[ANN][X11/X13] X11 (Darkcoin)/X13 (Marucoin) miner (based on sph-sgminer)

[Bawb3]

I have weird behaviour with x13, at first it starts to run around 10.2mh/s
within 4 hours it has linearly dropped to around 9.4mh/s.

today morning after waking up, after 8 hours, it had dropped to 8.7mh/s

and it seems to be linear drop, any suggestions?

I've noticed this behavior on one of my rigs too, it starts out at 13.5 Mh/s, 10 hours later its down to 12.7, temps look fine, and I'm only running single thread configs. It requires the whole system to be restarted in order to bring the hash rates back up, very weird. Using Win 8.1, Catalyst 13.1 (maybe I should try 14.4?)

[sharnalk]

Hi, i'd like a little help.

While running sgminer (the one i get here ) with x11 or x13 mod  i have no Accepted nor Rejected,  only hardware errors.

I tried many option (no option too) but always the same.

I'm running on 280x win7 64   tried catalyst 14.4 & 14.6 (but for most of people 14.4 works pretty well)

I tried to install the 2.9 SDK but still nothing i'm a bit lost i can't figure out the source of this issue

Still no idead about my problem ?

[chadwickx16]

I've noticed this behavior on one of my rigs too, it starts out at 13.5 Mh/s, 10 hours later its down to 12.7, temps look fine, and I'm only running single thread configs. It requires the whole system to be restarted in order to bring the hash rates back up, very weird. Using Win 8.1, Catalyst 13.1 (maybe I should try 14.4?)

Having the same issue on BAMT 1.6.2, with only one R9 290. In total I have 12 7950's, 1 7970, 2 R9 270's, and 4 R9 290's. This only happens on one of them. All of the other cards work perfect.

[deznuts]

To anyone who is not stable / crashes: Your card is running too fast and it crashes because you overclocked too much for your card to handle. Lower your gpu-engine and/or gpu-memclock. I have multiple 7950's and 7970's running X11 (previously) and now X13. 100% stable been running for days. Start with 1030 for core and 1400 for memory and work your way up little by little until you find the optimal speeds. This is not a "one size fits all". You actually need to put some work into it if you want optimal speeds.

[jimlite]

Has anyone got this working with a laptop with mobile 6770m radeon?
X11 works fine, but even x13modold is just hw errors.
Please post config.
Thanks

[hero18688]

[size]Can someone make the optimized sgminer for Jackpotcoin?
https://bitcointalk.org/index.php?topic=584481.0
 Nvidia card has much advantage mining this.That's abnormal.[/size]
Posted from Bitcointa.lk - #45tONu9S6jk2tlWt

[124C41]

Search this thread before you ask for help with settings, configs, etc.

You'll find that a lot of solutions were already posted earlier.

If not, create a post. But please search first.

[nwoolls]

Hi there - just a quick note to let you know the latest preview of MultiMiner 3.2 supports mining with LasyBear's SPH-SGMiner fork, including support for X13 and the modified X11 kernel:

https://bitcointalk.org/index.php?topic=248173.msg7225451#msg7225451

[ebliever]

To anyone who is not stable / crashes: Your card is running too fast and it crashes because you overclocked too much for your card to handle. Lower your gpu-engine and/or gpu-memclock. I have multiple 7950's and 7970's running X11 (previously) and now X13. 100% stable been running for days. Start with 1030 for core and 1400 for memory and work your way up little by little until you find the optimal speeds. This is not a "one size fits all". You actually need to put some work into it if you want optimal speeds.

How certain are you about this being the sole factor? I ran for months on Scrypt with my four 290X's at settings tailored for each one. Switched to X11 and was able to stay at the same speeds. But now going with Lasybears, like other 290/290X users I'm having a continuous struggle with sick GPU's. I've played with engine/memory settings, lowered the temperature settings, reduced intensity (and tried lower xintensity) and so forth, to no avail. It's getting frustrating. Oddly, only the 3 Sapphire 290X's are getting sick. I've never seen the stock reference 290X I have go sick while on this miner.

When the GPU goes sick it cuts the memory speed to 150 Mhz (i.e. about 10% normal), and manually trying to change it doesn't work, only rebooting does. And sometimes I get blue-screens instead. The X11 enhancement of this miner is critical to keeping me involved in mining and leasing my rig, but at the moment I'm dead in the water. Can't really lease my rig out under these conditions unless I badly derate it. The funny thing is that everything seemed to go south a week or so after using this miner, at a time not associated with any identifiable change point. If it was just one GPU I'd chalk it up to a marginal component getting flaky, but with three taking turns going bad, that doesn't seem as likely.

My current GPU settings for Engine/Memory (etc) are as follows. Can anyone with a 290 or 290X report settings they find to be stable?

"intensity" : "18",
"worksize" : "128",
"lookup-gap" : "2",
"thread-concurrency" : "25600",
"gpu-powertune" : "20",
"temp-cutoff" : "95",
"temp-overheat" : "85",
"temp-target" : "80",
"gpu-engine" : "1050,1040,1000,1010",
"gpu-memclock" : "1500,1498,1500,1500",
"auto-fan" : true,
"auto-gpu" : true,
"api-port" : "4028",
"expiry" : "1",
"failover-only" : true,
"gpu-threads" : "1",
"gpu-dyninterval" : "7",
"log" : "5",
"queue" : "1",
"scan-time" : "1",
"temp-hysteresis" : "3",
"kernel" : "x11mod"

[deznuts]

To anyone who is not stable / crashes: Your card is running too fast and it crashes because you overclocked too much for your card to handle. Lower your gpu-engine and/or gpu-memclock. I have multiple 7950's and 7970's running X11 (previously) and now X13. 100% stable been running for days. Start with 1030 for core and 1400 for memory and work your way up little by little until you find the optimal speeds. This is not a "one size fits all". You actually need to put some work into it if you want optimal speeds.

How certain are you about this being the sole factor? I ran for months on Scrypt with my four 290X's at settings tailored for each one. Switched to X11 and was able to stay at the same speeds. But now going with Lasybears, like other 290/290X users I'm having a continuous struggle with sick GPU's. I've played with engine/memory settings, lowered the temperature settings, reduced intensity (and tried lower xintensity) and so forth, to no avail. It's getting frustrating. Oddly, only the 3 Sapphire 290X's are getting sick. I've never seen the stock reference 290X I have go sick while on this miner.

When the GPU goes sick it cuts the memory speed to 150 Mhz (i.e. about 10% normal), and manually trying to change it doesn't work, only rebooting does. And sometimes I get blue-screens instead. The X11 enhancement of this miner is critical to keeping me involved in mining and leasing my rig, but at the moment I'm dead in the water. Can't really lease my rig out under these conditions unless I badly derate it. The funny thing is that everything seemed to go south a week or so after using this miner, at a time not associated with any identifiable change point. If it was just one GPU I'd chalk it up to a marginal component getting flaky, but with three taking turns going bad, that doesn't seem as likely.

My current GPU settings for Engine/Memory (etc) are as follows. Can anyone with a 290 or 290X report settings they find to be stable?

"intensity" : "18",
"worksize" : "128",
"lookup-gap" : "2",
"thread-concurrency" : "25600",
"gpu-powertune" : "20",
"temp-cutoff" : "95",
"temp-overheat" : "85",
"temp-target" : "80",
"gpu-engine" : "1050,1040,1000,1010",
"gpu-memclock" : "1500,1498,1500,1500",
"auto-fan" : true,
"auto-gpu" : true,
"api-port" : "4028",
"expiry" : "1",
"failover-only" : true,
"gpu-threads" : "1",
"gpu-dyninterval" : "7",
"log" : "5",
"queue" : "1",
"scan-time" : "1",
"temp-hysteresis" : "3",
"kernel" : "x11mod"

In 95% of cases the problem is simply your card is clocked too high. I ran into the same issue where some cards were 100% stable on scrypt, but no longer stable on x11 or x13 with certain settings. I had to back down a bit or up the voltage. Below was 100% stable when I ran 4 290x's on X11.

--thread-concurrency 27400 -I 19 -g 1 --worksize 64 --lookup-gap 2 --no-submit-stale --no-adl --kernel x11mod

The cards were all running between 1100-1120 core and 1450 on memory. The cards were all watercooled.

If the memory on your 290x's is Elpidia 99% chance that 1500 is too much which is causing your problem. Try turn it down back to the stock 1250 and see if your problem goes away. If it were me I would just remove gpu-engine and gpu-memclock and see if you are at least stable with stock clocks. If you are then you 100% know it's related to that.

[ebliever]

In 95% of cases the problem is simply your card is clocked too high. I ran into the same issue where some cards were 100% stable on scrypt, but no longer stable on x11 or x13 with certain settings. I had to back down a bit or up the voltage. Below was 100% stable when I ran 4 290x's on X11.

--thread-concurrency 27400 -I 19 -g 1 --worksize 64 --lookup-gap 2 --no-submit-stale --no-adl --kernel x11mod

The cards were all running between 1100-1120 core and 1450 on memory. The cards were all watercooled.

If the memory on your 290x's is Elpidia 99% chance that 1500 is too much which is causing your problem. Try turn it down back to the stock 1250 and see if your problem goes away. If it were me I would just remove gpu-engine and gpu-memclock and see if you are at least stable with stock clocks. If you are then you 100% know it's related to that.

Thanks for the feedback. Odd thing is that GPU 0 was the slow one under Scrypt (919/1500 speed), but it's the only one that tolerates 1050/1500 with no sick issues now. I think it's the one with the Elphidia memory.

I think your suggestion to drop back to stock engine/memory settings is an excellent one. Shouldn't take too many hours running to become confident that that's the problem. I derated my machine and set it to reboot whenever a GPU goes sick and landed a 72 hour lease, but I'll try to report back my results when I get a chance.

[LightningBlade]

www.coinking.io just added x13 multipool few days ago and the profitability is really good. Also have other x11 and other algo..

[bullus]

www.coinking.io just added x13 multipool few days ago and the profitability is really good. Also have other x11 and other algo..

I think every multipool who had cloakcoin(x13) in it had a realy good profitability..now it's over.

[junkun13]

any chance that jackpot alg will be optiomized?

+ 1 on thin

[Amph]

www.coinking.io just added x13 multipool few days ago and the profitability is really good. Also have other x11 and other algo..

the profitability in the front page is wrong, the real one is 0.0005, see under statistics

[yudhistira]

I often get this, whats wrong with http://www.drkpool.com

[xingqiaoyin]

Are you sure it is the pool or maybe you have crappy internet connection.
Or maybe your conn is fine but your router is not up to the job.
If you have too many devices (smartphones,pcs,tablets) connected to a home grade router you will have lots of disconnects like that.

If it is indeed the pool nothing you could do but switch.

[deznuts]

Anyone have x11modCypressglg2tc8000w128l4.bin made with driver 14.1 or 14.4? If so please quote me and post here. Thanks!!!

[Stenull]

Thanks lasybear and co! it runs ok stable, there is some cosmetic bugs when running for some hours but no biggies. Iv'e been running both x11 and x13 for 2-3 days in row without problems.

[timk225]

Some questions, I am getting my mining PCs changed over from the normal X11 for Darkcoin to X11mod for GDN.

I copied this config file from back on page 7 of this topic:

"intensity" : "15",
"worksize" : "256",
"lookup-gap" : "1",
"thread-concurrency" : "25601",
"expiry" : "1",
"auto-fan" : true,
"log" : "5",
"queue" : "0",
"scan-time" : "1",
"gpu-threads" : "10",
"vectors" : "1",
"temp-cutoff" : "95",
"temp-overheat" : "85",
"temp-target" : "75",
"temp-hysteresis" : "3",
"gpu-fan" : "40-100",
"gpu-engine" : "1055",
"gpu-memclock" : "1250",
"gpu-powertune" : "20",
"gpu-dyninterval" : "7",

What is "temp-hysteresis", and what can it do?

What about gpu-dyninterval?

With gpu-fan set for 40-100, that is its percentage of speed, so will it just adjust within that range for the temp target, overheat, and cutoff?

I saw in someone else's coding a line about "gpu 0,1" for 2 gpus, but I forget the exact syntax.  What's the proper line for a computer with two and another with five gpus?  I presume 0,1 and 0,1,2,3,4.