is definitely something wrong in 2.8.1's release and dynamic difficulty adjustments.  It seems to happen at diff=4 (or possibly when they move to diff=8).  The users reporting the problem are getting stuck at difficulty of 4, on rigs that were at difficulty=8 or difficulty=16 prior to upgrading to 2.8.1.  This is reflected by a drastic drop in their hash rates (50% or 75% drop).

My best guess is the integer difficulty isn't adjusting to the proper target.  So users are working on a harder target than what they should be, reducing their share submissions.

[os2sam]

Q: Which ATI SDK is the best for cgminer?
A: At the moment, versions 2.4 and 2.5 work the best. If you are forced to use
the 2.6 SDK, the phatk kernel will perform poorly, while the diablo or my
custom modified poclbm kernel are optimised for it.

I can't find any version of cgminer which would allow me to mine using Radeon HD4670 on WinXP SP3. I have Catalyst
version 11.12 installed, which is the last version supporting SDK on WinXP. SDK version is 2.5, OpenCL version is 1.1

I would like to setup both solo and pooled mining, with long-polling, as explained here:

https://bitcointalk.org/index.php?topic=111950.msg1215083#msg1215083

As I stated before I am willing to try and help you out with this since I have some experience with this GPU.

What error's are you getting with CG Miner?
Are you able to mine with any miner with this GPU?
Is your GPU AGP or PCIe? hopefully the latter.
Sam

[slush]

Are all of these users using Stratum proxy in the chain?

[GernMiester]

I don't care, OK?
[/quote]

OK.
LOL only an idiot would not care about wasting money day in and day out for as long as they are dumb enough to continue.

[os2sam]

I don't care, OK?

OK.
LOL only an idiot would not care about wasting money day in and day out for as long as they are dumb enough to continue.

Well, I can't speak to his motivations as to why he want's to use his 4670 for mining.

But I started with CPU mining and then GPU mining with a 4670 as well.  It was a great training exercise and nobody would help me with it, over a year and a half ago, either.  So I can see doing it now too for the learning experience.

Now whether he want's to keep on using that for mining for the foreseeable future or not is worth debating the wisdom of.

I agree with his implication that we should be more understanding and helpful with noob's.
Sam

[Joshwaa]

You will be better off just buying the coins. The Radeon HD4xxx series is not worth it. You will pay more for power.

I don't care, OK?

Sorry for offering the advice.

[-ck]

There is definitely something wrong in 2.8.1's release and dynamic difficulty adjustments.  It seems to happen at diff=4 (or possibly when they move to diff=8).  The users reporting the problem are getting stuck at difficulty of 4, on rigs that were at difficulty=8 or difficulty=16 prior to upgrading to 2.8.1.  This is reflected by a drastic drop in their hash rates (50% or 75% drop).

My best guess is the integer difficulty isn't adjusting to the proper target.  So users are working on a harder target than what they should be, reducing their share submissions.

Bug acknowledged. Working on a fix and hopefully will release a 2.8.2 in the near future correcting this and the reconnection issues.

[Cranky4u]

Gneral question...is there any thoughts / work being done WRT when a miner goes offline (display = OFF) and then having cgminer rescan (say every 30 sec) to refind mining hardware?

I ask becasue every few days one of my CM1s drops offline but the other keeps mining. As I am at a day job, it has to wait for a soft reset until I get home and thereby loosing mining time.

Cheers.

[slush]

ckolivas, reconnecting demo extended for "client.get_version" message (sent after subscribe) is here: http://pastebin.com/Ai2htr3R

[Askit2]

Gneral question...is there any thoughts / work being done WRT when a miner goes offline (display = OFF) and then having cgminer rescan (say every 30 sec) to refind mining hardware?

I ask becasue every few days one of my CM1s drops offline but the other keeps mining. As I am at a day job, it has to wait for a soft reset until I get home and thereby loosing mining time.

Cheers.

Could stop cgminer after a set number of shares, and in a launching file put the actual application launch into a loop so that when the shares are hit cgminer stops, there is a short wait time then it restarts with the hardware running again.

It would in an odd way fix your problem but it doesn't answer your question directly.

[PandaMiner]

Well, I can't speak to his motivations as to why he want's to use his 4670 for mining.
...
Now whether he want's to keep on using that for mining for the foreseeable future or not is worth debating the wisdom of.

We also don't know if he's getting free power, we just assume. :p

[-ck]

New release - 2.8.2, 11th October 2012

The code is getting close to a point where I can declare the stratum support stable. Those testing stratum support are strongly recommended to upgrade since the variable diff support is broken on earlier releases.


Human readable changelog

Fixed a scenario where stratum might declare a pool dead and never try to reconnect to it.
Fixed the broken target support for difficulties >2
Add a --fix-protocol option to prevent cgminer from redirecting to stratum if you wish to stick to http / getwork.
Added support for mining.reconnect in stratum for when pools wish to move their stratum server.
Added support for client.get_version in stratum.
Try again to fix the dynamic intensity support on windows by adding long term history but limiting the amount it can change at any one time.
If a pool is switched to its stratum version, the original config version is reported (which helps puppetmaster control).


