CCminer(SP-MOD) Modded GPU kernels.

[sp_]

Above 0,2 BTC per day for 1 gigahash.

you only need 60 980 cards (10 rigs), or 180 750TI Cards(30 rigs) to reach 1 gigahash.

$)

[jpouza]

There are many profitable quark coins f.ex:

BitQuark
MonetaryUnit
Securecoin
DimeCoin
Quarkcoin

with the latest miner (release 24), 2-3 more profitable than mining x11.  

Above 0,2 BTC per day for 1 gigahash.

you only need 60 980 cards (10 rigs), or 180 750TI Cards(30 rigs) to reach 1 gigahash.

$)

heheh just a few cards hum?

With BTC price = trash, nothing is profitable anymore.

[sp_]

They say that GPU mining is dead, but as long as you get more than the electic bill, and you have enough capital your farm can grow and grow.

[sp_]

With BTC price = trash, nothing is profitable anymore.

This is correct

[go6ooo1212]

Nice calculation , I like it

[PVmining]

This is correct for some algorithms. Didn't expect the price to go below $300 / BTC

with the current BTC price and 180 750ti you will make $60 a day when mining quarkcoins/clones.
Power consumption will be below 7WATT per MHASH (around 5000KWh a month)

income per month: 1830$ (6.1BTC)
Power cost per month (?): (@0.1$ per KWh) 500$

Profit: $1330

...and you will need 25months to return your hardware investments, which means "never ever" in the crypto game
I'm a crypto beliver, but the energy costs are that high at my place that I will shut down some of my rigs too as many of you.

sp_ thanks for your efforts... I mined other algos, but I liked to see your great miner performance upgrades.

[Handofmidas]

Tested Coinotron with release24. Unfortunately there are no graphs there, but logged in regularly and the hashrate was always between 14 and 15 MH, instead of the anticipated 16.7 MH. AFAIK Coinotron does not have vardiff, in contrast to coinmine, so I'm getting shares quite frequently there.

[squadbox]

I was only getting about 8 Mh/s effective hashrate at the pool with r24 as opposed to 10.5 Mh/s with r23.  I'll be sticking with r23 for now.

those hashrates are superb ... but as sp works on the miner and optimizations - there is bound to be some mishaps along the way ...

im still runing the latest git pull - and will stay on it for now - to test stability also ...

i too have issues with extranonce2 errors though ... westhash is the main one ... yaamp is down at the moment - but was not getting too many errors there ...

have you overclocked the cards to get those hashrates? ... if so - how? ... assuming you are using linux too as per your post ...

tanx ...

#crysx

Those hashrates are for two 750Ti's.. sorry if I wasn't clear about that!
10.5M/2 = 5250 khash/s effective at the pool (YAAMP) for quark. I get about 5550 +/- 50 khash/s reported from the miner itself.

[sp_]

I was only getting about 8 Mh/s effective hashrate at the pool with r24 as opposed to 10.5 Mh/s with r23.  I'll be sticking with r23 for now.

those hashrates are superb ... but as sp works on the miner and optimizations - there is bound to be some mishaps along the way ...

im still runing the latest git pull - and will stay on it for now - to test stability also ...

i too have issues with extranonce2 errors though ... westhash is the main one ... yaamp is down at the moment - but was not getting too many errors there ...

have you overclocked the cards to get those hashrates? ... if so - how? ... assuming you are using linux too as per your post ...

tanx ...

#crysx

Those hashrates are for two 750Ti's.. sorry if I wasn't clear about that!
10.5M/2 = 5250 khash/s effective at the pool (YAAMP) for quark. I get about 5550 +/- 50 khash/s reported from the miner itself.

Yaamp takes 3.9% in fee and the rest is rejects or?

[squadbox]

I've had very poor performance with r24 built from git on linux.  Lots of low difficulty shares and nonce issues concerning already being sent.

[2015-01-05 00:20:18] nonce 6537ae00 was already sent 2 seconds ago
[2015-01-05 00:20:26] nonce 47dd5f01 was already sent 2 seconds ago
I was only getting about 8 Mh/s effective hashrate at the pool with r24 as opposed to 10.5 Mh/s with r23.  I'll be sticking with r23 for now.

Some days ago I added 2 nonce sending in quark, but it had some bugs (double sending) and I reverted the changeset.
Make sure that you build from the github version that is tagged with the "release 24" changeset. Commits before the tagged release could have sideeffects. And commits after are not tested properly and could have errors.
Yesterday I fixed a bug when running benchmarkmode (Run the benchmark for a while and it starts to print 100 times as much text to the commandline). Seems to work, but untested on the pool.

I was building from the most recent commit (the benchmark fix), not the tagged one.  I'll checkout the proper one and report back.

[sp_]

I tested on

http://suchpool.pw/drk/


got this:

[2015-01-05 21:09:04] GPU #0: GeForce GTX 750 Ti, 2769 kH/s
[2015-01-05 21:09:04] GPU #2: GeForce GTX 750, 2283 kH/s
[2015-01-05 21:09:06] GPU #0: GeForce GTX 750 Ti, 2758 kH/s
[2015-01-05 21:09:06] accepted: 58/59 (98.31%), 5735 khash/s yay!!!
[2015-01-05 21:09:08] GPU #0: GeForce GTX 750 Ti, 2790 kH/s
[2015-01-05 21:09:08] nonce b7dd4e00 was already sent 2 seconds ago
[2015-01-05 21:09:08] GPU #2: GeForce GTX 750, 2274 kH/s
[2015-01-05 21:09:08] GPU #1: GeForce GTX 650, 673.33 kH/s
[2015-01-05 21:09:08] GPU #0: GeForce GTX 750 Ti, 2825 kH/s
[2015-01-05 21:09:10] GPU #0: GeForce GTX 750 Ti, 2808 kH/s
[2015-01-05 21:09:10] nonce b7dd4e00 was already sent 4 seconds ago
[2015-01-05 21:09:10] GPU #2: GeForce GTX 750, 2276 kH/s
[2015-01-05 21:09:10] nonce b7dd4e00 was already sent 4 seconds ago
[2015-01-05 21:09:11] GPU #1: GeForce GTX 650, 673.97 kH/s
[2015-01-05 21:09:12] GPU #1: GeForce GTX 650, 674.20 kH/s
[2015-01-05 21:09:13] accepted: 59/60 (98.33%), 5736 khash/s yay!!!
[2015-01-05 21:09:13] stratum connection reset
[2015-01-05 21:09:13] GPU #0: GeForce GTX 750 Ti, 2810 kH/s
[2015-01-05 21:09:13] GPU #2: GeForce GTX 750, 2279 kH/s
[2015-01-05 21:09:13] Stratum difficulty set to 0.024
[2015-01-05 21:09:14] stratum1.suchpool.pw:3335 x11 block 198222

I think it's because of the benchmark fix. Will try without it.

(Submittet to github. The benchmarkfix will only be run in benchmarkmode)

[squadbox]

I was only getting about 8 Mh/s effective hashrate at the pool with r24 as opposed to 10.5 Mh/s with r23.  I'll be sticking with r23 for now.

those hashrates are superb ... but as sp works on the miner and optimizations - there is bound to be some mishaps along the way ...

im still runing the latest git pull - and will stay on it for now - to test stability also ...

i too have issues with extranonce2 errors though ... westhash is the main one ... yaamp is down at the moment - but was not getting too many errors there ...

have you overclocked the cards to get those hashrates? ... if so - how? ... assuming you are using linux too as per your post ...

tanx ...

#crysx

Those hashrates are for two 750Ti's.. sorry if I wasn't clear about that!
10.5M/2 = 5250 khash/s effective at the pool (YAAMP) for quark. I get about 5550 +/- 50 khash/s reported from the miner itself.

Yaamp takes 3.9% in fee and the rest is rejects or?

I say effective hashrate as opposed to "actual" hashrate.  Rejects are low.  It's just a matter of how hashrates are calculated by the pool.  Multipool explains it well: "The hashrate your miner displays is its own internal hashrate for calculating difficulty 1 shares, and thus is fairly meaningless to the pool. The pool displays an average hashrate calculated based on your valid, accepted shares per minute."

5250/5550 = ~95% which isn't bad considering it's connected through a stratum proxy for leasing.

[sp_]

I say effective hashrate as opposed to "actual" hashrate.  Rejects are low.  It's just a matter of how hashrates are calculated by the pool.  Multipool explains it well: "The hashrate your miner displays is its own internal hashrate for calculating difficulty 1 shares, and thus is fairly meaningless to the pool. The pool displays an average hashrate calculated based on your valid, accepted shares per minute."
5250/5550 = ~95% which isn't bad considering it's connected through a stratum proxy for leasing.

I would say that is acceptable. If the network connection is slow on the pool, the numbers get lower. And you get some Job not found rejects.
I found a problem in the code now when the hashcounter sum get's higher than 2^32. for a 16000MHASH rig, the rig will hash fine for 5 mins, and then might run into trouble. Testing now at suchpool.pw.

[sp_]

I have reproduced the missing hashrate now on suchpool.pw:

[2015-01-05 22:09:40] accepted: 99/101 (98.02%), 5719 khash/s yay!!!
[2015-01-05 22:09:44] GPU #1 Found nounce 55d0b2a6
[2015-01-05 22:09:44] GPU #1: GeForce GTX 650, 673.24 kH/s
[2015-01-05 22:09:45] accepted: 100/102 (98.04%), 5719 khash/s yay!!!
[2015-01-05 22:10:00] GPU #2 Found nounce ad4ec4cb
[2015-01-05 22:10:00] GPU #2: GeForce GTX 750, 2275 kH/s
[2015-01-05 22:10:00] accepted: 101/103 (98.06%), 5719 khash/s yay!!!
[2015-01-05 22:10:01] GPU #0 Found nounce 04da9b14
[2015-01-05 22:10:01] GPU #0: GeForce GTX 750 Ti, 2780 kH/s
[2015-01-05 22:10:01] accepted: 102/104 (98.08%), 5719 khash/s yay!!!
[2015-01-05 22:10:03] GPU #2 Found nounce aaf7b02e
[2015-01-05 22:10:03] GPU #2: GeForce GTX 750, 2275 kH/s
[2015-01-05 22:10:03] accepted: 103/105 (98.10%), 5719 khash/s yay!!!
[2015-01-05 22:10:06] GPU #0 Found nounce 00cf770e
[2015-01-05 22:10:06] GPU #0: GeForce GTX 750 Ti, 2791 kH/s
[2015-01-05 22:10:06] accepted: 104/106 (98.11%), 5719 khash/s yay!!!
[2015-01-05 22:10:09] GPU #2 Found nounce ab665fea
[2015-01-05 22:10:09] GPU #2: GeForce GTX 750, 2264 kH/s
[2015-01-05 22:10:09] accepted: 105/107 (98.13%), 5719 khash/s yay!!!
[2015-01-05 22:10:42] GPU #0 Found nounce 05b997fb
[2015-01-05 22:10:42] GPU #0: GeForce GTX 750 Ti, 2776 kH/s
[2015-01-05 22:10:44] GPU #1: GeForce GTX 650, 672.34 kH/s
[2015-01-05 22:11:09] GPU #2: GeForce GTX 750, 2273 kH/s
[2015-01-05 22:11:19] GPU #0 Found nounce 06038a08
[2015-01-05 22:11:19] GPU #0: GeForce GTX 750 Ti, 2767 kH/s
[2015-01-05 22:11:21] accepted: 106/108 (98.15%), 5719 khash/s yay!!!
[2015-01-05 22:11:32] GPU #0 Found nounce 0219df7e
[2015-01-05 22:11:32] GPU #0: GeForce GTX 750 Ti, 2782 kH/s
[2015-01-05 22:11:34] GPU #2 Found nounce adeb3eaa
[2015-01-05 22:11:34] GPU #2: GeForce GTX 750, 2275 kH/s
[2015-01-05 22:11:37] GPU #0 Found nounce 00b73a85
[2015-01-05 22:11:37] GPU #0: GeForce GTX 750 Ti, 2784 kH/s
[2015-01-05 22:11:37] accepted: 107/109 (98.17%), 5719 khash/s yay!!!
[2015-01-05 22:11:45] GPU #1: GeForce GTX 650, 672.44 kH/s
[2015-01-05 22:11:57] GPU #0 Found nounce 0341a810
[2015-01-05 22:11:57] GPU #0: GeForce GTX 750 Ti, 2785 kH/s
[2015-01-05 22:11:58] accepted: 108/110 (98.18%), 5720 khash/s yay!!!
[2015-01-05 22:11:59] GPU #2 Found nounce adf4208b
[2015-01-05 22:11:59] GPU #2: GeForce GTX 750, 2273 kH/s
[2015-01-05 22:12:00] accepted: 109/111 (98.20%), 5720 khash/s yay!!!


Miner reports: 13 found
Pool: 11 accepted

no blockchange
no rejects.

shares are lost..

edit:

sometimes alot of shares are lost in a short period of time:

[2015-01-05 22:17:32] accepted: 136/139 (97.84%), 5722 khash/s yay!!!
[2015-01-05 22:17:37] GPU #2 Found nounce ab5e7b0f
[2015-01-05 22:17:37] GPU #2: GeForce GTX 750, 2277 kH/s
[2015-01-05 22:17:37] GPU #1 Found nounce 571569c9
[2015-01-05 22:17:37] GPU #1: GeForce GTX 650, 672.91 kH/s
[2015-01-05 22:17:55] GPU #0 Found nounce 07ff3047
[2015-01-05 22:17:55] GPU #0: GeForce GTX 750 Ti, 2781 kH/s
[2015-01-05 22:18:31] GPU #1 Found nounce 576de9ee
[2015-01-05 22:18:31] GPU #1: GeForce GTX 650, 673.20 kH/s
[2015-01-05 22:18:37] GPU #2: GeForce GTX 750, 2277 kH/s
[2015-01-05 22:18:40] GPU #0 Found nounce 07443c33
[2015-01-05 22:18:40] GPU #0: GeForce GTX 750 Ti, 2782 kH/s
[2015-01-05 22:18:56] GPU #0 Found nounce 026e9734
[2015-01-05 22:18:56] GPU #0: GeForce GTX 750 Ti, 2784 kH/s
[2015-01-05 22:18:56] accepted: 137/140 (97.86%), 5724 khash/s yay!!!
[2015-01-05 22:19:02] GPU #1 Found nounce 5693acef
[2015-01-05 22:19:02] GPU #1: GeForce GTX 650, 673.05 kH/s
[2015-01-05 22:19:02] accepted: 138/141 (97.87%), 5724 khash/s yay!!!

[Schleicher]

I have reproduced the missing hashrate now on suchpool.pw:

Yeah. With multipool.us as well:

Code:

[2015-01-05 23:37:35] GPU #0 nonce count = 28
[2015-01-05 23:37:35] GPU #0: GeForce GTX 970, 7103 kH/s
[2015-01-05 23:37:35] accepted: 28/28 (100.00%), 7099 khash/s yay!!!
[2015-01-05 23:37:37] GPU #0 nonce count = 29  
[2015-01-05 23:37:37] GPU #0: GeForce GTX 970, 7061 kH/s
[2015-01-05 23:37:39] GPU #0 nonce count = 30
[2015-01-05 23:37:39] GPU #0: GeForce GTX 970, 7021 kH/s
[2015-01-05 23:37:39] accepted: 29/29 (100.00%), 7095 khash/s yay!!!
[2015-01-05 23:37:41] GPU #0 nonce count = 31
[2015-01-05 23:37:41] GPU #0: GeForce GTX 970, 7020 kH/s
[2015-01-05 23:37:41] accepted: 30/30 (100.00%), 7092 khash/s yay!!!

[sp_]

fixed it now

No GPU codechanges in this version from build 24

- fixed command output in benchmark mode
- fixed hashrate on some pools. coinmine.pl, suchpool.pw etc.

The 25th public beta(windows executable) is available here: (05-jan-2015)

http://www.filedropper.com/release25

The sourcecode is available here:

https://github.com/sp-hash/ccminer

[chrysophylax]

With BTC price = trash, nothing is profitable anymore.

This is correct

its THIS that keeps it dead ...

there was a time when 'the market' ( ie 'the people' ) actually believed that btc was its OWN monetary unit and NOT a commodity for the government controlled fiat currencies ...

a time when btc was its OWN market - its OWN system - its OWN way of dealing selling buying and profiting ...

oh - hang on ... that time is still nigh ...

as i posted on @cryptobucket ...

no bloody wonder satoshi went underground ...

people have just 'lost' the whole ideal and intention that btc was built in the first place ...

ho hum ... the governments have won ...

for now ...

#crysx

[chrysophylax]

Above 0,2 BTC per day for 1 gigahash.

you only need 60 980 cards (10 rigs), or 180 750TI Cards(30 rigs) to reach 1 gigahash.

$)

ok - 30 rigs - here we come ...

#crysx

[chrysophylax]

I was only getting about 8 Mh/s effective hashrate at the pool with r24 as opposed to 10.5 Mh/s with r23.  I'll be sticking with r23 for now.

those hashrates are superb ... but as sp works on the miner and optimizations - there is bound to be some mishaps along the way ...

im still runing the latest git pull - and will stay on it for now - to test stability also ...

i too have issues with extranonce2 errors though ... westhash is the main one ... yaamp is down at the moment - but was not getting too many errors there ...

have you overclocked the cards to get those hashrates? ... if so - how? ... assuming you are using linux too as per your post ...

tanx ...

#crysx

Those hashrates are for two 750Ti's.. sorry if I wasn't clear about that!
10.5M/2 = 5250 khash/s effective at the pool (YAAMP) for quark. I get about 5550 +/- 50 khash/s reported from the miner itself.

still very nice ...

sp - you are doing an awesome job mate ...

#crysx

[chrysophylax]

I say effective hashrate as opposed to "actual" hashrate.  Rejects are low.  It's just a matter of how hashrates are calculated by the pool.  Multipool explains it well: "The hashrate your miner displays is its own internal hashrate for calculating difficulty 1 shares, and thus is fairly meaningless to the pool. The pool displays an average hashrate calculated based on your valid, accepted shares per minute."
5250/5550 = ~95% which isn't bad considering it's connected through a stratum proxy for leasing.

I would say that is acceptable. If the network connection is slow on the pool, the numbers get lower. And you get some Job not found rejects.
I found a problem in the code now when the hashcounter sum get's higher than 2^32. for a 16000MHASH rig, the rig will hash fine for 5 mins, and then might run into trouble. Testing now at suchpool.pw.

agreed - acceptable in my eyes also ...

the issue doesnt seem to be a problem here ... maybe im not connected long enough to notice it for any period of time? ...

#crysx