CCminer(SP-MOD) Modded GPU kernels.

[sp_]

The skein sp-mod #4 is around 7% faster than alexis miner. 0.05btc donation. want a copy?

[reb0rn21]

The skein sp-mod #4 is around 7% faster than alexis miner. 0.05btc donation. want a copy?

+ more OC and power use
+ 3% more stales
+ 3% more rejects due insanely high intensity
 

[nitrobg]

The skein sp-mod #4 is around 7% faster than alexis miner. 0.05btc donation. want a copy?

Well, alexis users could just increase the voltage by 3-4 straps or increase the power limit by few % and save those 0.05 BTC. Higher hashrate means higher power draw, so both ways even out.

[gemoganization]

Looking to get a copy of SP MOD....how does one go about getting this?

[sp_]

The sp-mod's doesn't normally use more power, and 5% faster hash can be traded with 20% less power.
You can run alexis miner with 100% tdp and break your cards, or the sp-mod at the same speed with 20% less power.

[nitrobg]

I was thinking more like 75% PL alexis and 70% yours. Would you mind testing the power draw theory by running GPU-Z logging for both tests?

I've used your leaked phi #2 and while it gives slightly higher hashrate compared to anxmod, the temperatures and power draw are higher as well. What's my incentive in paying 0.05-0.1 BTC when I can just increase the clocks of my cards? Not trolling, I'm genuinely curious.

[ivakar]

https://github.com/Enemy-Mine/ccminer/releases/tag/v1.03-RVN

Hope nobody paid sp_ for his "fastest x17 and x16r for big farms only" shit ))

I'm sorry, I've decided to download and check your miner, but unfortunately there is no such user on GitHub
and that is why it is not possible to download the miner, is there any other link?

[sp_]

I've used your leaked phi #2

Phi  sp-mod has only been released in one version. it does 40MHASH with +500 on the memclock and oc on the core 100% tdp. 1080ti

80% tdp 38 mhash. The Anx mod is doing 36MHASH with 100% tdp and the same config

[sp_]

I'm sorry, I've decided to download and check your miner, but unfortunately there is no such user on GitHub
and that is why it is not possible to download the miner, is there any other link?

The sp-mod kernels are private and there is no downloadlink.

[scryptr]

I'm sorry, I've decided to download and check your miner, but unfortunately there is no such user on GitHub
and that is why it is not possible to download the miner, is there any other link?

The sp-mod kernels are private and there is no downloadlink.

I THINK THE "ENEMY-MINE" LINK IS BOGUS--

I did not check it out earlier, but it does not work now.  Some of the recent posts and posters appear troll-oriented as well.       --scryptr

[rednoW]

https://github.com/Enemy-Mine/ccminer/releases/tag/v1.03-RVN

Hope nobody paid sp_ for his "fastest x17 and x16r for big farms only" shit ))

I'm sorry, I've decided to download and check your miner, but unfortunately there is no such user on GitHub
and that is why it is not possible to download the miner, is there any other link?

It was deleted from github for some reasons, maybe because source files was not correspond to binaries provided. But according to yiimp-based pools stats it is the most popular miner out there. So, there should be a link somewhere ... I have it on my personal cloud but won't post public. Don't want them ban me ))

[sp_]

I was thinking more like 75% PL alexis and 70% yours. Would you mind testing the power draw theory by running GPU-Z logging for both tests?

Here is a test that show my keccak sp-mod #3 is using 20% less power than the alexis miner to get the same hash.
When running on the same power setting sp-mod is around 6% faster.

keccak sp-mod #3 100% tdp oc vs alexis 1.0 cuda 7.5 visual studio 2013 32bit build (390.65 driver)



keccak sp-mod 80% tdp oc

[sp_]

Here is a snapshot of the nvprofiler. sp-mod #3 vs alexis 1.0 (80% tdp)

[rednoW]

I was thinking more like 75% PL alexis and 70% yours. Would you mind testing the power draw theory by running GPU-Z logging for both tests?

Here is a test that show my keccak sp-mod #3 is using 20% less power than the alexis miner to get the same hash.
When running on the same power setting sp-mod is around 6% faster.

keccak sp-mod #3 100% tdp oc vs alexis 1.0 cuda 7.5 visual studio 2013 32bit build (390.65 driver)

cuda7.5 is not a fastest one for alexis )) Stop fooling around

[sp_]

sp_ , what about the 20% faster skein or the 20% faster blake2s, over the "alexis slow code"  ?  
Show us your strength  

5-10% faster with 20% less power. That's around 20% more profit in a bear market with high power costs.

[sp_]

cuda7.5 is not a fastest one for alexis )) Stop fooling around

No cudatoolkit run faster than my mod. The alexis miner is compiled with compute 5.0 and 5.2 (maxwell) and on pascal the results will vary with the driver. When you upgrade the driver,  alexis legacy code is recompiled with the  latest NVIDIA developer optimalizations.

My test is conducted with one of the latest drivers. 390.65  

For keccak, the alexis miner is running best with cuda 7.5 32 bit and the latest driver.

On a more expensive 1080ti with higher intensity and more power my mod can do 1600MHASH, The alexis miner use more power and heat and trottle down to 1400

[NWMiningCo]

How can I access the binaries to test?  Happy to provide a donation for such an increase in hashrates. 

[Lumining]

I've seen someone mining LUX coin with this mining software: ccminer/alexis-1.0
Usually, everyone uses SPMOD1 for lux, can anyone tell me the difference between these?

[@enemy]

sp_ , what about the 20% faster skein or the 20% faster blake2s, over the "alexis slow code"  ?  
Show us your strength  

5-10% faster with 20% less power. That's around 20% more profit in a bear market with high power costs.

Nice ,   " 2 host threads in 1 gpu " again  or something  new?
j/k nvm

[sp_]

1 host thread. pure kernel optimalizations