Topic: [ANN] sgminer v5 - optimized X11/X13/NeoScrypt/Lyra2RE/etc. kernel-switch miner - page 199. (Read 877859 times)

I will upload my bin..7950@1040/1500 ~11mh/s..shortly

edit

https://mega.co.nz/#!LAdCXIra!KD_Dil8Dx8UQ0BrvpNWHP16hj_Sc6V1Yx1Ev9MYAaNg

using

Good luck 

My miners runs off a USB flash drive, I have only 10% free, If I try to update the driver, I risk run out of flash space, happened before and had to reinstall everything

the only option (unless someone can help with the bin files) is to install one of the miners from scratch on a hard drive, once everything is working, copy the bin files to the USB flash drive and get rid of the hard drive.

but it is a lot of work that I will really love to avoid if possible

Why not just update to 14.6?

Hello
I am mining Nist5 (talkcoin-mod) algo with 4x 7950, but can't get more than 5.2MH per card.
Using 14.4 Driver on Xubuntu 12.04, kindly anyone can provide a 14.6 BIN file for the 7950 on talkcoin-mod?

I have one but it says Nist5 instead of talkcoin-mod, rename it, but I get tons of HW errors, so obviously it is not compatible with this version or my cards, or the OS.

thank you!.

It's working great now. Thanks!

do
git submodule init
git submodule update
first
or when doing git
git clone --recursive -b v5_0 git://github.com/sgminer-dev/sgminer.git

I'm trying to run this on opensuse.
If I try compiling it myself, I get:

Making all in submodules
make[2]: Entering directory `/home/jacob/bin/sgminer/submodules'
Making all in jansson
make[3]: Entering directory `/home/jacob/bin/sgminer/submodules/jansson'
make[3]: *** No rule to make target `all'.  Stop.
make[3]: Leaving directory `/home/jacob/bin/sgminer/submodules/jansson'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/home/jacob/bin/sgminer/submodules'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/jacob/bin/sgminer'
make: *** [all] Error 2


If I try running the x86_64 binary from nicehash, I get a libcurl error: no version information available

I have libcurl and all dependencies.

Hi guys,
Sorry, I'm not able to read the whole thread regarding my problem. Probably someone has a quick answer to this:

I'm starting the miner with the following config file:

{
"pools" : [
   {
      "name" : "Betarigs X13",
      "url" : "stratum+tcp://r1182.g19.rigs.eu.betarigs.com:4420",
      "user" : "user",
      "pass" : "x",
      "pool-priority" : "0",
      "algorithm" : "marucoin-mod",
      "intensity" : "18",
      "gpu-threads" : "2",
      "worksize": "128"
   },   
   {
      "name" : "NH x13 auto",
      "url" : "stratum+tcp://stratum.nicehash.com:4337",
      "user" : "BTC address here",
      "pass" : "d=.01;f0=0;f2=0;f3=4.14;f4=3.18;f5=0;f6=2.66;f7=0",
      "algorithm" : "marucoin-mod",
      "intensity" : "18",
      "gpu-threads" : "2",
      "worksize": "128"
   },
   {
      "name" : "NH x11 auto",
      "url" : "stratum+tcp://stratum.nicehash.com:4336",
      "user" : "BTC address here",
      "pass" : "d=.04;f0=0;f2=0;f3=4.14;f4=3.18;f5=0;f6=2.66;f7=0",
      "algorithm" : "darkcoin-mod",
      "intensity" : "17",
      "gpu-threads" : "2",
      "worksize": "256"
   },
   {
      "name" : "NH x15 auto",
      "url" : "stratum+tcp://stratum.nicehash.com:4339",
      "user" : "BTC address here",
      "pass" : "d=.01;f0=0;f2=0;f3=4.14;f4=3.18;f5=0;f6=2.66;f7=0",
      "algorithm" : "bitblock",
      "intensity" : "16",
      "gpu-threads" : "2",
      "worksize": "64"
   },
   {
      "name" : "TradeMyBit E2 X13",
      "url" : "stratum+tcp://east02.us.trademybit.com:5550",
      "user" : "user",
      "pass" : "x",
      "algorithm" : "marucoin-mod",
      "intensity" : "18",
      "gpu-threads" : "2",
      "worksize": "128"
   }   
],

"hamsi-expand-big" : "1",
"extranonce-subscription" : true,
"show-coindiff" : true,

