Pages:
Author

Topic: Krnlx Nvidia xevan miner - 3.3+ mh on 1070, ~6mh on 80ti FREE, OPENSOURCE (Read 30073 times)

newbie
Activity: 231
Merit: 0
hello there,


if you have problems like :  "Cuda error in func 'quark_blake512_cpu_setBlock_80' at line 701 : unspecified launch failure"

the problem is your intensity is set to high , lower intensity then try again , it will work i am playing with this miner on a MSI GTX 750 2 GB and get with algo xevan : 548.60 KH/s !
not bad for an GTX 750 2GB and its not the i version


[edit]

this error "Cuda error in func 'quark_blake512_cpu_setBlock_80' at line 701 : unspecified launch failure , has to do with your memory allocation clock on your GPU/CPU

Question >: can some one ex plain what this means in the miner > :  [S/A/T]: 0/246/246, diff:
jr. member
Activity: 504
Merit: 3
Can someone share ccmner/krnlx-enemy.mod01? Please private message to me

yes please...anyone have the ccminer/krnlx-enemy.mod01  miner?
newbie
Activity: 162
Merit: 0
My 2 rigs were running this miner just fine until I added a new card to fill out the second rig. 

Now only this rig is consistently crashing with the error:

Cuda error in func 'quark_blake512_cpu_setBlock_80' at line 701 : unspecified launch failure.                         

I've tried removing OC, lower intensity and have had no luck.  Both rigs are now at 6 cards mixing 1070TIs and 1070s.

Any ideas?

Thx!

Hi Friend ,

I also have the same problem , my rig has 8 cards 1070ti and the miner is krnlx/ccminer-xevan but in the last three days i have the same error,,
Have you could to solve this trouble?
Thanks if you could share something with me.
 Grin



same issue with 1070TI freezing PC, fan 100%
Any suggestions?
newbie
Activity: 38
Merit: 0
Hi all! This is prealpha version of my xevan miner. Tested only on linux, cuda 7.5 and 1070 & 1080ti.

https://github.com/krnlx/ccminer-xevan

Donations are welcome.


UPD. For 1060 cards intensity sets up incorrectly - you must adjust it with -i param. -i 20 is best for 1070, 1080ti

UPD2. Windows version by Palgin

x86_64
https://mega.nz/#!NhhnGahJ!dGC_LNOi_a98BuHSuLZQB7k-YT-dM7I_We_svn1hCc0
x86
https://mega.nz/#!98xj3RyD!tfWT6yKHb7gQhzJMi0YYx7wwIX75wppE3Hw7grcMR-4

Hey, krnlx!
Any chance to rewrite/improve performance Xevan algo?
newbie
Activity: 6
Merit: 0
Can someone share ccmner/krnlx-enemy.mod01? Please private message to me
member
Activity: 92
Merit: 10
I more or less randomly faced the following error:

Code:
Cuda error in func 'quark_blake512_cpu_setBlock_80' at line 701 : unspecified launch failure.

After some Google'ing there are some people stating that this has something todo with over or underclocking the GPU's.
My 1080Rig is overclocked so I build myself the following systemd file to enable the watchdog function which in case restarts the xevan-ccminer:

Code:
[Unit]
Description=CCMiner
After=syslog.target network.target
StartLimitBurst=10
StartLimitIntervalSec=120

[Service]
Type=simple
ExecStart=/usr/bin/screen -S ccminer -L -Logfile /home/miner/ccminer.log -Dm /home/miner/ccminer-xevan/ccminer -a xevan -o [STRATUM_URL] -u [USERNAME] -p x --api-remote -q --no-color --api-bind=192.168.178.10:4068
ExecStop=/usr/bin/killall ccminer
User=miner
Restart=always
RestartSec=3

[Install]
WantedBy=multi-user.target

This can possibly help someone ...
member
Activity: 92
Merit: 10
anyone have the problem where the miner closes after a few minutes - every time? I'm not entirely sure what the error or if there even is one displayed because the command prompt disappears too quickly.

I do have some slight overclock on my gpus but most other algos with various other miners seem to mine fine. Was just wondering if anyone else had this problem and figured out what was going on.

How do yo start the miner? Are you using Windows or Linux? I would guess Windows Wink.
Can you put in the script you are using to start the miner?
member
Activity: 92
Merit: 10
I created an pull request because the xevan-ccminer does not compile with OpenSSL >= 1.1.0.
https://github.com/krnlx/ccminer-xevan/pull/11
newbie
Activity: 34
Merit: 0
newbie
Activity: 2
Merit: 0
anyone have the problem where the miner closes after a few minutes - every time? I'm not entirely sure what the error or if there even is one displayed because the command prompt disappears too quickly.

I do have some slight overclock on my gpus but most other algos with various other miners seem to mine fine. Was just wondering if anyone else had this problem and figured out what was going on.
sr. member
Activity: 518
Merit: 251
Trustee Wallet Signature Bounty
Hey, guys!

Are you planned to make compiler for Ubuntu?

