CCminer(SP-MOD) Modded GPU kernels.

[scryptr]

NOT PAID IN FULL--

@bensam--  The code was released without the Donation Goal being met.  The goal was 5 BTC, and the Neoscrypt account received about 2 BTC.  DJM34 released the code because he felt like it.

You owe a small contribution if you use the code.       --scryptr

[djm34]

NOT PAID IN FULL--

@bensam--  The code was released without the Donation Goal being met.  The goal was 5 BTC, and the Neoscrypt account received about 2 BTC.  DJM34 released the code because he felt like it.

You owe a small contribution if you use the code.       --scryptr

well he doesn't owe me anything, it is just that he criticizes a lot the way we do things, ask us for new stuff and he is the first to not contribute when we ask for contribution finding other dodgy argument to say that what we do is crap... (then ultimately using them...)

[sp_]

I submitted an x11 speedup. small AES/echo change.
Now the cuda 7.5 version is faster than cuda 6.5 on the 750ti. (+15khash)
on the
gtx 970 it is still 250khash slower(than release 74)
gtx 960 20khash slower.(than release 74)

[sp_]

My GTX 960 gets 340kh/s mining Neoscrypt with the latest NiceHash Miner, that makes my GTX 960 as fast as an AMD 280X with the latest NiceHash Miner.  They are still using your release dot 74.  If you did a hybrid compile (Pallas), NiceHash may incorporate your release dot 77+.  Neoscrypt is still a top-paying algo at NiceHash, and AMD code is really being optimized there.
I may do some benchmark tests.       --scryptr

I have done some tests. Some kernals are faster with cuda 6.5 and some are faster in 7.5. With a hybrid approach we get more hash.

I will create a hybrid projectfile for release 78 (windows only)
Small boost in all algos as usual

[pallas]

My GTX 960 gets 340kh/s mining Neoscrypt with the latest NiceHash Miner, that makes my GTX 960 as fast as an AMD 280X with the latest NiceHash Miner.  They are still using your release dot 74.  If you did a hybrid compile (Pallas), NiceHash may incorporate your release dot 77+.  Neoscrypt is still a top-paying algo at NiceHash, and AMD code is really being optimized there.
I may do some benchmark tests.       --scryptr

I have done some tests. Some kernals are faster with cuda 6.5 and some are faster in 7.5. With a hybrid approach we get more hash.

I will create a hybrid projectfile for release 78 (windows only)
Small boost in all algos as usual

Thanks :-)
I may have a look at the hybrid Makefile after the holidays, in case nobody does it before.

[sp_]

here are some timings from the x11 algo on a 750ti:

cuda 7.5 is faster: blake, skein,shavite,simd,echo
cuda 6.5 is faster: bmw,groestl,jhkeccak,luffacubehash

Code:

release 74:

22.6765s  8.1170ms          (19968 1 1)        (64 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  quark_blake512_gpu_hash_80(unsigned int, unsigned int, uint2*) [3277]
22.6846s  8.8407ms*          (39936 1 1)        (32 1 1)       128        0B        0B         -           -  GeForce GTX 750         2        14  quark_bmw512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [3283]
22.6935s  83.971ms*           (9984 1 1)       (512 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  quark_groestl512_gpu_hash_64_quad(unsigned int, unsigned int, unsigned int*, unsigned int const *) [3289]
22.7775s  10.594ms           (9984 1 1)       (128 1 1)        40        0B        0B         -           -  GeForce GTX 750         2        14  quark_skein512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [3295]
22.7881s  34.017ms*          (4992 1 1)       (256 1 1)        80        0B        0B         -           -  GeForce GTX 750         2        14  quark_jh512Keccak512_gpu_hash_64(unsigned int, unsigned int, unsigned int*) [3300]
22.8221s  59.762ms*          (4992 1 1)       (256 1 1)        71        0B        0B         -           -  GeForce GTX 750         2        14  x11_luffaCubehash512_gpu_hash_64(unsigned int, unsigned int, __int64*) [3305]
22.8819s  57.118ms           (3994 1 1)       (320 1 1)        64  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_shavite512_gpu_hash_64(unsigned int, unsigned int, __int64*) [3310]
22.9390s  55.778ms         (319488 1 1)        (32 1 1)        62        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_expand_64(unsigned int, unsigned int, __int64 const *, uint4*) [3316]
22.9948s  29.153ms          (39936 1 1)        (32 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_compress_64_maxwell(unsigned int, unsigned int, __int64*, uint4*) [3322]
23.0239s  93.088ms           (4992 1 1)       (256 1 1)        80  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_echo512_gpu_hash_64_final(unsigned int, unsigned int, __int64 const *, unsigned int*, unsigned int) [3330]

release 77:

3.83318s  7.1578ms*          (19968 1 1)        (64 1 1)        80        0B        0B         -           -  GeForce GTX 750         2        14  quark_blake512_gpu_hash_80(unsigned int, unsigned int, uint2*) [736]
3.84034s  9.1572ms          (39936 1 1)        (32 1 1)       146        0B        0B         -           -  GeForce GTX 750         2        14  quark_bmw512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [742]
3.84951s  84.421ms           (9984 1 1)       (512 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  quark_groestl512_gpu_hash_64_quad(unsigned int, unsigned int, unsigned int*, unsigned int const *) [748]
3.93394s  9.0119ms*           (9984 1 1)       (128 1 1)        40        0B        0B         -           -  GeForce GTX 750         2        14  quark_skein512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [754]
3.94295s  37.825ms           (4992 1 1)       (256 1 1)        79        0B        0B         -           -  GeForce GTX 750         2        14  quark_jh512Keccak512_gpu_hash_64(unsigned int, unsigned int, unsigned int*) [759]
3.98079s  64.982ms           (2496 1 1)       (512 1 1)        75        0B        0B         -           -  GeForce GTX 750         2        14  x11_luffaCubehash512_gpu_hash_64(unsigned int, unsigned int, __int64*) [764]
4.04578s  56.054ms*          (4992 1 1)       (256 1 1)        80  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_shavite512_gpu_hash_64(unsigned int, unsigned int, __int64*) [769]
4.10184s  47.404ms*         (319488 1 1)        (32 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_expand_64(unsigned int, unsigned int, __int64 const *, uint4*) [775]
4.14925s  25.791ms*          (39936 1 1)        (32 1 1)       128        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_compress_64_maxwell(unsigned int, unsigned int, __int64*, uint4*) [781]
4.17506s  91.801ms*          (4992 1 1)       (256 1 1)       128  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_echo512_gpu_hash_64_final(unsigned int, unsigned int, __int64 const *, unsigned int*, unsigned int) [789]

[chrysophylax]

aaand a massively happy new year to you all ...

2016 is going to be a huge one ...

tanx for all 'your' work - yes ALL the devs ...

cant wait to get back into it in the second week of jan ...

only a few hours left here in adelaide australia ...

its been a pleasure and an honor and a massive wakeup call with excitement to boot this year ...

2016 - here we come ...

see you ALL on the other side ...

#crysx

[sambiohazard]

Happy New Year to all of you & I hope you will evolve to be a better human being in this new year.

Regards

Sam

[Mek]

Free gpu miner for NVIDIA cards:

release 77++ (git)

980ti: 4.1GHASH
980:  3.05GHASH
970: 2,63MHASH
960:  1,71MHASH
750ti: 930MHASH

ccminer -a blakecoin

1.5.77(sp-MOD) is available here: (20-dec-2015)

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

The sourcecode is available here:

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

Hello, how do I test the hashrate for blakecoin with my card? I tried mining against my wallet (no pool) - it's running on port 21:

Code:

ccminer -a blakecoin --url=http://127.0.0.1:21 --userpass=myuser:mypass

I get this cryptic error:

Code:

[2015-12-31 17:52:14] NVAPI GPU monitoring enabled.
[2015-12-31 17:52:14] 1 miner thread started, using 'blakecoin' algorithm.
[2015-12-31 17:52:14] Binding thread 0 to cpu 0 (mask 1)
Cuda error in func 'blake256_cpu_setBlock_16' at line 703 : invalid device symbol.

My device is GTX670M in a gaming laptop with optimus.
Can you help, please?

[theLosers106]

Free gpu miner for NVIDIA cards:

release 77++ (git)

980ti: 4.1GHASH
980:  3.05GHASH
970: 2,63MHASH
960:  1,71MHASH
750ti: 930MHASH

ccminer -a blakecoin

1.5.77(sp-MOD) is available here: (20-dec-2015)

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

The sourcecode is available here:

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

Hello, how do I test the hashrate for blakecoin with my card? I tried mining against my wallet (no pool) - it's running on port 21:

Code:

ccminer -a blakecoin --url=http://127.0.0.1:21 --userpass=myuser:mypass

I get this cryptic error:

Code:

[2015-12-31 17:52:14] NVAPI GPU monitoring enabled.
[2015-12-31 17:52:14] 1 miner thread started, using 'blakecoin' algorithm.
[2015-12-31 17:52:14] Binding thread 0 to cpu 0 (mask 1)
Cuda error in func 'blake256_cpu_setBlock_16' at line 703 : invalid device symbol.

My device is GTX670M in a gaming laptop with optimus.
Can you help, please?

Your card is using Fermi based GPU so you would have to use this ccminer instead. Happened to me once 
https://github.com/tpruvot/ccminer/releases

[Mek]

Free gpu miner for NVIDIA cards:

release 77++ (git)

980ti: 4.1GHASH
980:  3.05GHASH
970: 2,63MHASH
960:  1,71MHASH
750ti: 930MHASH

ccminer -a blakecoin

1.5.77(sp-MOD) is available here: (20-dec-2015)

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

The sourcecode is available here:

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

Hello, how do I test the hashrate for blakecoin with my card? I tried mining against my wallet (no pool) - it's running on port 21:

Code:

ccminer -a blakecoin --url=http://127.0.0.1:21 --userpass=myuser:mypass

I get this cryptic error:

Code:

[2015-12-31 17:52:14] NVAPI GPU monitoring enabled.
[2015-12-31 17:52:14] 1 miner thread started, using 'blakecoin' algorithm.
[2015-12-31 17:52:14] Binding thread 0 to cpu 0 (mask 1)
Cuda error in func 'blake256_cpu_setBlock_16' at line 703 : invalid device symbol.

My device is GTX670M in a gaming laptop with optimus.
Can you help, please?

Your card is using Fermi based GPU so you would have to use this ccminer instead. Happened to me once 
https://github.com/tpruvot/ccminer/releases

Thanks, it works! I am getting around 150 MH/s (fyi).

[scryptr]

HAPPY NEW YEAR!!! --

And best wishes for the New Year! I am awaiting sp_'s newest release (dot 78) and hoping that this "New Year's Baby" does more than crawl!          --scryptr

[djm34]

HAPPY NEW YEAR!!! --

And best wishes for the New Year! I am awaiting sp_'s newest release (dot 78) and hoping that this "New Year's Baby" does more than crawl!          --scryptr

idem...
happy new year to every one 

[ldp5500]

Happy New Year

[bathrobehero]

Happy New Year!

[pallas]

here are some timings from the x11 algo on a 750ti:

cuda 7.5 is faster: blake, skein,shavite,simd,echo
cuda 6.5 is faster: bmw,groestl,jhkeccak,luffacubehash

Code:

release 74:

22.6765s  8.1170ms          (19968 1 1)        (64 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  quark_blake512_gpu_hash_80(unsigned int, unsigned int, uint2*) [3277]
22.6846s  8.8407ms*          (39936 1 1)        (32 1 1)       128        0B        0B         -           -  GeForce GTX 750         2        14  quark_bmw512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [3283]
22.6935s  83.971ms*           (9984 1 1)       (512 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  quark_groestl512_gpu_hash_64_quad(unsigned int, unsigned int, unsigned int*, unsigned int const *) [3289]
22.7775s  10.594ms           (9984 1 1)       (128 1 1)        40        0B        0B         -           -  GeForce GTX 750         2        14  quark_skein512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [3295]
22.7881s  34.017ms*          (4992 1 1)       (256 1 1)        80        0B        0B         -           -  GeForce GTX 750         2        14  quark_jh512Keccak512_gpu_hash_64(unsigned int, unsigned int, unsigned int*) [3300]
22.8221s  59.762ms*          (4992 1 1)       (256 1 1)        71        0B        0B         -           -  GeForce GTX 750         2        14  x11_luffaCubehash512_gpu_hash_64(unsigned int, unsigned int, __int64*) [3305]
22.8819s  57.118ms           (3994 1 1)       (320 1 1)        64  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_shavite512_gpu_hash_64(unsigned int, unsigned int, __int64*) [3310]
22.9390s  55.778ms         (319488 1 1)        (32 1 1)        62        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_expand_64(unsigned int, unsigned int, __int64 const *, uint4*) [3316]
22.9948s  29.153ms          (39936 1 1)        (32 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_compress_64_maxwell(unsigned int, unsigned int, __int64*, uint4*) [3322]
23.0239s  93.088ms           (4992 1 1)       (256 1 1)        80  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_echo512_gpu_hash_64_final(unsigned int, unsigned int, __int64 const *, unsigned int*, unsigned int) [3330]

release 77:

3.83318s  7.1578ms*          (19968 1 1)        (64 1 1)        80        0B        0B         -           -  GeForce GTX 750         2        14  quark_blake512_gpu_hash_80(unsigned int, unsigned int, uint2*) [736]
3.84034s  9.1572ms          (39936 1 1)        (32 1 1)       146        0B        0B         -           -  GeForce GTX 750         2        14  quark_bmw512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [742]
3.84951s  84.421ms           (9984 1 1)       (512 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  quark_groestl512_gpu_hash_64_quad(unsigned int, unsigned int, unsigned int*, unsigned int const *) [748]
3.93394s  9.0119ms*           (9984 1 1)       (128 1 1)        40        0B        0B         -           -  GeForce GTX 750         2        14  quark_skein512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [754]
3.94295s  37.825ms           (4992 1 1)       (256 1 1)        79        0B        0B         -           -  GeForce GTX 750         2        14  quark_jh512Keccak512_gpu_hash_64(unsigned int, unsigned int, unsigned int*) [759]
3.98079s  64.982ms           (2496 1 1)       (512 1 1)        75        0B        0B         -           -  GeForce GTX 750         2        14  x11_luffaCubehash512_gpu_hash_64(unsigned int, unsigned int, __int64*) [764]
4.04578s  56.054ms*          (4992 1 1)       (256 1 1)        80  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_shavite512_gpu_hash_64(unsigned int, unsigned int, __int64*) [769]
4.10184s  47.404ms*         (319488 1 1)        (32 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_expand_64(unsigned int, unsigned int, __int64 const *, uint4*) [775]
4.14925s  25.791ms*          (39936 1 1)        (32 1 1)       128        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_compress_64_maxwell(unsigned int, unsigned int, __int64*, uint4*) [781]
4.17506s  91.801ms*          (4992 1 1)       (256 1 1)       128  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_echo512_gpu_hash_64_final(unsigned int, unsigned int, __int64 const *, unsigned int*, unsigned int) [789]

I tried making a quick and dirty hybrid ccminer according to this table, to mine X11.  Unfortunately it is slower. Maybe it's because I'm on Linux...

[joblo]

here are some timings from the x11 algo on a 750ti:

cuda 7.5 is faster: blake, skein,shavite,simd,echo
cuda 6.5 is faster: bmw,groestl,jhkeccak,luffacubehash

Code:

release 74:

22.6765s  8.1170ms          (19968 1 1)        (64 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  quark_blake512_gpu_hash_80(unsigned int, unsigned int, uint2*) [3277]
22.6846s  8.8407ms*          (39936 1 1)        (32 1 1)       128        0B        0B         -           -  GeForce GTX 750         2        14  quark_bmw512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [3283]
22.6935s  83.971ms*           (9984 1 1)       (512 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  quark_groestl512_gpu_hash_64_quad(unsigned int, unsigned int, unsigned int*, unsigned int const *) [3289]
22.7775s  10.594ms           (9984 1 1)       (128 1 1)        40        0B        0B         -           -  GeForce GTX 750         2        14  quark_skein512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [3295]
22.7881s  34.017ms*          (4992 1 1)       (256 1 1)        80        0B        0B         -           -  GeForce GTX 750         2        14  quark_jh512Keccak512_gpu_hash_64(unsigned int, unsigned int, unsigned int*) [3300]
22.8221s  59.762ms*          (4992 1 1)       (256 1 1)        71        0B        0B         -           -  GeForce GTX 750         2        14  x11_luffaCubehash512_gpu_hash_64(unsigned int, unsigned int, __int64*) [3305]
22.8819s  57.118ms           (3994 1 1)       (320 1 1)        64  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_shavite512_gpu_hash_64(unsigned int, unsigned int, __int64*) [3310]
22.9390s  55.778ms         (319488 1 1)        (32 1 1)        62        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_expand_64(unsigned int, unsigned int, __int64 const *, uint4*) [3316]
22.9948s  29.153ms          (39936 1 1)        (32 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_compress_64_maxwell(unsigned int, unsigned int, __int64*, uint4*) [3322]
23.0239s  93.088ms           (4992 1 1)       (256 1 1)        80  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_echo512_gpu_hash_64_final(unsigned int, unsigned int, __int64 const *, unsigned int*, unsigned int) [3330]

release 77:

3.83318s  7.1578ms*          (19968 1 1)        (64 1 1)        80        0B        0B         -           -  GeForce GTX 750         2        14  quark_blake512_gpu_hash_80(unsigned int, unsigned int, uint2*) [736]
3.84034s  9.1572ms          (39936 1 1)        (32 1 1)       146        0B        0B         -           -  GeForce GTX 750         2        14  quark_bmw512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [742]
3.84951s  84.421ms           (9984 1 1)       (512 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  quark_groestl512_gpu_hash_64_quad(unsigned int, unsigned int, unsigned int*, unsigned int const *) [748]
3.93394s  9.0119ms*           (9984 1 1)       (128 1 1)        40        0B        0B         -           -  GeForce GTX 750         2        14  quark_skein512_gpu_hash_64(unsigned int, unsigned int, __int64*, unsigned int const *) [754]
3.94295s  37.825ms           (4992 1 1)       (256 1 1)        79        0B        0B         -           -  GeForce GTX 750         2        14  quark_jh512Keccak512_gpu_hash_64(unsigned int, unsigned int, unsigned int*) [759]
3.98079s  64.982ms           (2496 1 1)       (512 1 1)        75        0B        0B         -           -  GeForce GTX 750         2        14  x11_luffaCubehash512_gpu_hash_64(unsigned int, unsigned int, __int64*) [764]
4.04578s  56.054ms*          (4992 1 1)       (256 1 1)        80  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_shavite512_gpu_hash_64(unsigned int, unsigned int, __int64*) [769]
4.10184s  47.404ms*         (319488 1 1)        (32 1 1)        64        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_expand_64(unsigned int, unsigned int, __int64 const *, uint4*) [775]
4.14925s  25.791ms*          (39936 1 1)        (32 1 1)       128        0B        0B         -           -  GeForce GTX 750         2        14  x11_simd512_gpu_compress_64_maxwell(unsigned int, unsigned int, __int64*, uint4*) [781]
4.17506s  91.801ms*          (4992 1 1)       (256 1 1)       128  4.0000KB        0B         -           -  GeForce GTX 750         2        14  x11_echo512_gpu_hash_64_final(unsigned int, unsigned int, __int64 const *, unsigned int*, unsigned int) [789]

I tried making a quick and dirty hybrid ccminer according to this table, to mine X11.  Unfortunately it is slower. Maybe it's because I'm on Linux...

I'm not a big fan of the hybrid option, seems simpler to manage multiple executables, but I'll throw in my 2 satoshis.
I presume you compiled x11 sub-algos with the compiler that would provide the better performance. That should, theoretically,
give the best of both worlds.

I haven't seen any difference in performance between Linux & Windows so I don't suspect that is the problem.

Did you compile in the same order? IIRC you initially compiled with 7.5 then recompiled with 6.5. This would effectively link
mostly 7.5 object files with the 6.5 linker. Perhaps you used the 7.5 linker this time and some of the performance lost is due to it.

The card might also have an impact, SP tested with a 750, a 9xx might perrform differently.

It would be interesting to profile the hybrid performance compared to SP's data above.

[sp_]

I added 0.5MHASH to the qubit algo and 20MHASH in the doom algo(10% faster). on the gtx 970 @github

[sp_]

I tried making a quick and dirty hybrid ccminer according to this table, to mine X11.  Unfortunately it is slower. Maybe it's because I'm on Linux...

I am profiling the release 74 exe file. (If you want to build you need to extract the sourcecode from the 6.5 version wich is release 74)

the command line can be something like:

nvprof ccminer74.exe -a x11 --benchmark -d 0
and compare with:
nvprof ccminer77.exe -a x11 --benchmark -d 0

[sp_]

I'm not a big fan of the hybrid option, seems simpler to manage multiple executables, but I'll throw in my 2 satoshis.

I am not a fan eighter. So I try to make small adjustments to the code to aid the cuda7.5 compiler.