[ANN] sgminer with X11/X13/Nist5/Quark/Anime kernels compatible 14.6 amd drivers

[oktay50000]

i upgraded my farm with this miner and driver to 14.6
linux rigs works fine for some time and then droping hash rate after waiting work ....

windows is better but sometimes getting waitin for work

any fix for linux
?

[1713910801]

1713910801

[1713910801]

1713910801

[volodzya]

@bullus,

can you post your config ?

[bullus]

@bullus,

can you post your config ?

Code:

],
"kernel" : "x11mod",
"api-allow" : "W:127.0.0.1",
"api-listen" : true,
"api-mcast-port" : "4028",
"api-port" : "4028",
"expiry" : "28",
"failover-only" : true,
"kernel-path" : "/usr/local/bin",
"log" : "5",
"queue" : "1",
"scan-time" : "7",
"scrypt" : true,
"shares" : "0",
"failover-switch-delay" : "10",
"gpu-threads" : "2",
"gpu-dyninterval" : "7",
"gpu-engine" : "0",
"gpu-fan" : "55",
"gpu-platform" : "0",
"gpu-memclock" : "0",
"gpu-memdiff" : "0",
"gpu-powertune" : "8",
"gpu-vddc" : "0",
"intensity" : "18",
"temp-target" : "75",
"temp-overheat" : "85",
"temp-cutoff" : "95",
"temp-hysteresis" : "3",
"vectors" : "1",
"worksize" : "64",
"lookup-gap" : "0",
"shaders" : "1536",
"thread-concurrency" : "8000",
"no-pool-disable" : true,
"device" : "0"
}

6970@950/1375~2300kh/s x11

[volodzya]

@bullus,

can you post your config ?

Code:

],
"kernel" : "x11mod",
"api-allow" : "W:127.0.0.1",
"api-listen" : true,
"api-mcast-port" : "4028",
"api-port" : "4028",
"expiry" : "28",
"failover-only" : true,
"kernel-path" : "/usr/local/bin",
"log" : "5",
"queue" : "1",
"scan-time" : "7",
"scrypt" : true,
"shares" : "0",
"failover-switch-delay" : "10",
"gpu-threads" : "2",
"gpu-dyninterval" : "7",
"gpu-engine" : "0",
"gpu-fan" : "55",
"gpu-platform" : "0",
"gpu-memclock" : "0",
"gpu-memdiff" : "0",
"gpu-powertune" : "8",
"gpu-vddc" : "0",
"intensity" : "18",
"temp-target" : "75",
"temp-overheat" : "85",
"temp-cutoff" : "95",
"temp-hysteresis" : "3",
"vectors" : "1",
"worksize" : "64",
"lookup-gap" : "0",
"shaders" : "1536",
"thread-concurrency" : "8000",
"no-pool-disable" : true,
"device" : "0"
}

6970@950/1375~2300kh/s x11

how many cards do you have in the rig ? are all 6970 ? what cpu / ram have you ?

p.s. thnx for the reply

[Spacewide]

Hello

Sharing my x11 .bat for 5x R9 280X RoyalQueen Club3D.
4.45mh/s

Code:

setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
sgminer.exe -k x11mod -o stratum+tcp://eu.wafflepool.com:3331 -u xxx -p 0.16 -I 18,18,17,18,18 -w 256 --thread-concurrency 8192 --lookup-gap 2 -g 2 --gpu-engine 1150,1150,1150,1150,1150 --gpu-memclock 1500,1500,1500,1500,1500 --auto-fan --gpu-fan 55-85 --temp-cutoff 80 --temp-overheat 76 --temp-target 73 --no-submit-stale --no-pool-disable --api-listen --api-allow W:127.0.0.1

Runs like a tank.

G3220 intel CPU
8GB Ram
64GB SSD
Asrock BTC H81
x R9 280X RoyalQueen Club3D.


The api-listen and api-allow localhost is implemented to work with http://cgminermonitor.com, so obsolete.

[innergy]

@djm34

Thank you a lot for this great sgminer 

I fixed the idle/work resuming problem.

This is the download link for the fixed miner: https://mega.co.nz/#!ZsJTALQI!qDh9jCWULjabTDPYq0K4fZamdZrPwot7m7AAfTEOJPQ


I also opened a new thread (just in case) some will not notice this comment and the fixed version of the miner: https://bitcointalk.org/index.php?topic=658975.msg7412560#msg7412560

The fixed miner is not available anymore.
Could someone re-upload it. Thanks.

