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

[lmaonade80]

I don't see any issues and your connection issue is exactly the same as mine. It has to be on Nicehash's end. I have gotten 3 out of my 18 rigs to connect to nicehash through sheer luck, and the rest are hashing at LTCrabbit x11 port as the failover. I guess I just have to wait it out and hope they connect?

[lmaonade80]

For all of you who are using NiceHash: today we had another DDoS (purely network related), I'm sorry for that. Nothing was lost, but as you've experienced there has been lots of disconnects both from providers rigs an buyers orders. Some miners might still having issues by not being able to connect due to several disconnects and invalid shares resulting into unwanted bans - this will be sorted by the end of the day by automatic un-ban (for valid miners). We'll improve our DDoS protection by moving to another, better DDoS protected location. Service is now working normally and all download links are alive. Thanks for understanding and thanks for using NiceHash.

I am thinking that they did not unban quite a few IP's that are legitimate but confused with aiding with the DDOS attack. ie, I have 3 rigs in my apt that can connect, but the 16 at my farm cannot do anything with NiceHash. I'm getting a little frustrated.

[MrBobC]

I'm still at 15 at Nicehash and 8 at Wafflepool.  When did this start for you?  For me the connection issues happened immediately after my miner upgrade, and interestingly enough, when I go back to the old miner, it still doesn't connect.  For me this was when X15 became active on betarigs, I think Saturday the 28th.  I guess I can't understand the issue because I'm using the nicehash build of sgminer which they sponsored to make work with their website.  I will say there was one day one time when I restarted the miners and they connected instantly to nicehash, but that was only one time.

[mannyg]

I've been noticing lately that Nicehash, when connected, stays hashing fine.... that is until you restart the miner or something happens where your rig reboots, then it wont connect for a looooong while.  I had to set my CGWatcher to keep trying indefinitely else it would just sit there after 3 failed attempts to start.

Has anyone managed to get their 280x's or 290's X15 speed to hash at 3.5x times their Scrypt hash rate?  I cant get my 280x's barely at 2.3MH/s and my 290s at 2.4MH/s.  I *should* be closer to 2.5MH/s and 3MH/s. 

I'm using 13.12 drivers and Win8 x64, bitblock.cl kernel.  Using the same engine, memclock and intensity as X11 & X13

[bullus]

U can try my bitblock.bin I'm getting 2450kh/s with a 7950@1035/1500

https://mega.co.nz/#!zM8FXJwZ!E4u3wCEKR0y0BeXMwvEIDPGrYdgLcuVw5Zep13w4rNw

config
https://bitcointalk.org/index.php?topic=386554.msg7646091#msg7646091

[MrBobC]

I agree Mannyg, once they connect they're fine, no issues, but if you restart... Also, I have my 280x hashing around 2.6Mh on 14.6 beta drivers in BAMT.  I use the same settings for x11/x13/x15, the below conf is for my Sapphire Vapor-X 280x.

"intensity" : "16",
"worksize" : "256",
"thread-concurrency" : "8192",
"gpu-threads" : "2",
"gpu-engine" : "1080",
"gpu-memclock" : "1450",
"failover-only" : true,
"no-pool-disable" : true,
"api-network" : true,
"api-listen" : true,
"api-port" : "4028",
"gpu-powertune" : "0"

Sure wish Nicehash would chime in and let us know what is up with these connection issues!

[oranges411]

my 7850 isn't working with any bitblock kernel
Commit from 7 hours ago: 27709ced8364b164c6342a8f38ffdedd9aa91d90
7850 with 1GB of GPU memory, xubuntu 13.10, catalyst driver 14.6

In debug, after if receives work from a pool, it keeps saying the number of staged work items is greater than the maximum (which is set at whatever queue is), and then discards the work item.

Code:

Selecting BITBLOCKPOOL for work
Generated stratum merkle [.......]
Generated stratum header [........]
Work job_id [....] nonce2 [..] ntime [......]
Generated target [.......]
Generated stratum work
Pushing work from BITBLOCKPOOL to hash queue
35.0 C  F: 20%(1203RPM)  E: 860MHz  M: 1200Mhz  V: 1.138V  A: 99%  P: 0%
32.0 C  F: 20%(1132RPM)  E: 860MHz  M: 1200Mhz  V: 1.138V  A: 99%  P: 0%
34.0 C  F: 20%(1047RPM)  E: 860MHz  M: 1200Mhz  V: 1.138V  A: 99%  P: 0%   
Staged work: total (2) > max (1), discarding
Discarded work 

^^^ repeated again and again

If I change the queue number, the number of staged work items is still always one more than the "max" and the work get discarded.  I've tried tons of different settings, and if I don't have it in text mode, the hashrates all stay at 0 and after a few minutes the cards are declared sick and hung.

advice?

[MrBobC]

I'm noticing another interesting issue.  So, based on the conf I provided earlier, you can see I rent on betarigs.  I am noticing that when I rent out my rigs, the only ones that click over to betarigs are the ones that did eventually connect to nicehash.  All the other ones just stay on wafflepool and don't hash.  This is completely unacceptable, because it means when I rent only 1/3 of the rigs have been kicking over, and the rest of them essentially go dead.  I never had this problem with the old software.  Anyone else?

[guzzzi]

How to activate the Logging into a File? Sometimes SGMiner just close without giving an Information why. Would like to Log everything to see what happen.

[Trillium]

I agree Mannyg, once they connect they're fine, no issues, but if you restart... Also, I have my 280x hashing around 2.6Mh on 14.6 beta drivers in BAMT.  I use the same settings for x11/x13/x15, the below conf is for my Sapphire Vapor-X 280x.

"intensity" : "16",
"worksize" : "256",
"thread-concurrency" : "8192",
"gpu-threads" : "2",
"gpu-engine" : "1080",
"gpu-memclock" : "1450",
"failover-only" : true,
"no-pool-disable" : true,
"api-network" : true,
"api-listen" : true,
"api-port" : "4028",
"gpu-powertune" : "0"

Sure wish Nicehash would chime in and let us know what is up with these connection issues!

Seems to vary a lot with driver version, my 280x only getting about 2.1 MH/s with pretty much identical settings. I'm on 13.12 still, not sure I can be bothered to play uninstall/reinstall/reconfigure-all-the-displays games right now though..

[JuanHungLo]

How to activate the Logging into a File? Sometimes SGMiner just close without giving an Information why. Would like to Log everything to see what happen.

The way I log activity is to start my miners with a batch file:

Code:

setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
del *.bin
sgminer 2>logfile.txt --config nicehashx15.conf 

the "sgminer 2>logfile.txt" will write the activity to the same directory as the sgminer.

but lately my miners have been shutting down or freezing or rebooting without a clue left in the logfile.

Something's not right with the x15 config.  I just can't figure it out.

[guzzzi]

Yes right thats my Problem too. Seems X15 shut down SGminer after a while.

Many thanks für the Hint with the Log.

[unitedminers]

How do I get the column with "A:" below (S)ettings (D)isplay in the compiled by myself or compiled by Nicehash SGMiner 5 back?

I get lots of "Accepted ..... Diff ... GPU2 ... at ...." but on top is everything at 0 "A:0 R:0 HW:0", too.

SPH-SGminer shows the column "A:" but doesn't work with X15.

Compiled SGMiner 5 (Ubuntu 13.10) with this howto: http://blog.shindasingh.com/2014/01/sgminer-4-0/comment-page-1/

Thanks for every hint!

[mfqrs3]

Has someone compared power consumption between X11,X13 and X15 ? I find that X15 is more voltage hungry, stable clocks from X13 just don't work here stable enough.

[mannyg]

I agree Mannyg, once they connect they're fine, no issues, but if you restart... Also, I have my 280x hashing around 2.6Mh on 14.6 beta drivers in BAMT.  I use the same settings for x11/x13/x15, the below conf is for my Sapphire Vapor-X 280x.

"intensity" : "16",
"worksize" : "256",
"thread-concurrency" : "8192",
"gpu-threads" : "2",
"gpu-engine" : "1080",
"gpu-memclock" : "1450",
"failover-only" : true,
"no-pool-disable" : true,
"api-network" : true,
"api-listen" : true,
"api-port" : "4028",
"gpu-powertune" : "0"

Sure wish Nicehash would chime in and let us know what is up with these connection issues!

Seems to vary a lot with driver version, my 280x only getting about 2.1 MH/s with pretty much identical settings. I'm on 13.12 still, not sure I can be bothered to play uninstall/reinstall/reconfigure-all-the-displays games right now though..

I've read that the 14.4 drivers increase X11 hash rate, and the 14.6 ones break everything without some voodoo file swapping and modding, etc.  I dont want to have to bother with all that nonsense, it's very frustrating to not see a straight forward driver install for my GPU's and have everything work the same as with 13.12.   

I've started to notice a new thing too, ... occasional random HW errors popping up on X15.   Same configs i used with X11 and X13 that didnt have HW errors at alll... ever. 

Does thread-concurrency matter for X11/X13/X15?

[bullus]

Has someone compared power consumption between X11,X13 and X15 ? I find that X15 is more voltage hungry, stable clocks from X13 just don't work here stable enough.

https://bitcointalk.org/index.php?topic=669806.msg7582139#msg7582139

[mannyg]

May have just solved the random X15 HW errors by changing 2 things just for X15,... worksize and thread-concurrency.  I made the worksize for the 280x's from 256 to 64 and for the 290's from 512 to 128.  Also my tc's were 8193 and 20481, i changed them to 8192 and 20480.  Keeping my fingers crossed and hope it gets rid of them.

[mannyg]

I'm new to nicehash and config files (I usually play with bat files). How exactly do I run the config file in sgminer? I have spend 30 minute on youtube and google for the answer but everytime I run the sgminer it just close down.

I using the sgminer from https://www.nicehash.com/software/sgminer-5.0-pre-release-2014-06-30-win32.zip

Is there a 7970 specific config file? Could someone check if I configure the it correctly?

Code:

{
"pools" : [
        {
                "name" : "NiceHash_Scrypt",
                "url" : "stratum+tcp://stratum.nicehash.com:3333",
                "user" : "b",
                "pass" : "x",
                "algorithm" : "zuikkis",
                "nfactor" : "10",
                "xintensity" : "4",
                "thread-concurrency" : "8192",
                "gpu-threads" : "2"
        },
        {
                "name" : "NiceHash_Scrypt-N",
                "url" : "stratum+tcp://stratum.nicehash.com:3335",
                "user" : "b",
                "pass" : "x",
                "algorithm" : "zuikkis",
                "nfactor" : "11",
                "xintensity" : "4",
                "thread-concurrency" : "8192",
                "gpu-threads" : "2"
        },
        {
                "name" : "NiceHash_X11",
                "url" : "stratum+tcp://stratum.nicehash.com:3336",
                "user" : "b",
                "pass" : "x",
                "algorithm" : "darkcoin-mod",
                "nfactor" : "10",
                "xintensity" : "300",
                "thread-concurrency" : "8960",
                "gpu-threads" : "2"
        },
        {
                "name" : "NiceHash_X13",
                "url" : "stratum+tcp://stratum.nicehash.com:3337",
                "user" : "b",
                "pass" : "x",
                "algorithm" : "marucoin-mod",
                "nfactor" : "10",
                "xintensity" : "300",
                "thread-concurrency" : "8192",
                "gpu-threads" : "2"
        },
        {
                "name" : "NiceHash_Keccak",
                "url" : "stratum+tcp://stratum.nicehash.com:3338",
                "user" : "b",
                "pass" : "x",
                "algorithm" : "maxcoin",
                "nfactor" : "10",
                "xintensity" : "300",
                "thread-concurrency" : "8960",
                "gpu-threads" : "1"
        },
        {
                "name" : "NiceHash Scrypt_Backup",
                "url" : "stratum+tcp://stratum.nicehash.com:3333",
                "user" : "b",
                "pass" : "x",
                "algorithm" : "zuikkis",
                "nfactor" : "10",
                "xintensity" : "4",
                "gpu-threads" : "2",
                "thread-concurrency" : "8192",
        },
        {
                "name" : "Clevermining",
                "url" : "stratum+tcp://us.clevermining.com:3333",
                "user" : "b",
                "pass" : "x",
                "algorithm" : "zuikkis",
                "nfactor" : "10",
                "xintensity" : "4",
                "gpu-threads" : "2",
                "thread-concurrency" : "8192"
        }
],
"xintensity" : "4",
"vectors" : "1",
"algorithm" : "zuikkis",
"lookup-gap" : "2",
"shaders" : "1792",
"thread-concurrency" : "8192",
"device" : "0"
"log" : "5",
"log-dateformat" : "1",
"failover-only" : true,
"failover-switch-delay" : "30",
"no-pool-disable" : true,
"scrypt" : true,
"queue" : "0",
"scan-time" : "7",
"expiry" : "28",
"api-listen" : true,
"api-mcast-port" : "4028",
"api-port" : "4001",
"api-allow" : "W:127.0.0.1"
}

I'm using one of two of my GPU to mine, device = 0

A specific config file for 7970 would be the same as a config file for 280x.  To launch the config file, you would have to add "--config NameOfYourConfigFile.conf" after sgminer at a command prompt.  either '-c' or '--config' will work.

Your config has port numbers that start with 3xxx for each algorithm.  This wont work for the auto switching, they should be 4xxx. 

Also, gpu-thread entries for each algorithm may cause issues if you are on windows

I launch mine from a bat file with contents that look like this:

Code:

setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1

sgminer -c nicehash-multialgo.conf
pause

[xpalax]

Anyone having the issue or know what causes it.  Every 30 seconds sgminer will switch to the next pool in line.

Heres my config:

Code:

{
"pools" : [
{
		"name" : "NiceHash Scrypt",
		"url" : "stratum+tcp://stratum.nicehash.com:4333",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "0",
		"algorithm" : "zuikkis",
		"nfactor" : "10",
		"xintensity" : "4",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "8192",
		"state" : "disabled"	},
{
		"name" : "NiceHash Scrypt-N",
		"url" : "stratum+tcp://stratum.nicehash.com:4335",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "1",
		"algorithm" : "zuikkis",
		"nfactor" : "11",
		"intensity" : "12",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "4096",
		"state" : "disabled"	},
{
		"name" : "NiceHash X11",
		"url" : "stratum+tcp://stratum.nicehash.com:4336",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "2",
		"algorithm" : "darkcoin-mod",
		"nfactor" : "10",
		"xintensity" : "4",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "192",
		"state" : "disabled"	},
{
		"name" : "NiceHash X13",
		"url" : "stratum+tcp://stratum.nicehash.com:4337",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "3",
		"algorithm" : "marucoin-mod",
		"nfactor" : "10",
		"xintensity" : "4",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "8192",
		"state" : "disabled"	},
{
		"name" : "NiceHash X15",
		"url" : "stratum+tcp://stratum.nicehash.com:4339",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "4",
		"algorithm" : "bitblock",
		"nfactor" : "10",
		"xintensity" : "4",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "8192",
		"state" : "disabled"	},
{
		"name" : "NiceHash Keccak",
		"url" : "stratum+tcp://stratum.nicehash.com:4338",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "5",
		"algorithm" : "maxcoin",
		"nfactor" : "10",
		"xintensity" : "300",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "8960",
		"state" : "disabled"	},
{
		"name" : "NiceHash  NIST5",
		"url" : "stratum+tcp://stratum.nicehash.com:4340",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "6",
		"algorithm" : "talkcoin-mod",
		"nfactor" : "10",
		"xintensity" : "4",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "8192",
		"state" : "disabled"	},
{
		"name" : "Wafflepool X13",
		"url" : "stratum+tcp://uswest.wafflepool.com:3330",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "7",
		"algorithm" : "marucoin-mod",
		"nfactor" : "10",
		"xintensity" : "4",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "8192",
		"state" : "enabled"	},
{
		"name" : "Wafflepool X11",
		"url" : "stratum+tcp://uswest.wafflepool.com:3331",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "8",
		"algorithm" : "darkcoin-mod",
		"nfactor" : "10",
		"xintensity" : "4",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "8960",
		"state" : "enabled"	},
{
		"name" : "Wafflepool Scrypt-N",
		"url" : "stratum+tcp://uswest.wafflepool.com:3332",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "9",
		"algorithm" : "zuikkis",
		"nfactor" : "11",
		"intensity" : "12",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "4096",
		"state" : "enabled"	},
{
		"name" : "Wafflepool Scrypt",
		"url" : "stratum+tcp://uswest.wafflepool.com:3333",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "10",
		"algorithm" : "zuikkis",
		"nfactor" : "10",
		"xintensity" : "4",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "8192",
		"state" : "enabled"	},
{
		"name" : "Magicpool",
		"url" : "stratum+tcp://mine.magicpool.org:7777",
		"user" : "x",
		"pass" : "x",
		"pool-priority" : "11",
		"algorithm" : "zuikkis",
		"nfactor" : "10",
		"xintensity" : "4",
		"gpu-engine" : "1000",
		"gpu-memclock" : "1250",
		"thread-concurrency" : "8192",
		"state" : "enabled"	}

],
"api-allow" : "W:127.0.0.1",
"api-listen" : true,
"failover-only" : true,
"queue" : "0",
"rotate" : "0",
"gpu-threads" : "2",
"gpu-powertune" : "20",
"vectors" : "1",
"worksize" : "128",
"shaders" : "1792",
"thread-concurrency" : "8192",
"failover-switch-delay" : "60"
}

[mfqrs3]

Has someone compared power consumption between X11,X13 and X15 ? I find that X15 is more voltage hungry, stable clocks from X13 just don't work here stable enough.

https://bitcointalk.org/index.php?topic=669806.msg7582139#msg7582139

Thanks!