Full changelog

- Reinstate the history on dynamic intensity mode to damp fluctuations in
intensity but use an upper limit on how much the value can increase at any time
to cope with rare overflows.
- Create a fix-protocol option which prevents cgminer from switching to stratum
if it's detected.
- Simplify target generation code.
- Add support for client.get_version for stratum.
- Use a 64 bit unsigned integer on the diff target to generate the hex target.
- Update reconnect message to show whole address including port.
- Look for null values and parse correct separate array entries for url and port
with client reconnect commands for stratum.
- The command for stratum is client.reconnect, not mining.reconnect.
- Only copy the stratum url to the rpc url if an rpc url does not exist.
- Implement rudimentary mining.reconnect support for stratum.
- Ignore the value of stratum_active on calling initiate_stratum and assume
we're always trying to reinitiate it, and set the active flag to false in that
function.
- stratum auth can be unset if we fail to authorise on subsequent calls to
auth_stratum which undoes the requirement of setting it in one place so set it
in pool_active.

[kano]

cut/paste ...

2.8.2
An Xubuntu 11.04 x86_64 executable is in my github downloads called cgminer-2.8.2a
https://github.com/kanoi/cgminer/downloads
(it also works on Fedora 16 and 17)

For anyone who didn't realise, it's just the executable file to put in place of 'cgminer'
Nothing else needs changing
First get and extract the full binary release from ckolivas and then copy my file in place of 'cgminer'

No problems so far on my BFL or my '2xGPU+2xIcarus' (for 7 minutes so far ) on normal pools

The same configure options as cvolivas' binary version
In case anyone was wondering:
CFLAGS="-O2 -W -Wall" ./autogen.sh --enable-icarus --enable-bitforce --enable-ztex --enable-modminer --enable-scrypt
make clean
make

[dietwice]

Tried a windows version of 2.8.2 on 1 x CM1 + 1 x BFL Single on Win7 x64.
It starts as usual, attaches to stratum and makes a look of hard working, but never makes any shares, just takes Icarus Re-estimations periodically.
Ran it for 5 minutes - U is still 0, no shares.

O_o

Got back to 2.7.6 through mining proxy - it works well as usual.

[ice_chill]

Tried a windows version of 2.8.2 on 1 x CM1 + 1 x BFL Single on Win7 x64.
It starts as usual, attaches to stratum and makes a look of hard working, but never makes any shares, just takes Icarus Re-estimations periodically.
Ran it for 5 minutes - U is still 0, no shares.

O_o

Got back to 2.7.6 through mining proxy - it works well as usual.

I have the same problem with BFL Single and Windows 7 64bit, it looks like working, but no shares submitted and all values R/E/U etc are 0.

Will revert to earlier version.

[Frizz23]

How do I connect to BTC Guild using cgminer and Stratum protocol?

What I thought should work:
cgminer -o stratum+tcp://stratum.btcguild.com:9332 -u myWorker_01 -p secret -w 256 -v 2 -I 9

But with this command line cgminer only shows

Code:

 cgminer version 2.8.2 - Started: [2012-10-11 16:15:35]
--------------------------------------------------------------------------------
 (5s):773.1 (avg):815.5 Mh/s | Q:9  A:0  R:0  HW:0  E:0%  U:0.0/m
 TQ: 0  ST: 5  SS: 0  DW: 5  NB: 1  LW: 58  GF: 0  RF: 0  WU: 12.3
 Connected to stratum.btcguild.com with stratum as user myWorker_01
 Block: 000002995d7f5a59ea441833957a0f0d...  Started: [16:15:39]
--------------------------------------------------------------------------------
 [P]ool management [G]PU management [S]ettings [D]isplay options [Q]uit
 GPU 0:  68.5C 2140RPM | 419.7/423.3Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
 GPU 1:  71.0C 2032RPM | 410.6/412.8Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
--------------------------------------------------------------------------------

 [2012-10-11 16:31:32] Started cgminer 2.8.2
 [2012-10-11 16:31:33] Probing for an alive pool
 [2012-10-11 16:32:52] Stratum from pool 0 detected new block
 [2012-10-11 16:40:27] Stratum from pool 0 detected new block
 [2012-10-11 16:41:00] Stratum from pool 0 detected new block
 [2012-10-11 16:41:58] Stratum from pool 0 detected new block

Why do accepts/rejects stay at 0 ?

What am I doing wrong?
(My system: Win7x64, 2x6970)

[EDIT] Just downloaded 2.8.1 - and this version works OK.

[Steffer]

2.8.2 does not seem to work for me either.

Win 7 x64 with 1 7970

Copied old .conf from 2.8.1 ..