Looking forward to this! It'll make some very nice inroads on that new masternode, indeed.

I'm not sure I have the capability of blind walking you through directory copying and file editing on Windows if you've never done it before.  Usually my "go to" solution for this situation is to recommend hiring a kid in the neighborhood to come do it for you for a few bucks. Kids always know how to hack any computer.

Mostly you want to right click on the folder/directory, choose the 'copy' from the dropdown menu, hold down the ctrl key and tap the 'v' key. That should create a new directory renamed with a " - Copy" added into the name. You can edit your .json file in there for your remaining GPU's and then run the miner bat file from that new directory.

To hit the other cards, you can create a new .json file and modify its "DEVICES" entry to be [4,5,6] instead of "AUTO" (or whatever numbers your remaining cards are - FYI, I get core dumps if I select a number that goes beyond my available cards)
I just copied the entire directory into a new directory, edited the copied .json and ran lolMiner in that new directory so that it picked up the edited .json by default. I believe you can also just copy the .json file to a new name in the same directory and use the --usercfg command line option to point to the new .json file. I didn't get too elaborate with my config.

Big, there's a known problem that occurs sometimes with using 4+ cards at the same time under a single lolMiner process which may be causing your issue.

Try adjusting your .json from:
"DEVICES" : "AUTO"
to:
"DEVICES" : [0,1]

and see if that runs without problems. If so, you can add additional GPU's in there until it becomes unstable. I'm currently running just 3 GPU's per lolMiner process.

Nice progress, Lolliedieb. I've since changed from AMD driver to ROCM 1.9, so the comparison to the numbers above may not be apples to apples, but here are the results using ROCM 1.9 on Ubuntu 18.04 with .06 alpha for stand-alone review:

Vega 56 (fully powered, only compute option set) :
Average speed (300s): 40.8 sol/s | 38.7 sol/s | 40.2 sol/s Total: 119.6 sol/s

Vega 64 (fully powered, only compute option set) :
Average speed (300s): 45.0 sol/s | 45.5 sol/s Total: 90.5 sol/s

Sounds good, Lolliedieb! It would be great if you could begin tweaking it to run well on Linux ROCm 1.9.x if you have time available. It seems like that's the direction AMD linux support is heading...

Version .41 was getting 18-19 sol/s on Vega 56's (144_5).

Also, I undervolted. Fully powered Vega 56's are getting:
Average speed (300s): 32.7 sol/s | 32.9 sol/s | 38.6 sol/s | 36.9 sol/s | 38.7 sol/s Total: 179.8 sol/s

Ignore the first two numbers - I had to undervolt my 64's as full watts swamped my ps. The last three are 56's, so it's averaging around 37-39 sol/s fully powered. Just don't want to spend an additional 60w per card to get another 3-5 sol/s

The fully powered (240w) Vega 64's (first two numbers) are giving me 41-43 sol/s:
Average speed (300s): 41.8 sol/s | 42.5 sol/s | 31.2 sol/s | 29.1 sol/s | 31.0 sol/s Total: 175.5 sol/s

Decent progress so far...

On Ubuntu 18.04.1 running AMD 18.30 drivers, I'm getting the following hashrate for two vega 64s (first two) and two vega 56s on 144_5 (ZEL):
Average speed (300s): 36.4 sol/s | 38.0 sol/s | 34.8 sol/s | 33.2 sol/s Total: 142.4 sol/s
Average speed (300s): 37.7 sol/s | 37.5 sol/s | 34.6 sol/s | 32.8 sol/s Total: 142.6 sol/s
Average speed (300s): 36.6 sol/s | 38.3 sol/s | 34.3 sol/s | 33.0 sol/s Total: 142.3 sol/s

Undervolted the 64s to 130w, and the 56s to 110w and changed them to compute - no other changes (not quite convinced other changes have any effect in this driver).

--- edit: add 192_7 numbers ---
Same undervolting 56:110, 64:130 (with an additional vega56) for 192_7 (GENX). First two 64s, last 3 56s:
Average speed (300s): 23.5 sol/s | 24.1 sol/s | 20.7 sol/s | 19.8 sol/s | 20.4 sol/s Total: 108.4 sol/s
Average speed (300s): 23.9 sol/s | 24.1 sol/s | 20.8 sol/s | 19.6 sol/s | 20.1 sol/s Total: 108.7 sol/s

Ok, my bad. I tried to double up a single power cable to one of the Vega 64's instead of using two separate power inputs. Correcting that config resolved the crash (which never had a kernel, panic, btw. It would simply reboot, which spoke to hardware level behavior).

Sorry for the false alarm.

So I'm seeing an odd behavior with 144.5 mining (zelcash in this case, 2 Vega64's, 3 Vega56's). If/when I lose connection to the pool, the moment the worker is reauthorized, my linux box crashes and reboots:
-----------------
Average speed (5s): 34.0 sol/s | 32.6 sol/s | 33.4 sol/s | 30.8 sol/s | 27.6 sol/s Total: 158.3 sol/s
Submitting share
Share accepted
Submitting share
Submitting share
Share accepted
Share accepted
Average speed (5s): 34.2 sol/s | 32.0 sol/s | 30.6 sol/s | 29.0 sol/s | 28.2 sol/s Total: 154.0 sol/s
Submitting share
Share accepted
Lost connection to stratum server equigems.online:9000 or server not reachable.
Trying to connect in 5 seconds
Lost connection to stratum server equigems.online:9000 or server not reachable.
Trying to connect in 5 seconds
Connected to equigems.online:9000
Subscribed to stratum server
New target received: 00a0000000000000
New job received: cccd
Authorized worker:
-----------------
and then the immediate reboot. I don't see the kernel panic in the log, but I see ubuntu 18 uses syslog caching by default. I've removed caching and we'll see if it catches something on the next crash.

This is more of an FYI unless you see something in your reconnect code that might cause a panic. However, I'm not even sure whether the lost connection is a symptom of an imminent crash or the actual cause. Who knows, it could simply be instability with one of my risers...

I'll see if I can track down more info in the logs next time.

edit: I also notice that if I ctrl-c out of the miner, it's a 50/50 toss up that restating the miner causes a reboot on reconnect. So this is likely related. Perhaps something about the state that the card is left in during an abnormal stop? Reuse that card and cause the panic? I find that I usually have to power off the clear the cards before the system is stable again. Just speculating...

I'm undervolting and using the latest amd 18.30 drivers for ubuntu.

Thanks Lolliedieb, working great as usual. I'm still mining away on HEPTA with my vega/rx570 rig. I have a request, and not sure if it's possible, but I notice that my pool rated efficiency goes down as I start undervolting the cards (I presume from bad blocks I'm sending in?). Is there any way we can get some sort of percentage of valid results from your internal checks for each of the cards that you display sol/s on? Something I could use to tell if I've undervolted too far on one card or if any particular card is returning more bad blocks  than another? Is that even possible?

Just a thought. Otherwise, I'm always happy to see any performance improvement that you could gain from future patches!

Thanks,
Cyg

edit: Also, would there be any benefit to adding a CPU option to the miner? I'm not sure how many sol/s we could expect from it but I feel like I need the cpu on this rig working towards something instead of sitting idle... 

I'm able to get lolMiner working with two Vega 56's on Ubuntu 18.04.4 with amdgpu-pro-18.30-633530.tar.xz (legacy and pal both installed) by symlinking libamdoclsc64.so to libamdocl64.so and making certain I had the 'video' group on the user running the miner.

However, I'm with wojcieh and am only able to get 31-33 sol/s out of each of my two stock MSI airboost vega 56s (I undervolt them, but even at normal voltage I get max of 33). Are you seeing that 56 - 63 sol/s performance only on Windows? Did you make vbios changes?

That's a pretty amazingly dramatic performance improvement.

Wow, nice improvement in speed! Now running:

---------------------------------------------
Average speed (60s): 29.4 sol/s | 31.0 sol/s | 17.5 sol/s | 19.0 sol/s Total: 97.0 sol/s
---------------------------------------------

Excellent work, Lolliedieb!

For SnowGem I used:
"POOL" : "mining.miningspeed.com",
"PORT" : "3022",

Here's an example of what I use on Ubuntu. Is this what you're looking for?

$ cat user_config.json
{
        "DEFAULTS" : {
                "DEVICES" : "AUTO",
                "APIPORT" : 0
        },

   "HEPTA_NIBIRU" :
   {
      "COIN" : "HEPTA",
      "POOLS" : [
         {"POOL" : "hepta.nibirupool.com",
          "PORT" : "8138",
          "USER" : ".",
          "PASS" : "any"}
      ]
   }
}

$ cat run_hepta_nibiru.sh
#!/bin/bash

#################################
## Begin of user-editable part ##
#################################

PROFILE=HEPTA_NIBIRU

export GPU_FORCE_64BIT_PTR=1
export GPU_MAX_HEAP_SIZE=100
export GPU_USE_SYNC_OBJECTS=1
export GPU_MAX_ALLOC_PERCENT=100
export GPU_SINGLE_ALLOC_PERCENT=100
#################################
##  End of user-editable part  ##
#################################

cd "$(dirname "$0")"
while true
do
  ./lolMiner -profile=$PROFILE $@
  if [ $? -eq 134 ]
  then
    break
  fi
done

Yup.   Even running at normal voltage:

~$ cat /sys/class/drm/card0/device/hwmon/hwmon0/power1_cap
165000000
~$ cat /sys/class/drm/card2/device/hwmon/hwmon1/power1_cap
90000000
~$ cat /sys/class/drm/card3/device/hwmon/hwmon2/power1_cap
165000000
~$ cat /sys/class/drm/card4/device/hwmon/hwmon3/power1_cap
90000000

Only gets me a avg additional 10 sol/s or so...

Average speed (5s): 21.8 sol/s | 24.8 sol/s | 13.0 sol/s | 12.6 sol/s Total: 72.2 sol/s
Average speed (5s): 23.6 sol/s | 21.8 sol/s | 10.8 sol/s | 12.8 sol/s Total: 69.0 sol/s
Average speed (5s): 23.8 sol/s | 19.4 sol/s | 12.4 sol/s | 13.4 sol/s Total: 69.0 sol/s
Average speed (5s): 19.4 sol/s | 16.6 sol/s | 11.6 sol/s | 12.2 sol/s Total: 59.8 sol/s
Average speed (5s): 19.2 sol/s | 23.4 sol/s | 13.6 sol/s | 14.0 sol/s Total: 70.2 sol/s
---------------------------------------------
Average speed (60s): 19.7 sol/s | 20.4 sol/s | 13.2 sol/s | 13.0 sol/s Total: 66.3 sol/s
---------------------------------------------

...and does a great impression of a kerosene room heater.

Welcome back Lolliedieb, hope your vacation was both fun and restful.

Your version 0.41 has been running on my Ubuntu 18 rig (HEPTA) without fail for days, now. Undervolted two Vega 54's (to 90w) and two RX 570's (to 80w)

Average speed (5s): 15.4 sol/s | 21.2 sol/s | 9.6 sol/s | 10.0 sol/s Total: 56.2 sol/s
Average speed (5s): 22.2 sol/s | 18.8 sol/s | 10.2 sol/s | 11.4 sol/s Total: 62.6 sol/s
Average speed (5s): 16.6 sol/s | 19.6 sol/s | 9.2 sol/s | 11.6 sol/s Total: 57.0 sol/s
Average speed (5s): 17.0 sol/s | 19.4 sol/s | 10.2 sol/s | 9.6 sol/s Total: 56.2 sol/s
Average speed (5s): 16.2 sol/s | 18.2 sol/s | 12.2 sol/s | 9.6 sol/s Total: 56.2 sol/s
---------------------------------------------
Average speed (60s): 17.5 sol/s | 18.5 sol/s | 10.0 sol/s | 10.7 sol/s Total: 56.8 sol/s
---------------------------------------------

I'm looking forward to any performance improvements you can make, though. Good luck!

On a headless Ubuntu 18.04 server (updated to latest patch) using amdgpu-pro-18.20-606296 (had to hack it a bit by symlinking libamdoclsc64.so to libamdocl64.so and fixing some permissions - strace is a wonderful tool) using two stock MSI Vega 56's, it's working. Mostly.

I get:
Average speed (60s): 21.2 sol/s | 19.9 sol/s Total: 41.1 sol/s

Although one of the 56's locked up after about an hour or so and zombied the process. Had to hard reboot the system:
===============================================
Average speed (5s): 22.0 sol/s | 21.4 sol/s Total: 43.4 sol/s
New job received:
Submitting share
Share accepted
Average speed (5s): 18.2 sol/s | 19.2 sol/s Total: 37.4 sol/s
New job received:
Average speed (5s): 4.6 sol/s | 7.4 sol/s Total: 12.0 sol/s
---------------------------------------------
Average speed (60s): 17.2 sol/s | 18.6 sol/s Total: 35.9 sol/s
---------------------------------------------
Average speed (5s): 0.0 sol/s | 0.0 sol/s Total: 0.0 sol/s
Average speed (5s): 0.0 sol/s | 0.0 sol/s Total: 0.0 sol/s
Error: lolMiner detected that device 0 is stucked
Shutting down lolMiner
===============================================