Thanks!
newbie
Activity: 172
Merit: 0
What intensivety better for p106-100 and 1066?
newbie
Activity: 10
Merit: 0
hello guys
i try use
--quiet and some another
--plimit=
--tlimit=85
--max-temp=
not work miner just not accept this command...

command worked well
--no-color        disable colored output

just very need --quiet  and --tlimit or --max-temp
today freezee afterbernet and temp go to 80C.. in another miner Ccminer on another algo no this problem
sr. member
Activity: 420
Merit: 250
I guess this krnlx ccminer no longer being updated.
But still works for xevan
2 1080ti and 1 1070ti getting about 14.2 mh
However I don't see what each individual card is doing
Does my hash rate sound good?


yes, it is a pity this great miner stopped to be updated, I hope there could be some more improvements to be achieved
maybe tpruvot could look further into in and get it done? dont't no..
but still this is the best miner we have for xevan for nv cards...

newbie
Activity: 394
Merit: 0
Tell me, how can I download the miner haven for Windows!The link from the main page does not work?
member
Activity: 144
Merit: 10

What does it mean GPU#0:NONCE FOUND? (marked in red) Does it even work (gpu 0)?


dont worry ... its same like block found
newbie
Activity: 1
Merit: 0
Hi all!
Ubuntu 16.04 LTS, Cuda 8.0, Nvidia Xserver (prefer high performance), 5 x 1060 3Gb Gigabyte
ccMiner today biuld from Github. readme.txt: release 1.7.1 (Jan 2015) "Sibcoin & Whirlpool midstate"

sudo ./ccminer -a xevan -o stratum+tcp://stratum.gos.cx:3738 -u xxx -p c=XHM
2018-02-27 13:16:01] GPU#0:NONCE FOUND
[2018-02-27 13:16:01] GPU#1:target 0 ffc00000 3f
[2018-02-27 13:16:01] GPU#1:target 3fffc00000
[2018-02-27 13:16:01] [S/A/T]: 0/304/304, diff: 0.036, 9832.42kH/s yes!
[2018-02-27 13:16:08] GPU#0:NONCE FOUND
[2018-02-27 13:16:08] GPU#2:GeForce GTX 1060 3GB, 1981.43kH/s
[2018-02-27 13:16:08] GPU#2:target 0 ffc00000 3f
[2018-02-27 13:16:08] GPU#2:target 3fffc00000
[2018-02-27 13:16:08] [S/A/T]: 0/305/305, diff: 0.018, 9831.88kH/s yes!
[2018-02-27 13:16:10] GPU#0:NONCE FOUND
[2018-02-27 13:16:10] GPU#0:target 0 ffc00000 3f
[2018-02-27 13:16:10] GPU#0:target 3fffc00000
[2018-02-27 13:16:10] [S/A/T]: 0/306/306, diff: 0.025, 9831.53kH/s yes!
[2018-02-27 13:16:11] GPU#0:NONCE FOUND
[2018-02-27 13:16:11] GPU#4:GeForce GTX 1060 3GB, 1964.88kH/s
[2018-02-27 13:16:11] GPU#4:target 0 ffc00000 3f
[2018-02-27 13:16:11] GPU#4:target 3fffc00000
[2018-02-27 13:16:11] [S/A/T]: 0/307/307, diff: 0.293, 9831.06kH/s yes!

What does it mean GPU#0:NONCE FOUND? (marked in red) Does it even work (gpu 0)?

WBR.
newbie
Activity: 10
Merit: 0
My 2 rigs were running this miner just fine until I added a new card to fill out the second rig. 

Now only this rig is consistently crashing with the error:

Cuda error in func 'quark_blake512_cpu_setBlock_80' at line 701 : unspecified launch failure.                         

I've tried removing OC, lower intensity and have had no luck.  Both rigs are now at 6 cards mixing 1070TIs and 1070s.

Any ideas?

Thx!

Hi Friend ,

I also have the same problem , my rig has 8 cards 1070ti and the miner is krnlx/ccminer-xevan but in the last three days i have the same error,,
Have you could to solve this trouble?
Thanks if you could share something with me.
 Grin

newbie
Activity: 47
Merit: 0
I havent been able to compile it for ubuntu 16.04.
I have 1080TI and cuda 9.1 installed, with nvidia-390 drivers.

I get an error:
Code:
ccminer.cpp:45:26: fatal error: cuda_runtime.h: No such file or directory

I know that other miners, like ccminer, requires editing the nvcc version and keeping only sm_61 for my graphics card, but in Makefile.am its not even listed.
I tried to add manually, but stumble to the same error.

Anyone have a clue about that?


I had the same problem. after hours of banging my head off the wall I found the configure.sh file was looking for cuda 7.5. I just changed it to my current version and bingo off to the races
newbie
Activity: 2
Merit: 0
I havent been able to compile it for ubuntu 16.04.
I have 1080TI and cuda 9.1 installed, with nvidia-390 drivers.

I get an error:
Code:
ccminer.cpp:45:26: fatal error: cuda_runtime.h: No such file or directory

I know that other miners, like ccminer, requires editing the nvcc version and keeping only sm_61 for my graphics card, but in Makefile.am its not even listed.
I tried to add manually, but stumble to the same error.

Anyone have a clue about that?
Pages:
Jump to: