Author

Topic: [ANN] sgminer v5 - optimized X11/X13/NeoScrypt/Lyra2RE/etc. kernel-switch miner - page 219. (Read 877859 times)

sr. member
Activity: 547
Merit: 250
badman, I just noticed your github issue 286 and squirrel-humped it with the fanspeed issue 287.
sr. member
Activity: 547
Merit: 250

I don't see how that would not allow you to either a) manually switch to another alive pool b) crash out the miner in 60 seconds c) force your starting fanspeeds to 99% after a crash out or d) change anything from the previous build from 06.16.2014

This is kind of shitty now if the entire github repo is polluted with this config parsing bullshit if it broke the miner for so many different people.
it switches fine for me but if i am on a profile other than the one i set as default it will time out after 60 sec
edit: the version from nicehash does exactly the same

Great.  For about the last two weeks we were making steady fucking improvements.  Somebody gets the great idea to make a prettier config file and we're back to the first fucking bugs from pages 1-10 of the thread; complaints of ckolivas only, reports of crash outs after 60s, and the same pseudo-overheat on exit

Fuck...
hero member
Activity: 658
Merit: 500
phzi, badman74 likely compiled his own Windows binaries & I am running the Windows binaries from here https://www.nicehash.com/software/sgminer-5.0-pre-release-2014-06-22-win32.zip
It isn't the extra-nonce subscription causing issues maybe is it?  I do have "no-extranonce-subscribe" : true, in all my profiles as well as globally.

I don't see how that would not allow you to either a) manually switch to another alive pool b) crash out the miner in 60 seconds c) force your starting fanspeeds to 99% after a crash out or d) change anything from the previous build from 06.16.2014

This is kind of shitty now if the entire github repo is polluted with this config parsing bullshit if it broke the miner for so many different people.
it switches fine for me but if i am on a profile other than the one i set as default it will time out after 60 sec
edit: the version from nicehash does exactly the same
also if you don't use the profile options it seems to work normally
sr. member
Activity: 547
Merit: 250
phzi, badman74 likely compiled his own Windows binaries & I am running the Windows binaries from here https://www.nicehash.com/software/sgminer-5.0-pre-release-2014-06-22-win32.zip
It isn't the extra-nonce subscription causing issues maybe is it?  I do have "no-extranonce-subscribe" : true, in all my profiles as well as globally.

I don't see how that would not allow you to either a) manually switch to another alive pool b) crash out the miner in 60 seconds c) force your starting fanspeeds to 99% after a crash out or d) change anything from the previous build from 06.16.2014

This is kind of shitty now if the entire github repo is polluted with this config parsing bullshit if it broke the miner for so many different people.
sr. member
Activity: 547
Merit: 250
Fuck this thing it just crashes the drivers repeatedly.
try removing TC on all of them first to see if they will run without
then you can tweak based off the bis it makes

I'd agree, but I can't even force it to try and build a bin by switching to an alive pool...
i ran your config without the TC and it switches for me but times out at 60 sec....


This is a nice idea and all, the profiles and a pretty config file, but if the shit doesn't fucking work... I'm at a loss to speculate how useful it is

I'm sliding back over to that hybrid mis-match drag-and-drop multi you made a few days back badman...
hero member
Activity: 700
Merit: 500
phzi, badman74 likely compiled his own Windows binaries & I am running the Windows binaries from here https://www.nicehash.com/software/sgminer-5.0-pre-release-2014-06-22-win32.zip
It isn't the extra-nonce subscription causing issues maybe is it?  I do have "no-extranonce-subscribe" : true, in all my profiles as well as globally.
sr. member
Activity: 547
Merit: 250
Fuck this thing it just crashes the drivers repeatedly.
try removing TC on all of them first to see if they will run without
then you can tweak based off the bis it makes

I'd agree, but I can't even force it to try and build a bin by switching to an alive pool...

Oh, and there's this bullshit going on with the fucking fanspeeds too.  After you crash out you start up, fans at fucking 99% regardless of temps.
I have neither of these issues on my linux install with 14.6 drivers.  The config I posted above is working great, switching no problems.  If I force quit sgminer and restart it, there's no issue.

I did run into odd problems with previous installs while I was messing with configs tho.  Had to actually remove the entire sgminer folder and build from git again to get it to work (and no, it wasn't old bin files or anything obvious).

phzi, badman74 likely compiled his own Windows binaries & I am running the Windows binaries from here https://www.nicehash.com/software/sgminer-5.0-pre-release-2014-06-22-win32.zip
hero member
Activity: 700
Merit: 500
Fuck this thing it just crashes the drivers repeatedly.
try removing TC on all of them first to see if they will run without
then you can tweak based off the bis it makes

I'd agree, but I can't even force it to try and build a bin by switching to an alive pool...

Oh, and there's this bullshit going on with the fucking fanspeeds too.  After you crash out you start up, fans at fucking 99% regardless of temps.
I have neither of these issues on my linux install with 14.6 drivers.  The config I posted above is working great, switching no problems.  If I force quit sgminer and restart it, there's no issue.

I did run into odd problems with previous installs while I was messing with configs tho.  Had to actually remove the entire sgminer folder and build from git again to get it to work (and no, it wasn't old bin files or anything obvious).

Edit: it isn't the extra-nonce subscription maybe is it?  I do have "no-extranonce-subscribe" : true, in all my profiles as well as globally.
hero member
Activity: 658
Merit: 500
Fuck this thing it just crashes the drivers repeatedly.
try removing TC on all of them first to see if they will run without
then you can tweak based off the bis it makes

I'd agree, but I can't even force it to try and build a bin by switching to an alive pool...
i ran your config without the TC and it switches for me but times out at 60 sec....
sr. member
Activity: 547
Merit: 250
Fuck this thing it just crashes the drivers repeatedly.
try removing TC on all of them first to see if they will run without
then you can tweak based off the bis it makes

I'd agree, but I can't even force it to try and build a bin by switching to an alive pool...

Oh, and there's this bullshit going on with the fucking fanspeeds too.  After you crash out you start up, fans at fucking 99% regardless of temps.
sr. member
Activity: 547
Merit: 250
I might be a fucking idiot the 4xxx ports are still in the config gimme a sec.

NOPE, absolutely not, this miner now refuses to manually switch pools.
hero member
Activity: 658
Merit: 500
sr. member
Activity: 547
Merit: 250
havent tried the pool switch but i can confirm it works perfect with 14.6
i am getting 5mh/s on my 290's on x11

I'm with you, now try and manually switch to a 3xxx nicehash port and so you can build another .bin

It won't happen.

Check above config for port numbers if you need.
hero member
Activity: 658
Merit: 500
havent tried the pool switch but i can confirm it works perfect with 14.6
i am getting 5mh/s on my 290's on x11
hero member
Activity: 700
Merit: 500
sr. member
Activity: 547
Merit: 250
and you got this to work for more than 60 sec?
that is the only problem i have run into....
Yup - stable for several hours with no issue.
found the problem...
i was missing
"default-profile": "x11",
without the default profile specified it was closing at 60 sec

Found it!

Ok, so default-profile must match your first alive pool.

However, I'm running into a significant problem that IDK has been addressed yet.  After sgminer_5 crashes out of the 60 seconds, it leaves all cards on 'overheat' and the fans start at 100%.  What the fuck.

Also, can't fucking switch pools manually across profiles, not sure what's going on here.
sr. member
Activity: 547
Merit: 250
newbie
Activity: 8
Merit: 0
I'm having trouble creating a config with different gpu threads to handle my two 280x and one 290x (with darkcoin). I'm using the latest v5 with profiles which I understand is per pool, so is there a way to extend it per GPU?  ubuntu 13.04 / cata 13.12

Code:
{
"pools" : [
        {
                "name" : "HashFaster Darkcoin",
                "url" : "stratum+tcp://stratum1.dark.hashfaster.com:4444",
                "user" : "jamin84.shelf2",
                "pass" : "x",
                "profile" : "280x"
        },
        {
                "name" : "DarkPool",
                "url" : "stratum+tcp://drkpool.com:333",
                "user" : "jamin84.shelf2",
                "pass" : "x",
                "profile" : "280x"
        }
],
"profiles" : [
        {
                "name" : "280x",
                "algorithm" : "darkcoin-mod",
                "nfactor" : "10",
                "intensity" : "16",
                "worksize" : "256",
                "gpu-threads" : "2",
                "gpu-engine" : "1050-1050",
                "gpu-memclock" : "1500-1500",
                "thread-concurrency" : "22272"
        },
        {
                "name" : "290x",
                "algorithm" : "darkcoin-mod",
                "nfactor" : "10",
                "intensity" : "20",
                "thread-concurrency" : "40100",
                "worksize" : "256",
                "gpu-threads" : "1",
                "gpu-engine" : "938-1050",
                "gpu-memclock" : "1500-1500",
                "thread-concurrency" : "40100"
        }
],
"default-profile" : "280x",
"failover-only" : true,
"failover-switch-delay" : "30",
"no-pool-disable" : true,
"no-submit-stale" : true,
"kernel-path" : "/usr/local/bin",
"kernel" : "alexkarold,ckolivas,zuikkis",
"lookup-gap" : "2,2,2",
"gpu-powertune" : "0,0,0",
"auto-fan" : true,
"temp-target" : "86,86,86",
"temp-overheat" : "92,92,92",
"temp-cutoff" : "95",
"queue" : "1",
"scan-time" : "7",
"expiry" : "28",
"log" : "5"
}


New build is beautiful, getting 3mH/s on each 280x and dropped temps 10+ degrees.
hero member
Activity: 658
Merit: 500
and you got this to work for more than 60 sec?
that is the only problem i have run into....
Yup - stable for several hours with no issue.
found the problem...
i was missing
"default-profile": "x11",
without the default profile specified it was closing at 60 sec
hero member
Activity: 700
Merit: 500
and you got this to work for more than 60 sec?
that is the only problem i have run into....
Yup - stable for several hours with no issue.
Jump to: