CCminer(SP-MOD) Modded GPU kernels.

[sp_]

Spreadcoin version 9 has been sendt to the donators.

happy mining

[zjy]

Spreadcoin version 9 has been sendt to the donators.

happy mining

hey SP  SpreadCoin profit, 750ti? Can anyone donate 0.1BTC  to get the optimized version?

[sp_]

Spreadcoin can sometimes be very profitable. Anyone can donate and recieve the optimized miner.

here is a screenshot with overclocking:

[sp_]

submitted another optimalization in bmw512.
Visible improvement in the the quark algo the gtx960 is up 50-100khash

[antonio8]

Spreadcoin can sometimes be very profitable. Anyone can donate and recieve the optimized miner.

here is a screenshot with overclocking:

What settings are you running and OC?

[Grout]

I donated a beer last week and you didn't even thank me :'-(
My feelings are so hurt that I will keep donating one day's worth of mining every two weeks ;-)
If you want your 7% to represent more beers, increase my hashrate :-p
Keep up the good job SP...

Thanks, you have donated multiple times

Note that if you reach 0.1BTC I will send you the optimized spreadcoin miner with full sourcecode (linux compatible)

A new release spreadcoinminer version 9 is soon ready for the donators.

According to my log, I gave 0.074 so far. And I missed a few weeks. That's what you get for not including an optional donation percentage in your miner

I might reach 0.1 in about a month. Do you have other private miners that you give at higher thresholds?

[rednoW]

submitted another optimalization in bmw512.
Visible improvement in the the quark algo the gtx960 is up 50-100khash

+10-15 khash with gtx750 )))

[lawrencelyl]

submitted another optimalization in bmw512.
Visible improvement in the the quark algo the gtx960 is up 50-100khash

Compiled and observed about 100++kH/s speed increase for gtx980

[chrysophylax]

I donated a beer last week and you didn't even thank me :'-(
My feelings are so hurt that I will keep donating one day's worth of mining every two weeks ;-)
If you want your 7% to represent more beers, increase my hashrate :-p
Keep up the good job SP...

Thanks, you have donated multiple times

Note that if you reach 0.1BTC I will send you the optimized spreadcoin miner with full sourcecode (linux compatible)

A new release spreadcoinminer version 9 is soon ready for the donators.

According to my log, I gave 0.074 so far. And I missed a few weeks. That's what you get for not including an optional donation percentage in your miner

I might reach 0.1 in about a month. Do you have other private miners that you give at higher thresholds?

our donation links are in the final stages of testing - and will be ready to put on permanently ( or at least as permanent as we can afford to - hehehe - which should be a long time ) ...

we have just come across a couple of issue with three of our servers - and are waiting for the feedback from our provider as to what is actually happening with it all ... when we do and can rectify the issues - we will publish all the system details ...

being the weekend though - i would not expect very much to happen for another few days ...

on another note - the spreadminer that sp had originally sent out was good - the next hashed better - this one should be a better one again ...

being solo mining is a pain though - but with the the hashrate the way it is - it shouldnt be too difficult to hit blocks if you are patient ...

#crysx

[chrysophylax]

submitted another optimalization in bmw512.
Visible improvement in the the quark algo the gtx960 is up 50-100khash

+10-15 khash with gtx750 )))

stock? ...

or you overclock also? ...

#crysx

[pallas]

This little patch is giving quark +300 Kh/s on my 970:

diff --git a/groestl_functions_quad.cu b/groestl_functions_quad.cu
index c39e81d..3ac3c5d 100644
--- a/groestl_functions_quad.cu
+++ b/groestl_functions_quad.cu
@@ -285,7 +285,7 @@ __device__ __forceinline__ void G256_MixFunction_quad(uint32_t *r)

 __device__ __forceinline__ void groestl512_perm_P_quad(uint32_t *const r)
 {
-
+#pragma unroll
        for(int round=0;round<14;round++)
     {
         G256_AddRoundConstantP_quad(r[7], r[6], r[5], r[4], r[3], r[2], r[1], r[0], round);

[chrysophylax]

This little patch is giving quark +300 Kh/s on my 970:

diff --git a/groestl_functions_quad.cu b/groestl_functions_quad.cu
index c39e81d..3ac3c5d 100644
--- a/groestl_functions_quad.cu
+++ b/groestl_functions_quad.cu
@@ -285,7 +285,7 @@ __device__ __forceinline__ void G256_MixFunction_quad(uint32_t *r)

 __device__ __forceinline__ void groestl512_perm_P_quad(uint32_t *const r)
 {
-
+#pragma unroll
        for(int round=0;round<14;round++)
     {
         G256_AddRoundConstantP_quad(r[7], r[6], r[5], r[4], r[3], r[2], r[1], r[0], round);

where does the patch go? ...

#crysx

[pallas]

This little patch is giving quark +300 Kh/s on my 970:

diff --git a/groestl_functions_quad.cu b/groestl_functions_quad.cu
index c39e81d..3ac3c5d 100644
--- a/groestl_functions_quad.cu
+++ b/groestl_functions_quad.cu
@@ -285,7 +285,7 @@ __device__ __forceinline__ void G256_MixFunction_quad(uint32_t *r)

 __device__ __forceinline__ void groestl512_perm_P_quad(uint32_t *const r)
 {
-
+#pragma unroll
        for(int round=0;round<14;round++)
     {
         G256_AddRoundConstantP_quad(r[7], r[6], r[5], r[4], r[3], r[2], r[1], r[0], round);

where does the patch go? ...

#crysx

"patch -p0" should work.
If you wonna do it manually, the file name is at the start of the patch.
I'm curious to see if it's good for the 750 as well.

[sp_]

I'm curious to see if it's good for the 750 as well.

no good for the 750ti. A drop of 400khash.

[pallas]

I'm curious to see if it's good for the 750 as well.

no good for the 750ti. A drop of 400khash.

I'm new to cuda: is there an ifdef or something we can put to enable the unroll for 9xx cards only?

[djm34]

I'm curious to see if it's good for the 750 as well.

no good for the 750ti. A drop of 400khash.

I'm new to cuda: is there an ifdef or something we can put to enable the unroll for 9xx cards only?

#pragma unroll n (and #pragma nounroll )

[sp_]

   #if __CUDA_ARCH__ > 500
   #pragma unroll
   #endif   

[rednoW]

stock? ...
or you overclock also? ...
#crysx

all me results are for max possible overclock for the algo.
For quark it is 1510/1600
6114 khash new record

[go6ooo1212]

Just did a small donation to SP_ , TxID:
58b1cd0e556f708bd6da1a7d5cad19b3a7dbca88b75d39ce63ccc4b921f7fb9f

[sp_]

Thanks for the support guys.

I found another improvement:

   x8 = ROTL32(0x4D42C787, 7);

compiles to:

   // inline asm
   mov.u32    %r4599, 1296222087;
   // inline asm
   shf.l.wrap.b32 %r4597, %r4599, %r4599, %r4756;

stupid compiler.

x11 improvement coming soon. (luffacubehash512)