Pages:
Author

Topic: [ANN] ccminer 2.3 - opensource - GPL (tpruvot) - page 60. (Read 500193 times)

sr. member
Activity: 392
Merit: 266
EthMonitoring.com
Latest ccminer 2.2 works like a charm with https://ethmonitoring.com (supported)
member
Activity: 70
Merit: 10
skunk algo

spmod5 228mh
cc 2.2 244mh
full member
Activity: 350
Merit: 100
I built 2.2 with cuda 8.0 on ubuntu 16.04 64, the rig has 9 cards, but I always run in to below error.
if I use -d 0.1.2.3.4.5.6.7 or -d 1,2,3,4,5,6,7,8 to run 8 cards only, it works well.:
Code:
*** ccminer 2.2 for nVidia GPUs by tpruvot@github ***^M
    Built with the nVidia CUDA Toolkit 8.0 64-bits^M
^M
  Originally based on Christian Buchner and Christian H. project^M
  Include some algos from alexis78, djm34, sp, tsiv and klausT.^M
^M
BTC donation address: 1AJdfCpLWPNoAMDfHF1wD5y8VgKSSTHxPo (tpruvot)^M
^M
[2017-08-14 02:23:29]^[[36m Starting on stratum+tcp://sigt.suprnova.cc:7106^[[0m^M
[2017-08-14 02:23:29] NVML GPU monitoring enabled.^[[0m^M
[2017-08-14 02:23:29] 9 miner threads started, using 'skunk' algorithm.^[[0m^M
[2017-08-14 02:23:31]^[[33m Stratum difficulty set to 1^[[0m^M
[2017-08-14 02:23:31] GPU #3: Intensity set to 20, 1048576 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #5: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #1: Intensity set to 21, 2097152 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #4: Intensity set to 20, 1048576 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #6: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #2: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #8: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #0: Intensity set to 21, 2097152 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #7: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:37] GPU #1: GeForce GTX 1080, 713.70 kH/s^[[0m^M
[2017-08-14 02:23:37] GPU #3: GeForce GTX 1070, 708.17 kH/s^[[0m^M
[2017-08-14 02:23:37] GPU #3: GeForce GTX 1070, 708.17 kH/s^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: unspecified launch failure^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: skunk_setBlock_80:693 unspecified launch failure^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[/code/
full member
Activity: 350
Merit: 100
I built 2.2 with cuda 8.0 on ubuntu 16.04 64, the rig has 9 cards, but I always run in to below error.
if I use -d 0.1.2.3.4.5.6.7 or -d 1,2,3,4,5,6,7,8, it works well.:
Code:
*** ccminer 2.2 for nVidia GPUs by tpruvot@github ***^M
    Built with the nVidia CUDA Toolkit 8.0 64-bits^M
^M
  Originally based on Christian Buchner and Christian H. project^M
  Include some algos from alexis78, djm34, sp, tsiv and klausT.^M
^M
BTC donation address: 1AJdfCpLWPNoAMDfHF1wD5y8VgKSSTHxPo (tpruvot)^M
^M
[2017-08-14 02:23:29]^[[36m Starting on stratum+tcp://sigt.suprnova.cc:7106^[[0m^M
[2017-08-14 02:23:29] NVML GPU monitoring enabled.^[[0m^M
[2017-08-14 02:23:29] 9 miner threads started, using 'skunk' algorithm.^[[0m^M
[2017-08-14 02:23:31]^[[33m Stratum difficulty set to 1^[[0m^M
[2017-08-14 02:23:31] GPU #3: Intensity set to 20, 1048576 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #5: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #1: Intensity set to 21, 2097152 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #4: Intensity set to 20, 1048576 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #6: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #2: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #8: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #0: Intensity set to 21, 2097152 cuda threads^[[0m^M
[2017-08-14 02:23:31] GPU #7: Intensity set to 18, 262144 cuda threads^[[0m^M
[2017-08-14 02:23:37] GPU #1: GeForce GTX 1080, 713.70 kH/s^[[0m^M
[2017-08-14 02:23:37] GPU #3: GeForce GTX 1070, 708.17 kH/s^[[0m^M
[2017-08-14 02:23:37] GPU #3: GeForce GTX 1070, 708.17 kH/s^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: unspecified launch failure^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: skunk_setBlock_80:693 unspecified launch failure^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[2017-08-14 02:23:37]^[[33m GPU #3: result does not validate on CPU!^[[0m^M
[/code/
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
yes, i guess ocminer tweaked it on a 1060. A message about that in its channel generaly fix it Wink
member
Activity: 81
Merit: 12
Not a single miner I have tried is reported accurately on suprnova.cc, so I am guessing it has to be supernova. I see reported hash from 11 MH/s to 45 MH/s and I keep at 17MH/s in reality.

SIGT has been a very hard coin to predict in general no calculators are right either, only way to track is by how much you get paid in the end.
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
yep :/ ZEC bounty empoisoned the opensource work... new devs gave ideas to sp

even nicehash stopped publishing their work

damniatx: no... i already tried, their wallet is weird and not compatible with bitcoin style protocols
member
Activity: 81
Merit: 12
Intensity 17 for ccminer 2.2 seems to lower power draw and temps nicely, 83-84% power, 66-67C with 67% Fan Speed.

Hash is still above others @ 17.7-17.8 MH/s or so.

Thanks for your work btw, too bad so many violate the GPL3 licensing with the private builds.
member
Activity: 109
Merit: 10
The diff change can trigger a small pause (500ms) that generally doesnt hurt Wink

are you looking to add NEXUS (SHA3) algo in the future ?
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
The diff change can trigger a small pause (500ms) that generally doesnt hurt Wink
member
Activity: 81
Merit: 12
Nice, nearly 19 MH/s on a GTX 1060 6GB with intensity at 18 with -202 memory clock ccminer-x86-2.2: https://github.com/tpruvot/ccminer/releases

Was getting 17 MH/s with just about every miner before that on exact same settings.

Power has some drops however, any idea why it keeps happening? Have same thing with krnlx. Goes from 91-89 to 34-35 every minute or so, seem to utilize more power as well compared to others which are 83-86. Would this cause any issues long term?

ccminer-x86-2.2:


3 other miners compared earlier:


legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
try to mine other coins with krnlx miner.. and see the difference, he optimised skunk yes (and i reused that) but he brokes most other algos too
member
Activity: 109
Merit: 10
if there is an issue with that, its suprnova fault.. or a lie of sp.. i dont mine there

and no, i didnt "simply" put knrlx code


woa. I'm so sorry. I got the same hash reporting on with the latest build.
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
if there is an issue with that, its suprnova fault.. or a lie of sp.. i dont mine there

and no, i didnt "simply" put knrlx code
member
Activity: 109
Merit: 10
hero member
Activity: 630
Merit: 502
Is this miner also affected by the inaccurate hashrate reporting?
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
its very close to cuda 8, but requires the latest drivers
hero member
Activity: 714
Merit: 500
I'm using CUDA 8.0, anyone try CUDA 9.0? Just wondering if any improvement on performance?
member
Activity: 121
Merit: 61
yep, actually vs2013 is the only way to support (and compare) all the recent CUDA versions (6.5 to 9.0)
Just now was succefully compiled latest ccminer 2.2 under vs2015. But don't yet tested it, because haven't NVidia GPU on build PC. I make the following patch to the sources to make it compatible with vs2015:

0001-make-sources-vs2015-compatible.patch (it's 12 Mb, because it includes libcurl_a.lib and libeay32MT.lib for both 32/64-bit platforms).

libcurl_a.lib compiled by me from sources and precompiled libeay32MT.lib from here. Both x86 and x64 releases compiled fine.

Everyone can test build under vs2015 with this patch and make a report here. To make patch use:
Code:
git apply 0001-make-sources-vs2015-compatible.patch

p.s. Already tested, all works fine.
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
yep, actually vs2013 is the only way to support (and compare) all the recent CUDA versions (6.5 to 9.0)
Pages:
Jump to: