[ANN] sgminer v5 - optimized X11/X13/NeoScrypt/Lyra2RE/etc. kernel-switch miner - page 196.

jch9678

full member

Activity: 169

Merit: 100

Quote from: platinum4 on July 16, 2014, 12:51:53 AM

Quote from: jch9678 on July 16, 2014, 12:33:02 AM

I'm new to sgminer 5 and x11, x13 etc coins but I've got a 290x mining on an x11 pool hashing around 5.3 mh/s. However I can't adjust the core clock with Afterburner. I can adjust the memory clock but not the core. Has anyone seen this behavior?

Also I cannot get NiceHash's versions of SGMiner 5 and SGMiner 4.1.271 to work with my P2Pool. I noticed others having this problem too. I've added "no-extranonce" : "true", to my config and that doesn't fix it either. Could this functionality have been removed because it wasn't needed for NiceHash's servers? I'm just guessing.

Known driver issue on some of the latest sgminer5 builds.

I can set higher clocks in X15GPU builds and sph_sgminer, some of the latest commits lock the gpu-engine on 14.6+ drivers for some reason.

Gotcha. Thanks. Any idea about my p2pool issue? Also thanks for posting your configs. I'm using yours and badman 74's as a base. Though I'm using 1 GPU thread while you guys seem to prefer 2 threads for x11...

GunPowder

full member

Activity: 168

Merit: 100

Quote from: GunPowder on July 14, 2014, 05:15:20 PM

Quote from: nicehashdev on July 14, 2014, 05:13:52 PM

That can depend on checking interval. What kind of miner do you use?

latest sgminer-5_0 from git. Should I use NH-version?

So, may I have an answer?

platinum4

sr. member

Activity: 547

Merit: 250

Quote from: jch9678 on July 16, 2014, 12:33:02 AM

I'm new to sgminer 5 and x11, x13 etc coins but I've got a 290x mining on an x11 pool hashing around 5.3 mh/s. However I can't adjust the core clock with Afterburner. I can adjust the memory clock but not the core. Has anyone seen this behavior?

Also I cannot get NiceHash's versions of SGMiner 5 and SGMiner 4.1.271 to work with my P2Pool. I noticed others having this problem too. I've added "no-extranonce" : "true", to my config and that doesn't fix it either. Could this functionality have been removed because it wasn't needed for NiceHash's servers? I'm just guessing.

Known driver issue on some of the latest sgminer5 builds.

I can set higher clocks in X15GPU builds and sph_sgminer, some of the latest commits lock the gpu-engine on 14.6+ drivers for some reason.

jch9678

full member

Activity: 169

Merit: 100

I'm new to sgminer 5 and x11, x13 etc coins but I've got a 290x mining on an x11 pool hashing around 5.3 mh/s. However I can't adjust the core clock with Afterburner. I can adjust the memory clock but not the core. Has anyone seen this behavior?

Also I cannot get NiceHash's versions of SGMiner 5 and SGMiner 4.1.271 to work with my P2Pool. I noticed others having this problem too. I've added "no-extranonce" : "true", to my config and that doesn't fix it either. Could this functionality have been removed because it wasn't needed for NiceHash's servers? I'm just guessing.

BombaUcigasa

legendary

Activity: 1442

Merit: 1005

Quote from: badman74 on July 15, 2014, 05:56:21 PM

in my experience when you have a miner that runs fine for a while then suddenly goes sick the clocks are set just a tad too high

This is solid advice. Even 10Mhz down will reduce the hangup events.

opet

newbie

Activity: 57

Merit: 0

Quote from: badman74 on July 15, 2014, 09:34:01 PM

edit: silly question but you are setting your username right?
edit2: err why do you have "failover-only": true, twice and
"extranonce-subscription": true,
"no-extranonce": true,
maybe that is why it doesn't work...
you tell the pool to use extranonce subscription then turn it off?

Alright, I removed both extranonce lines and the second failover-true line. (they were left in from a copy/paste I did of someone else's working conf). I then found an error in a username entry...

GOT IT WORKING! THANK YOU!! Grin

badman74

hero member

Activity: 658

Merit: 500

Quote from: opet on July 15, 2014, 09:25:23 PM

Quote from: platinum4 on July 15, 2014, 09:19:44 PM

Those ports may be closed if nist5 is the most profitable, which it has been for the last day or two.

All you have are profitability ports, with no fixed port as a failover.

How do I add a fixed port as failover? What would I need to add to the above?

EDIT:
I just added this pool to the list:

Code:

{
"name" : "nist5 failover",
"url" : "stratum+tcp://east01.us.trademybit.com:7770",
"user" : "xxx",
"pass" : "x",
"profile" : "nist5"
}

Still get the same URL error for all five. Sad

just add a pool on to the bottom of your pool list that is not using a profit switching port

Code:

{
"name" : "x11 Failover",
"url" : "stratum+tcp://east01.us.trademybit.com:4440",
"user" : "xxx",
"pass" : "x",
"profile" : "x11"
}

edit: silly question but you are setting your username right?
edit2: err why do you have "failover-only": true, twice and
"extranonce-subscription": true,
"no-extranonce": true,
maybe that is why it doesn't work...
you tell the pool to use extranonce subscription then turn it off?

opet

newbie

Activity: 57

Merit: 0

Quote from: platinum4 on July 15, 2014, 09:19:44 PM

Those ports may be closed if nist5 is the most profitable, which it has been for the last day or two.

All you have are profitability ports, with no fixed port as a failover.

How do I add a fixed port as failover? What would I need to add to the above?

EDIT:
I just added this pool to the list:

Code:

{
"name" : "nist5 failover",
"url" : "stratum+tcp://east01.us.trademybit.com:7770",
"user" : "xxx",
"pass" : "x",
"profile" : "nist5"
}

Still get the same URL error for all five. Sad

platinum4

sr. member

Activity: 547

Merit: 250

Those ports may be closed if nist5 is the most profitable, which it has been for the last day or two.

All you have are profitability ports, with no fixed port as a failover.

opet

newbie

Activity: 57

Merit: 0

Any ideas what might cause this URL error when launching sgminer5?

Code:

[21:07:35] Please check the details from the list below of the servers you have
input
[21:07:35] Most likely you have input the wrong URL, forgotten to add a port, or
 have not set up workers
[21:07:35] Pool: 0  URL: stratum+tcp://east01.us.trademybit.com:4010  User: xxx  Password: x
[21:07:35] Pool: 1  URL: stratum+tcp://east01.us.trademybit.com:4011  User: xxx  Password: x
[21:07:35] Pool: 2  URL: stratum+tcp://east01.us.trademybit.com:4012  User: xxx  Password: x
[21:07:35] Pool: 3  URL: stratum+tcp://east01.us.trademybit.com:4013  User: xxx  Password: x
[21:07:35] Press any key to exit, or sgminer will try again in 15s.

My sgminer.conf follows:

Code:

{
"pools" : [
{
"name" : "x11 multi",
"url" : "stratum+tcp://east01.us.trademybit.com:4010",
"user" : "xxx",
"pass" : "x",
"profile" : "x11"
}
,
{
"name" : "x13 multi",
"url" : "stratum+tcp://east01.us.trademybit.com:4011",
"user" : "xxx",
"pass" : "x",
"profile" : "x13"
}
,
{
"name" : "x15 multi",
"url" : "stratum+tcp://east01.us.trademybit.com:4012",
"user" : "xxx",
"pass" : "x",
"profile" : "x15"
}
,
{
"name" : "nist5 multi",
"url" : "stratum+tcp://east01.us.trademybit.com:4013",
"user" : "xxx",
"pass" : "x",
"profile" : "nist5"
}
]
,
"profiles" : [
"name" : "x11",
"algorithm" : "darkcoin-mod",
"intensity" : "21",
"gpu-threads" : "1",
"thread-concurrency" : "24000",
"worksize": "256"
}
,
{
"name" : "x13",
"algorithm" : "marucoin-mod",
"intensity" : "18",
"gpu-threads" : "4",
"thread-concurrency" : "24000",
"worksize": "256"
}
,
{
"name" : "x15",
"algorithm" : "bitblock",
"intensity" : "18",
"gpu-threads" : "2",
"thread-concurrency" : "8192",
"worksize": "256"
}
,
{
"name" : "nist5",
"algorithm" : "talkcoin-mod",
"intensity" : "20",
"gpu-threads" : "4",
"thread-concurrency" : "8192",
"worksize": "256"
}
]
,
"look-up gap": 2,
"gpu-powertune": "20",
"gpu-engine": "1050",
"gpu-memclock": "1570",
"failover-only": true,
"auto-fan": true,
"gpu-fan": "50-85",
"temp-cutoff" : "95",
"temp-overheat" : "85",
"temp-target" : "75",
"api-listen": true,
"api-port": "4028",
"api-mcast-port": "4028",
"api-allow": "W:127.0.0.1",
"queue": "0",
"expiry": "1",
"show-coindiff": true,
"no-submit-stale": true,
"failover-only": true,
"failover-switch-delay": "30",
"extranonce-subscription": true,
"no-extranonce": true,
"hotplug": "5",
"no-restart": true,
"remove-disabled": true
}

I'm new to this multi-agro version, so it's probably just something stupid that I missed. Any help would be much appreciated!

mannyg

full member

Activity: 142

Merit: 101

Ive spent weeks fine tuning and finding the happy engine clock for the GPUs, heres what I'm running without any random SICK or DEAD ones popping up.

All GPUs are MSI Gaming. gpu-engine/gpu-memclock

280x:
X11 -- 1130/1500
X13 -- 1120/1500
X15 -- 1130/1500
NIST5 -- 1125/1500

290:
X11 --1040/1250
X13 -- 1030/1250
X15 -- 1020/1250
NIST5 -- 1050/1250

The 290 is notoriously tough to overclock without it just totally crapping out the driver. The above settings have maintained a miner stable for well over a period of days on the same algorithm.

badman74

hero member

Activity: 658

Merit: 500

Bajawah

newbie

Activity: 59

Merit: 0

badman74

hero member

Activity: 658

Merit: 500

Bajawah

newbie

Activity: 59

Merit: 0

Quote from: badman74 on July 15, 2014, 12:50:34 PM

Quote from: Bajawah on July 15, 2014, 12:33:07 PM

Has anyone found a solution for the full machine lockup?

I searched through this thread and found a few people also running into this, but without resolution.

I'm assuming it has to do with a GPU going sick and bringing down the whole shabang.

My mine is headless, so it's hard to tell what happened other than after about 24 hours it sometimes becomes unresponsive and requires a power cycle. :-/

hard to say without knowing more about your setup but generally it has to do with setting TC or clocks too high

Quote from: mannyg on July 15, 2014, 02:09:47 PM

I cant stress enough how important it is to curb your GPU clocks to keep the switching working right. If you are pushing one algo too hard and a switch comes along, chances are the card will go SICK and DEAD after the driver crashes.

Is there a log file that I can look through to see what is causing an issue?

I added a restart crontab as a hacky fix for now. Unsure if it will work.

On Asus DCII 280x x 6 :

Code:

{
    "pools": [
        {
            "name": "TradeMyBit_X11",
            "url": "stratum+tcp://west01.us.trademybit.com:4010",
            "user": "x",
            "pass": "x",
            "nfactor": "10",
            "algorithm": "darkcoin-mod",
            "profile": "X-Coins"
        },
        {
            "name": "TradeMyBit_X13",
            "url": "stratum+tcp://west01.us.trademybit.com:4011",
            "user": "x",
            "pass": "x",
            "nfactor": "10",
            "algorithm": "marucoin-mod",
            "profile": "X-Coins"
        },
        {
            "name": "TradeMyBit_X15",
            "url": "stratum+tcp://west01.us.trademybit.com:4012",
            "user": "x",
            "pass": "x",
            "algorithm": "bitblock",
            "profile": "X-Coins"
        },
        {
            "name": "TradeMyBit_Nist5",
            "url": "stratum+tcp://west01.us.trademybit.com:4013",
            "user": "x",
            "pass": "x",
            "algorithm": "talkcoin-mod",
            "profile": "Nist5"
        },
        {
            "name": "TradeMyBit_Scrypt-N",
            "url": "stratum+tcp://west01.us.trademybit.com:2220",
            "user": "x",
            "pass": "x",
            "profile": "Scrypt-N"
        },
        {
            "name": "TradeMyBit_Scrypt",
            "url": "stratum+tcp://west01.us.trademybit.com:3330",
            "user": "x",
            "pass": "x",
            "nfactor": "10",
            "algorithm": "zuikkis"
        }
    ],
    "profiles": [
        {
            "name": "X-Coins",
            "intensity": "16",
            "gpu-engine": "1100",
            "gpu-memclock": "1800"
        },
        {
            "name": "Scrypt-N",
            "intensity": "13",
            "gpu-threads": "1",
            "gpu-engine": "1000",
            "gpu-memclock": "1500",
            "thread-concurrency": "10240"
        },
        {
            "name": "Nist5",
            "worksize": "64",
            "intensity": "18",
            "hamsi-expand-big": "1",
            "gpu-powertune": "8"
        }
    ],
    "hamsi-expand-big": "4",
    "worksize": "256",
    "thread-concurrency": "8192",

    "gpu-threads": "2",
    "gpu-powertune": "20",
    "gpu-engine": "1070",
    "gpu-memclock": "1600",

    "auto-fan": true,
    "gpu-fan": "50-85",
    "temp-target": "75",
    "temp-cutoff": "95",
    "temp-overheat": "85",

    "api-listen": true,
    "api-port": "4028",
    "api-mcast-port": "4028",
    "api-allow": "W:127.0.0.1",

    "queue": "0",
    "expiry": "1",
    "show-coindiff": true,
    "no-submit-stale": true,
    "failover-only": true,
    "failover-switch-delay": "30",
    "extranonce-subscription": true,
    "no-extranonce": true,
    "hotplug": "5",
    "no-restart": true,
    "remove-disabled": true
}

badman74

hero member

Activity: 658

Merit: 500

Quote from: atomicchaos on July 15, 2014, 03:37:00 PM

Quote from: badman74 on July 15, 2014, 03:00:30 PM

Quote from: risingtide on July 15, 2014, 02:36:14 PM

I'm not getting any GPUs recognized (Win7 x64 w/ 14.6 drivers) as mentioned here.

Code:

clDevicesNum returned error, no GPUs usable

Wondering if anyone else had encountered this problem?

have you tried --gpu-platform 1 in the .bat file or "gpu-platform" : "1", in your .conf

This will solve it. See? I knew someone else had the same issue that I did. This just appeared in SGMiner 5.

Did they change the default? I believe it used to default to "1". If the above doesn't work, do --gpu-platform 0

On my 50 rigs, I had to have a mixture.. Some worked by default, others had an error unless I specified. I submitted a bug report in Github, but I guess it was thought that this would fix the "new" issue, which it does, but it doesn't explain why the behavior changed between 4.1 and 5.

it has something to do with sgminer 5 not working with the onboard video (gpu-platform 0 if it there)

atomicchaos

hero member

Activity: 630

Merit: 500

Quote from: badman74 on July 15, 2014, 03:00:30 PM

Quote from: risingtide on July 15, 2014, 02:36:14 PM

I'm not getting any GPUs recognized (Win7 x64 w/ 14.6 drivers) as mentioned here.

Code:

clDevicesNum returned error, no GPUs usable

Wondering if anyone else had encountered this problem?

have you tried --gpu-platform 1 in the .bat file or "gpu-platform" : "1", in your .conf

This will solve it. See? I knew someone else had the same issue that I did. This just appeared in SGMiner 5.

Did they change the default? I believe it used to default to "1". If the above doesn't work, do --gpu-platform 0

On my 50 rigs, I had to have a mixture.. Some worked by default, others had an error unless I specified. I submitted a bug report in Github, but I guess it was thought that this would fix the "new" issue, which it does, but it doesn't explain why the behavior changed between 4.1 and 5.

badman74

hero member

Activity: 658

Merit: 500

Quote from: risingtide on July 15, 2014, 02:36:14 PM

I'm not getting any GPUs recognized (Win7 x64 w/ 14.6 drivers) as mentioned here.

Code:

clDevicesNum returned error, no GPUs usable

Wondering if anyone else had encountered this problem?

have you tried --gpu-platform 1 in the .bat file or "gpu-platform" : "1", in your .conf

mannyg

full member

Activity: 142

Merit: 101

Quote from: risingtide on July 15, 2014, 02:36:14 PM

I'm not getting any GPUs recognized (Win7 x64 w/ 14.6 drivers) as mentioned here.

Code:

clDevicesNum returned error, no GPUs usable

Wondering if anyone else had encountered this problem?

Double check for your on-board video and disable it if it is enabled.

risingtide

member

Activity: 143

Merit: 10

I'm not getting any GPUs recognized (Win7 x64 w/ 14.6 drivers) as mentioned here.

Code:

clDevicesNum returned error, no GPUs usable

Wondering if anyone else had encountered this problem?

Topic: [ANN] sgminer v5 - optimized X11/X13/NeoScrypt/Lyra2RE/etc. kernel-switch miner - page 196. (Read 877844 times)