[gmerk]

I found the opencl files in AMD folder inside \Packages\Drivers\Display\WB6A_INF\B171094 (after decompressing the 14.6 driver package without installing it, or with 7zip)
I don't know which are needed, so I copied everything that had "opencl" or "amdocl" in the filename into the sgminer folder (8 files, dlls & exes).
Deleted all previous bins, and now it's even better (+3%) than using the bins from 14.6 (I have 14.4 drivers installed).

Did this... 14.4 drivers with 14.6 opencl files in sgminer folder. Msi 7790s up from 1.17mh to 1.62mh. Sapphire 270s up from 2mh to 2.72mh. If anyone is interested in my configs I'll post them upon request.

I use a 7790 (seven-seven-nine-zero, a bit weak I know, but only ~70W).
With 14.6 it went from 1.2MH/s to 1.8MH/s but with locked memory clock (core clock not locked).
I removed 14.6, kept the .bin files, reinstalled 14.4 and everything is fine.
If I recreate the .bin with 14.4 I get my old low hashrates again...

Yes these little things are great power savers, and they run so cool...wish I had bought more last December when they were .1btc each. I'm very curious to know whats in your 1.8mh config...

I have a gigabyte 7790 with 1Gb GDDR5. Core @1150 Mem @1200 (for less heat)
I'm getting 1.830 MH/s and 0.027 WU/m (currently at fractalcoin)
This is my conf. It's nothing special...

Code:

]
,
"xintensity" : "100",
"vectors" : "1",
"worksize" : "128",
"kernel" : "x11mod",
"thread-concurrency" : "8000",
"lookup-gap" : "2",
"gpu-threads" : "4",
"gpu-engine" : "1150",
"gpu-memclock" : "1200",
"queue" : "2",
"scan-time" : "4",
"expiry" : "8",
"no-pool-disable" : true,
"failover-only" : true
}

TC doesn't play such important role as in scrypt. I also use 8192 or 2048, without any significant change.
Memory clock also doesn't affect my rate much, so I keep it low to drop the temps.
You can also try more threads (6 or 8 ).
Q/S/E  based mostly on the coin and pool and my experience/hope/illusion/paranoia...

Update: After a full day of mining, the memory clock is again stuck at stock speed (1500).
I restarted the miner but its still stuck. I'm using djm34 version.
I switched to kozmos version (with bins from 14.6), and the clock is again working ok.
I think this is getting too confusing for me to follow...

From what I've read so far, it seems series 5xxx & 6xxx don't benefit as much by 14.6 or the new miners.

Update2:
If the 14.6 opencl files are in sgminer folder, when I run sgminer, mem clock gets stuck @ stock freq.
If I remove them (after I make the bins), then mem clock acts normal. Weird or not, it works.
I've tried this with both djm34 & kozmos versions for both x11 and x13...

[MacHead2006]

When I try to ./configure new miner in PIMP (14.6) I got: no OpenCL found at the end of ./configure
I have downloaded 14.6 drivers and updated pimp 1.5.5.-14.4, but all was ok with prev. version of djm34 (nist5 mod) miner.

Any ideas ?
maybe copy any files from somewhere ? I see some copy win files but dont know what to do in Linux. 

[Kablam]

When I try to ./configure new miner in PIMP (14.6) I got: no OpenCL found at the end of ./configure
I have downloaded 14.6 drivers and updated pimp 1.5.5.-14.4, but all was ok with prev. version of djm34 (nist5 mod) miner.

Any ideas ?
maybe copy any files from somewhere ? I see some copy win files but dont know what to do in Linux.  

i think you have the wrong opencl verison or adl sdk after updating your drivers you have to manully delete and reinstall that.

no need to reinvent the wheel...

go to #pimp on irc freenode and ask for the 14.6b image of pimp and you dont have to do it yourself there is a workign image allrdy.

Kablam

[MacHead2006]

http://files.getpimp.org/pimp-testing/

that was a link.

[aadje93]

Does anybody have a fix for zlib1.dll error. Can't find a solution on google to fix it. Dont know from which package it is

EDIT:

Found it in sgminer_x11mod version

thanks man, hashing @ 3,9 now instead of 3,1 on 7950's. Hope this is also more stable, x11_mod had alot of "sgminer stopped working"

Imagine, i came from 1,93 on magicOS (magicpool ubuntu mining os) to 3.1 on the x11_mod and now 3,90. It insane!!

[kangjunzi]

@bullus,

can you post your config ?

Code:

],
"kernel" : "x11mod",
"api-allow" : "W:127.0.0.1",
"api-listen" : true,
"api-mcast-port" : "4028",
"api-port" : "4028",
"expiry" : "28",
"failover-only" : true,
"kernel-path" : "/usr/local/bin",
"log" : "5",
"queue" : "1",
"scan-time" : "7",
"scrypt" : true,
"shares" : "0",
"failover-switch-delay" : "10",
"gpu-threads" : "2",
"gpu-dyninterval" : "7",
"gpu-engine" : "0",
"gpu-fan" : "55",
"gpu-platform" : "0",
"gpu-memclock" : "0",
"gpu-memdiff" : "0",
"gpu-powertune" : "8",
"gpu-vddc" : "0",
"intensity" : "18",
"temp-target" : "75",
"temp-overheat" : "85",
"temp-cutoff" : "95",
"temp-hysteresis" : "3",
"vectors" : "1",
"worksize" : "64",
"lookup-gap" : "0",
"shaders" : "1536",
"thread-concurrency" : "8000",
"no-pool-disable" : true,
"device" : "0"
}

6970@950/1375~2300kh/s x11

Great, this is a good step towards even better community.

[Kablam]

http://www.xxxxxxx.xxx/xxxx/

that was a link.

dont publish that link here there was reason for you to go to the irc channel and ask for it!
the 1.5.51-14.6b image isnt public since there is some bugs on it like sgminer 5 not working....

 

[BITCOINDENMARK]

got 4.400 my r9 280x toxic. thank you.

Is that on windows 32 or 64 ?  (Does it mater, using 32/64?)

[bullus]

@bullus,

can you post your config ?

Code:

],
"kernel" : "x11mod",
"api-allow" : "W:127.0.0.1",
"api-listen" : true,
"api-mcast-port" : "4028",
"api-port" : "4028",
"expiry" : "28",
"failover-only" : true,
"kernel-path" : "/usr/local/bin",
"log" : "5",
"queue" : "1",
"scan-time" : "7",
"scrypt" : true,
"shares" : "0",
"failover-switch-delay" : "10",
"gpu-threads" : "2",
"gpu-dyninterval" : "7",
"gpu-engine" : "0",
"gpu-fan" : "55",
"gpu-platform" : "0",
"gpu-memclock" : "0",
"gpu-memdiff" : "0",
"gpu-powertune" : "8",
"gpu-vddc" : "0",
"intensity" : "18",
"temp-target" : "75",
"temp-overheat" : "85",
"temp-cutoff" : "95",
"temp-hysteresis" : "3",
"vectors" : "1",
"worksize" : "64",
"lookup-gap" : "0",
"shaders" : "1536",
"thread-concurrency" : "8000",
"no-pool-disable" : true,
"device" : "0"
}

6970@950/1375~2300kh/s x11

how many cards do you have in the rig ? are all 6970 ? what cpu / ram have you ?

p.s. thnx for the reply

This rig is only 1x 6970 (gamepc)

intel cpu 2600K
8gb mem
Asus Maximus IV
win8.1

Other rig with 4x7950 + 1x6970 is hashing the same for the 6970.

G3220 intel CPU
8GB Ram
Asrock BTC H81
win8.1

[DedRes]

Hi, guys.
Hi
I'm a newbie in mining and I'd highly appreciate some help with the sgminer on win7
For some reason I'm unable to get connected to any pool (I tried several).
Drivers are 14.6

And here's how it runs:

D:\ISR\sgminer_14.6_fixed\kernel>setx GPU_MAX_ALLOC_PERCENT 100
SUCCESS: Specified value was saved.
D:\ISR\sgminer_14.6_fixed\kernel>setx GPU_USE_SYNC_OBJECTS 1
SUCCESS: Specified value was saved.
D:\ISR\sgminer_14.6_fixed\kernel>
D:\ISR\sgminer_14.6_fixed\kernel>..\sgminer.exe --kernel x11mod -g 1 -w 128 -I 18 -o stratum+tcp://west01.us.trademybit.com:4440 -u xxxxxx.1 -p x --no-submit-stale --verbose

--text-only
[13:54:16] Started sgminer 4.1.0
[13:54:17] CL Platform 0 vendor: Advanced Micro Devices, Inc.
[13:54:17] CL Platform 0 name: AMD Accelerated Parallel Processing
[13:54:17] CL Platform 0 version: OpenCL 1.2 AMD-APP (1526.3)
[13:54:17] Platform 0 devices: 1
[13:54:17]      0       Caicos
[13:54:17] Number of ADL devices: 1
[13:54:18] ATI ADL Overdrive5 API found.
[13:54:18] ATI ADL Overdrive6 API found.
[13:54:18] Found 4 ADL adapters
[13:54:18] ADL index 0, id 321458240 - BIOS partno.: BR41330.001, version: 013.0
09.000.010, date: 2011/07/20 06:08
[13:54:18] GPU 0 assigned: iAdapterIndex:0 iPresent:1 strUDID:PCI_VEN_1002&DEV_6760&SUBSYS_161E103C&REV_00_4&2179970E&0&0008A iBusNumber:1 iDeviceNumber:0 iFunctionNumber:0

iVendorID:1002 name:AMD Radeon HD 7400M Series
[13:54:18] ADL index 1, id 321458240 - BIOS partno.: BR41330.001, version: 013.0
09.000.010, date: 2011/07/20 06:08
[13:54:18] ADL index 2, id 321458240 - BIOS partno.: BR41330.001, version: 013.0
09.000.010, date: 2011/07/20 06:08
[13:54:18] ADL index 3, id 321458240 - BIOS partno.: BR41330.001, version: 013.0
09.000.010, date: 2011/07/20 06:08
[13:54:18] GPU 0 AMD Radeon HD 7400M Series hardware monitoring enabled
[13:54:18] ADL GPU 0 is Adapter index 0 and maps to adapter id 321458240
[13:54:18] GPU 0 BIOS partno.: BR41330.001, version: 013.009.000.010, date: 2011/07/20 06:08
[13:54:18] Failed to ADL_Overdrive5_FanSpeed_Get for default value
[13:54:18] Init GPU thread 0 GPU 0 virtual GPU 0
[13:54:18] CL Platform vendor: Advanced Micro Devices, Inc.
[13:54:18] CL Platform name: AMD Accelerated Parallel Processing
[13:54:18] CL Platform version: OpenCL 1.2 AMD-APP (1526.3)
[13:54:18] List of devices:
[13:54:18]      0       Caicos
[13:54:18] Selected 0: Caicos
[13:54:18] Kernel x11mod is experimental.
[13:55:15] Initialising kernel x11mod.cl with bitalign, 1 vectors and worksize 128
[13:55:15] initCl() finished. Found Caicos
[13:55:15] Probing for an alive pool
[13:55:15] Testing Pool 0
[13:55:25] Waiting for work to be available from pools.
[13:55:25] Setting GPUs to idle performance.
[13:55:58] Closing socket for stratum Pool 0
[13:56:15] No servers were found that could be used to get work from.
[13:56:15] Please check the details from the list below of the servers you have input
[13:56:15] Most likely you have input the wrong URL, forgotten to add a port, or  have not set up workers
[13:56:15] Pool: 0  URL: stratum+tcp://west01.us.trademybit.com:4440  User: xxxxxx.1  Password: x
[13:56:15] No servers could be used! Exiting.


Where am I wrong with that?
Thanks in advance

[Hank.H]

i encountered some issues with this 14.4 fixed version.

I downgraded to 14.4 but when i ran it, only 1 GPU work.

like :

GPU0 : OFF
GPU1 : 100%
GPU2 : OFF
GPU3 : OFF

Any ideas ?  

thx

(4x R9 290)

[rustybeamer]

Thanks to all concerned (too many posts to quote).

Took my MSI R9 290 from 3.7 to 4.99Mh/s on x11.

 Used the method of extracting (but not installing) the 14.6 AMD package, copying the 8 required files from the AMD/AMD-Catalyst-14.6-Beta-V1.0-Windows-May23/Packages/Drivers/Display/WB6A-INF/B172153 folder and putting them in the sgminer folder with any previous bins removed before starting up the miner.

Getting WU 0.069 and 0.7% rejects with 1010 core, 1300 mem, intensity 20 on trademybit pool.  Gonna let it run for a few hours before tweaking some more.

Cheers again  

[fra55]

i encountered some issues with this 14.4 fixed version.

I downgraded to 14.4 but when i ran it, only 1 GPU work.

like :

GPU0 : OFF
GPU1 : 100%
GPU2 : OFF
GPU3 : OFF

Any ideas ?  

thx

(4x R9 290)

Check the .conf file, it will easily contain the string GPU: 1, so only the GPU n.1 will be activated.
Change it to 0,1,2,3

[Hank.H]

i encountered some issues with this 14.4 fixed version.

I downgraded to 14.4 but when i ran it, only 1 GPU work.

like :

GPU0 : OFF
GPU1 : 100%
GPU2 : OFF
GPU3 : OFF

Any ideas ?  

thx

(4x R9 290)

Check the .conf file, it will easily contain the string GPU: 1, so only the GPU n.1 will be activated.
Change it to 0,1,2,3

Yes ! it works thank you very much