[ANN] ccminer 2.3 - opensource - GPL (tpruvot)

[Epsylon3]

its similar to a 970 OC, between a 970 and a 980 150~200W

[1713432515]

1713432515

[1713432515]

1713432515

[1713432515]

1713432515

[bensam1231]

Are you going to look into fixing the NeoS bug on Windows on the 1XXX?

DJM is MIA and SP is pretending he doesn't exist.

[Epsylon3]

nope, i dont work on "hobby" projects

[bensam1231]

nope, i dont work on "hobby" projects

Ah, one of the highest earning algos is a 'hobby project'. I see... ~_~

[antonio8]

I replaced just my 970 with a 1080 and the temp in my room has been night and day difference.

The only thing I hate is the price but oh well.

As Epsylon3 said above maybe down the road the cards can be taking advantage of for mining.

antonio What wattage does it draw?  Thx

Go with what Epsylon3 posted above.

I don't have a wattage meter. Never used one to  test any rigs.

[bensam1231]

Speaking of NeoS performance, someone is talking about improving performance in Windows for memory hard algos with the next WDDM 2.1 update, but will require some tweaking by devs.

Thank you for a well informed and informational post.

Does that mean developers need to tweak some stuff for it to work in W7/8?

Epsolyn does some work on CCminer, he might be interested. SP as well, but he is 'retired'.

Check: https://bitcointalk.org/index.php?topic=770064.0

Also not sure how much this will influence Lyra2v2, but check Nanashi's thread: https://bitcointalk.org/index.php?topic=1519266

Yes, devs need to modify the way algo access memory to improve things on memory dependent algos (eth and neoscrypt to my knowledge)
Pascal is 4k pte by default in win 7/8 (where maxwell is 128kb) but can be 2m pte (full memory coverage)
On win 10 current - pascal and maxwell on new drivers are 64kb but pascal will be 2m by default with next driver AND win 10 insider preview and further win 10 july update (which didnt help neoscrypt for some reason - i think i saw 900khs but couldnt repro it - maybe I missed some build or setting i was using)

https://bitcointalk.org/index.php?topic=1368785.msg15363869#msg15363869

[th3.r00t]

@Epsylon3

https://github.com/tpruvot/ccminer/blob/2ee8bc9791be912206c2120b15a72206ec2476b3/configure.sh
Fixing cuda 6.5 in configure prevents from successful build on cuda 7.5

[antho281]

I replaced just my 970 with a 1080 and the temp in my room has been night and day difference.

The only thing I hate is the price but oh well.

As Epsylon3 said above maybe down the road the cards can be taking advantage of for mining.

antonio What wattage does it draw?  Thx

Go with what Epsylon3 posted above.

I don't have a wattage meter. Never used one to  test any rigs.

This is what we get from nvidia-smi. Not the most precise, but still better than nothing

Actually, I'm running them all (6) on a 1300W PSU

[Epsylon3]

@Epsylon3

https://github.com/tpruvot/ccminer/blob/2ee8bc9791be912206c2120b15a72206ec2476b3/configure.sh
Fixing cuda 6.5 in configure prevents from successful build on cuda 7.5

yep indeed, it should be /usr/local/cuda

its how i select the cuda version on linux, and that shoudnt have been commited

[skunk]

hi, just to inform that current head (on decred algo and since this commit) only mines on one random card out of 5 installed:

Code:

$ ./ccminer -n
*** ccminer 1.8-dev for nVidia GPUs by tpruvot@github ***
    Built with the nVidia CUDA Toolkit 7.5

  Originally based on Christian Buchner and Christian H. project
  Include some of the work of djm34, sp, tsiv and klausT.

BTC donation address: 1AJdfCpLWPNoAMDfHF1wD5y8VgKSSTHxPo (tpruvot)

GPU #0: SM 5.2 GeForce GTX 950
GPU #1: SM 5.2 GeForce GTX 950
GPU #2: SM 5.2 GeForce GTX 950
GPU #3: SM 5.2 GeForce GTX 950
GPU #4: SM 5.2 GeForce GTX 950

Code:

$ ./ccminer -a decred -o stratum+tcp://dcr.pool.mn:4800 -u skunk.the -p 123 -D
*** ccminer 1.8-dev for nVidia GPUs by tpruvot@github ***
    Built with the nVidia CUDA Toolkit 7.5

  Originally based on Christian Buchner and Christian H. project
  Include some of the work of djm34, sp, tsiv and klausT.

BTC donation address: 1AJdfCpLWPNoAMDfHF1wD5y8VgKSSTHxPo (tpruvot)

[2016-06-29 12:56:56] POOL 0: dcr.pool.mn:4800 USER skunk.the -s 10
[2016-06-29 12:56:56] Starting on stratum+tcp://dcr.pool.mn:4800
[2016-06-29 12:56:56] restart_threads
[2016-06-29 12:56:56] CUDA GPU 0 matches NVML GPU 0 by busId 1
[2016-06-29 12:56:56] CUDA GPU 1 matches NVML GPU 1 by busId 3
[2016-06-29 12:56:56] CUDA GPU 2 matches NVML GPU 2 by busId 5
[2016-06-29 12:56:56] CUDA GPU 3 matches NVML GPU 3 by busId 6
[2016-06-29 12:56:56] CUDA GPU 4 matches NVML GPU 4 by busId 7
[2016-06-29 12:56:56] NVML GPU monitoring enabled.
[2016-06-29 12:56:56] 5 miner threads started, using 'decred' algorithm.
[2016-06-29 12:56:56] Binding thread 4 to cpu 0 (mask 1)
[2016-06-29 12:56:56] Binding thread 2 to cpu 0 (mask 1)
[2016-06-29 12:56:56] Binding thread 3 to cpu 1 (mask 2)
[2016-06-29 12:56:56] Binding thread 1 to cpu 1 (mask 2)
[2016-06-29 12:56:56] Binding thread 0 to cpu 0 (mask 1)
[2016-06-29 12:56:56] Stratum difficulty set to 1
[2016-06-29 12:56:56] dcr.pool.mn:4800 decred block 41175
[2016-06-29 12:56:56] restart_threads
[2016-06-29 12:56:57] job 773a956 9d54 target change: ffff0000 (1.0)
[2016-06-29 12:56:57] GPU #4: start=00000000 end=3fffffff range=3fffffff
[2016-06-29 12:56:57] job 773a956 9d54 target change: ffff0000 (1.0)
[2016-06-29 12:56:57] GPU #3: start=00000000 end=3fffffff range=3fffffff
[2016-06-29 12:56:57] job 773a956 9d54 target change: ffff0000 (1.0)
[2016-06-29 12:56:57] GPU #2: start=00000000 end=3fffffff range=3fffffff
[2016-06-29 12:56:57] job 773a956 9d54 target change: ffff0000 (1.0)
[2016-06-29 12:56:57] GPU #0: start=00000000 end=3fffffff range=3fffffff
[2016-06-29 12:56:57] job 773a956 9d54 target change: ffff0000 (1.0)
[2016-06-29 12:56:57] GPU #1: start=00000000 end=3fffffff range=3fffffff
[2016-06-29 12:56:59] found => f722221b 1b2222f7
[2016-06-29 12:56:59] GPU #3: GeForce GTX 950, 228.06 MH/s
[2016-06-29 12:56:59] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:56:59] restart_threads
[2016-06-29 12:56:59] accepted: 1/1 (100.00%), 228.06 MH/s yes!
[2016-06-29 12:57:00] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:05] GPU #3: GeForce GTX 950, 771.25 MH/s
[2016-06-29 12:57:05] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:10] GPU #3: GeForce GTX 950, 771.16 MH/s
[2016-06-29 12:57:10] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:10] found => 356d1c05 051c6d35
[2016-06-29 12:57:10] GPU #3: GeForce GTX 950, 771.00 MH/s
[2016-06-29 12:57:10] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:10] outdated job 9d54, new 253 stales=1
[2016-06-29 12:57:10] stale work detected, discarding
[2016-06-29 12:57:11] found => 5a256316 1663255a
[2016-06-29 12:57:11] GPU #3: GeForce GTX 950, 770.81 MH/s
[2016-06-29 12:57:11] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:11] outdated job 9d54, new 253 stales=2
[2016-06-29 12:57:11] stale work detected, discarding
[2016-06-29 12:57:12] dcr.pool.mn:4800 decred block 41176
[2016-06-29 12:57:12] restart_threads
[2016-06-29 12:57:12] job 773a988 61d target change: ffff0000 (1.0)
[2016-06-29 12:57:12] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:17] GPU #3: GeForce GTX 950, 771.04 MH/s
[2016-06-29 12:57:17] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:19] found => 8299a150 50a19982
[2016-06-29 12:57:19] GPU #3: GeForce GTX 950, 770.92 MH/s
[2016-06-29 12:57:19] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:19] accepted: 2/2 (100.00%), 693.46 MH/s yes!
[2016-06-29 12:57:24] GPU #3: GeForce GTX 950, 770.89 MH/s
[2016-06-29 12:57:24] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:29] GPU #3: GeForce GTX 950, 767.62 MH/s
[2016-06-29 12:57:29] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:31] found => 0023a04a 4aa02300
[2016-06-29 12:57:31] GPU #3: GeForce GTX 950, 765.11 MH/s
[2016-06-29 12:57:31] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:31] accepted: 3/3 (100.00%), 715.79 MH/s yes!
[2016-06-29 12:57:36] GPU #3: GeForce GTX 950, 764.96 MH/s
[2016-06-29 12:57:36] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:37] found => 5212630e 0e631252
[2016-06-29 12:57:37] GPU #3: GeForce GTX 950, 764.70 MH/s
[2016-06-29 12:57:37] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:37] accepted: 4/4 (100.00%), 723.96 MH/s yes!
[2016-06-29 12:57:42] GPU #3: GeForce GTX 950, 764.96 MH/s
[2016-06-29 12:57:42] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:45] found => 24a11b7e 7e1ba124
[2016-06-29 12:57:45] GPU #3: GeForce GTX 950, 764.91 MH/s
[2016-06-29 12:57:45] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:45] accepted: 5/5 (100.00%), 729.81 MH/s yes!
[2016-06-29 12:57:50] found => 4b295cd6 d65c294b
[2016-06-29 12:57:50] GPU #3: GeForce GTX 950, 764.69 MH/s
[2016-06-29 12:57:50] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:50] accepted: 6/6 (100.00%), 732.14 MH/s yes!
[2016-06-29 12:57:54] GPU #3: GeForce GTX 950, 764.38 MH/s
[2016-06-29 12:57:54] GPU #3: start=00000000 end=ffffffff range=ffffffff
[2016-06-29 12:57:56] SIGINT received, exiting
[2016-06-29 12:57:56] restart_threads

[DragonSlayer]

hi, just to inform that current head (on decred algo and since this commit) only mines on one random card out of 5 installed:

I am not seeing this with a pair of 980's.  I just did a fresh build 2 days ago.

I will say that yiimp site is only seeing 1 card for the benchmarks.  It does show total hash correctly.

[skunk]

hi, just to inform that current head (on decred algo and since this commit) only mines on one random card out of 5 installed:

I am not seeing this with a pair of 980's.  I just did a fresh build 2 days ago.

i can reliably reproduce the issue on a 5x950 and a 6x970 rig, on both rigs only one card starts hashing (even if the cards are specified with the -d flag or if the number of cards are specified with the -t flag)

I will say that yiimp site is only seeing 1 card for the benchmarks.  It does show total hash correctly.

it has nothing to do with yiimp, i'm not mining there...

[Epsylon3]

never seen a problem like that...

else playing with leds... (windows only)

https://youtu.be/xFso96j7y1M

[bathrobehero]

never seen a problem like that...

else playing with leds... (windows only)

https://youtu.be/xFso96j7y1M

That looks cool!

[pallas]

looks like you're having fun with leds :-)
do you think it may work on linux as well?

[Epsylon3]

it require the nvapi which is only on windows... but maybe i could trace what is doing the nvapi dll

[skunk]

never seen a problem like that...

maybe it happens just on linux? i don't have windows rigs to try...

[Epsylon3]

i only have one pc on windows most are on linux so...

[skunk]

strange, i can reproduce it with "-a blake", "-a blakecoin" beside "-a decred" (even with --benchmark)
blake2s, vanilla and other algos are fine
well, i'll stick with alexis78's ccminer for decred...

[Epsylon3]

should be the same with alexis one

Code:

[2016-07-04 21:59:16] GPU #2: MSI GTX 960, 982.33 MH/s
[2016-07-04 21:59:17] decred block 42664, diff 831049.304
[2016-07-04 21:59:20] GPU #0: ASUS GTX 970, 1557.27 MH/s
[2016-07-04 21:59:21] GPU #2: MSI GTX 960, 982.29 MH/s
[2016-07-04 21:59:22] GPU #1: EVGA GT 740, 103.48 MH/s
[2016-07-04 21:59:22] accepted: 29/29 (diff 7.369), 2627.27 MH/s yes!
[2016-07-04 21:59:22] GPU #0: ASUS GTX 970, 1554.41 MH/s

this is on linux if you really don't believe me