"gpu-engine" : "1140,1140,1120,1150",
"gpu-memclock" : "1400,1400,1400,1400",
"gpu-powertune" : "0",
"gpu-vddc" : "0",
"auto-fan" : true,
"gpu-fan" : "10-100,10-100,10-100,10-100",

"temp-target" : "69,70,71,72",
"temp-overheat" : "86",
"temp-cutoff" : "95",
"temp-hysteresis" : "2",

"disable-rejecting" : true,
"failover-only" : true,
"failover-switch-delay" : "30",
"expiry" : "1",
"gpu-dyninterval" : "7",
"hotplug" : "5",
"log" : "5",
"queue" : "0",
"scan-time" : "1",
"no-submit-stale" : false,
"no-restart" : true,

"api-port" : "4028",
"api-listen" : true,
"api-allow" : "W:127.0.0.1,W:192.168.2.1/25"

}
 
The miner starts and silently shuts down after couple of seconds. When I start it with -T option it prints text below and shuts down after I press any key:

C:\Utils\Sgminer5>C:\Utils\Sgminer5\sgminer.exe -T --config C:\Utils\Sgminer5\sg
miner-auto.conf
[13:48:50] Started sgminer 4.2.2-265-g363f-dirty
[13:48:50] * using Jansson 2.6
[13:48:50] Loaded configuration file C:\Utils\Sgminer5\sgminer-auto.conf

[13:48:52] Probing for an alive pool
[13:48:53] Switching to NH x11 auto - first alive pool
[13:48:53] Startup GPU initialization... Using settings from pool NH x11 auto.

[13:48:53] Startup Pool No = 1

C:\Utils\Sgminer5>pause
Press any key to continue . . .

If I start it with "old-fashion" config file, without auto-switching options specified under the "pools" section, it works perfectly. What could it be? Thank you.

Just wondering if anyone else is getting the following error while trying to mine on X13 on Sgminer V5:

"/tmp/OCL774T11.cl", line 930: error: identifier "mixtab2" is undefined
      FUGUE512_3((hash.h4[0x0]), (hash.h4[0x1]), (hash.h4[0x2]));
      ^

"/tmp/OCL774T11.cl", line 930: error: identifier "mixtab3" is undefined
      FUGUE512_3((hash.h4[0x0]), (hash.h4[0x1]), (hash.h4[0x2]));
      ^

"/tmp/OCL774T11.[09:45:37] Failed to init GPU thread 1, disabling device 1
[09:45:37] thread_prepare failed for thread 1Segmentation fault

This isn't priority because this is only on my linux miner running two 7850s using 14.6 currently and 13.12 in the past and I don't get this error on the other linux miner running two R9 270s, one R9 270x and one R9 280 using 14.6.

I compile the source all the time and the 7850 miner has never worked on V5 for X13, so I have to go to different Sgminer and compile that instead of multi-algo mining with same miner.

Just compiled today using version sgminer 4.2.2-280-g5bd2-dirty and same error.  Used both marucoin-mod and marucoin-modold just to be sure.

This occurs whether the config is X13 only or a multi-algo config.  With the multi-algo config it errors like this when I switch to the X13 pool.  I've used fully configured config down to barebones config with same results along with same config used on working sgminer with the exception of algo name change from X13mod to marucoin-mod.

Prior to error screen it says building binary marucoin-modPitcairngw12814big4.bin .  Also copied the bin from the working sgminer, renamed it and only difference is lack of "building" message.

Just let me know if I have to submit a bug error as I don't have a github account right now.

Thank you

Thanks, it work

i found this out because my rig runs at 88 for its operating temp and it was throttling my cards to nothing
i really need to buy some new heat sinks or something... they are really loud and run at max temp even on dark

Interesting, thanks for pointing it out.

Both my rigs have been running seemingly problem free but I've changed that now so will see if it makes any difference...

To get cgwatcher working you need to make two files, like these.

profiles.conf:


sgminer.conf

this does not work well either as cgminer does not see anything in that second file
it will have no information on your gpu-temp settings ant will likely throttle your intensity to nothing

cgwatcher does like the use of "profiles", so everything after your pools you should put into a separate file and then "include" that file into the pools config file
[/quote]

Really?  Since when?  1.3.9 and 1.4.0 explicitly deny profiles.conf

Post a working config please?

[/quote]

cgwatcher does like the use of "profiles", so everything after your pools you should put into a separate file and then "include" that file into the pools config file