CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 1125.

fenomenhaa

sr. member

Activity: 248

Merit: 250

Quote from: sp_ on April 24, 2015, 03:23:54 AM

Remember to send some bits to:

KlausT @github:
BTC 1H2BHSyuwLP9vqt2p3bK9G3mDJsAi7qChw
DRK XcM9FXrvZS275pGyGmfJmS98tHPZ1rjErM

Since he helped to fix the broken hash.

will wait your new relase ))). İ'm so lazy to rebuilt source code ))) .Thanks for your sharing and for your work Sp ))

chrysophylax

legendary

Activity: 2912

Merit: 1091

--- ChainWorks Industries ---

Quote from: sp_ on April 24, 2015, 03:22:08 AM

The hashrates fluxate but it seems to be doing bether than 11MHASH on the pool so far. 12-15MHASH

the -g parameter and the latest optimalizations seems to give 20-30% more quark (verifed on the pool). Let's run it for a while and check stabilty

it does look that way - yes ...

if there is more optimization that can still be done - we will compile as soon as we possibly can and test ...

there will be a weekend where the whole farm will be pointed to both your btc addresses and mining for you both - as we usually do to donate ...

quark will be the algo - since that is the one you are working on at the moment and its probably one of the most 'profitable' ...

i think ill go to the office tonight and reset the router - its probably the issue here - so that we can continue testing ...

tanx sp ...

#crysx

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Remember to send some bits to:

KlausT @github:
BTC 1H2BHSyuwLP9vqt2p3bK9G3mDJsAi7qChw
DRK XcM9FXrvZS275pGyGmfJmS98tHPZ1rjErM

Since he helped to fix the broken hash.

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

The hashrates fluxate but it seems to be doing bether than 11MHASH on the pool so far. 12-15MHASH

the -g parameter and the latest optimalizations seems to give 20-30% more quark (verifed on the pool). Let's run it for a while and check stabilty

chrysophylax

legendary

Activity: 2912

Merit: 1091

--- ChainWorks Industries ---

Quote from: sp_ on April 24, 2015, 02:45:43 AM

-g 4 -i 23 gave higher rate, but crashed the rig after a while (2x750ti)

How is the power consumption, is it up with more gpu threads?

-g 5 is also working, but a bit slower.

im not too sure at the moment mate ...

ive lost remote ( ssh ) access for some reason and because im not there - i cant read the power meter ...

cant wait till i can do this full time - and fix all these little shitty issues

...

ill let you know when i can either 1 - get access to the farm or 2 - physically go there tonight and fix this issue ...

the test machine is on westhash at the moment if you want to have a look at -g3 -i 22 parameters with quark - two gisgabyte 750ti oc lp - stock clocks ...

seems stable enough ...

https://www.westhash.com/index.jsp?p=miners&a=12&d=1&addr=15umzHXF8NzXA4FywmeFbrDHgL8WcPs3wx ( thats our general / common btc address ) ...

#crysx

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

-g 4 -i 23 gave higher rate, but crashed the rig after a while (2x750ti)

How is the power consumption, is it up with more gpu threads?

-g 5 is also working, but a bit slower.

chrysophylax

legendary

Activity: 2912

Merit: 1091

--- ChainWorks Industries ---

Quote from: chrysophylax on April 24, 2015, 01:17:07 AM

Quote from: sp_ on April 24, 2015, 01:15:37 AM

Fixed now . The new blake was missing a cpu_init call

tanx sp ...

new clone then compile ...

will let you know how it goes in a few minutes ...

#crysx

all compiled and seems good so far sp ...

testing x11 at the moment - but will test on quark later today and see how it goes ...

tanx ...

edit ...

testing quark now Wink

... seems to be stable so far and a higher hashrate ...

-g 3 -i 22 on gigabyte 750ti oc lp ... stock clocks ... westhash ...

[2015-04-24 16:49:49] GPU #1: GeForce GTX 750 Ti, 5395
[2015-04-24 16:49:49] GPU #1: GeForce GTX 750 Ti, 6080
[2015-04-24 16:49:49] GPU #0: GeForce GTX 750 Ti, 6641
[2015-04-24 16:49:50] GPU #1: GeForce GTX 750 Ti, 5518
[2015-04-24 16:49:50] GPU #0: GeForce GTX 750 Ti, 6209
[2015-04-24 16:49:50] GPU #0: GeForce GTX 750 Ti, 5486
[2015-04-24 16:49:52] GPU #1: GeForce GTX 750 Ti, 5951
[2015-04-24 16:49:52] accepted: 24/24 (100.00%), 11444 khash/s yay!!!
[2015-04-24 16:49:55] stratum.westhash.com:3345/ quark block 1563963

#crysx

totoy

hero member

Activity: 623

Merit: 500

Quote from: jjjordan on April 23, 2015, 09:36:45 PM

Yeah, I definitely won't be able to do it with these instructions... Wink

If anybody has step by step ones I'll try.

here http://cryptomining-blog.com/tag/compile-cudaminer-for-windows/

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Quote from: scryptr on April 23, 2015, 07:18:41 PM

INITIAL RESULTS--

I cloned git commit #725 and ran it on my 2 rigs. THE "...does not validate on CPU" ERRORS are GONE! The hash rate still fluctuates. Hash rate is printed to the screen only with the physical GPU number. Using only "-g 4" and default intensity (no -i flag used), the 8 threads on my 2-card rig are reported at launch, as are the 24 threads on my 6 card rig. Then the hash rate fluctuates to higher than normal values, but without error. My 2-card 970 FTW+ rig is averaging near 35Mh/s, and my 6-card 750ti FTW rig is averaging around 60Mh/s. Both rig's hash rates swing wildly.

My total rate for all three machines mining on yaamp was about 80Mh/s before, mining with single threads as reported on the previous page. This was reflected in the pool rate. The pool rate still remains at the same level with commit #725, but the errors are gone. The local rates total over 100Mh/s, including a steady, single-threaded, rate for my 960 SSC of about 9.4Mh/s. The 960 SSC is somewhat like a control sample, I suppose. --scryptr

The stats have been corrected now, the miner should display around 70-80Mh/s (wich is the rate you get at the pool). Perhaps a bit more if you play with the -g and the -i parameter.

chrysophylax

legendary

Activity: 2912

Merit: 1091

--- ChainWorks Industries ---

Quote from: sp_ on April 24, 2015, 01:15:37 AM

Fixed now . The new blake was missing a cpu_init call

tanx sp ...

new clone then compile ...

will let you know how it goes in a few minutes ...

#crysx

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Fixed now . The new blake was missing a cpu_init call (x13,x15,x17,nist5 should work)

chrysophylax

legendary

Activity: 2912

Merit: 1091

--- ChainWorks Industries ---

Quote from: sp_ on April 24, 2015, 12:35:36 AM

Quote from: chrysophylax on April 24, 2015, 12:24:33 AM

Quote from: sp_ on April 23, 2015, 11:57:08 PM

quark. the others are not tested

just compiled and tested x11 - miner crashed on start ...
reverted back to previous version - all ok ...
#crysx

-i 22 is too high for x11 on the 750ti

no other parameters but -i 20.5 ... which is fine for the 750ti oc lp ...

the test machine cannot use -g parameter at all for x11 ... it crashes ccminer on startup ...

-g3 -i 22 is fine for quark - but wild hashrate fluctuations still ...

edit ...

this is the error sp ...

-------

Cuda error in func 'quark_blake512_cpu_setBlock_80' at line 476 : invalid argument.
[2015-04-24 15:20:24] GPU #0: cuda_check_cpu_init driver shutting down

-------

#crysx

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Quote from: chrysophylax on April 24, 2015, 12:24:33 AM

Quote from: sp_ on April 23, 2015, 11:57:08 PM

quark. the others are not tested

just compiled and tested x11 - miner crashed on start ...
reverted back to previous version - all ok ...
#crysx

-i 22 is too high for x11 on the 750ti

chrysophylax

legendary

Activity: 2912

Merit: 1091

--- ChainWorks Industries ---

Quote from: sp_ on April 23, 2015, 11:57:08 PM

quark. the others are not tested

just compiled and tested x11 - miner crashed on start ...

reverted back to previous version - all ok ...

#crysx

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Submitted bether launch congfig/intensity for compute 5.2 devices. (quark/blake)

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

quark. the others are not tested

chrysophylax

legendary

Activity: 2912

Merit: 1091

--- ChainWorks Industries ---

Quote from: sp_ on April 23, 2015, 11:44:04 PM

Just submitted 2 checkins merged from the klaus_t fork

-faster blake
-fixed stats

Seems to run ok with -i 22 -g 2

those parameters under any algo - or specific ones sp? ...

#crysx

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Just submitted 2 checkins merged from the klaus_t fork

-faster blake
-fixed stats

Seems to run ok with -i 22 -g 2

AliMan

hero member

Activity: 1974

Merit: 502

Vave.com - Crypto Casino

I'm unable to run either a single card or four cards on the -g 4 flag as it keeps crashing on the latest update.

Schleicher

hero member

Activity: 675

Merit: 514

Quote from: jjjordan on April 23, 2015, 08:52:52 PM

Do you guys know where I can find instructions to compile from github?
I have Win7 x64 on my PC. Also I have never done this (compiling) and don't have a github account.
It would be nice to try these versions between releases.

Install Visual Studio 2013
Install Nvidia CUDA 6.5 Toolkit
Download the ccminer source code
Unzip it
Open the ccminer.sln file
Make sure the compiler is set to "release" and "Win32"
Rebuild solution

Then you should see the ccminer.exe in the release folder

Topic: CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 1125. (Read 2347601 times)