CCminer(SP-MOD) Modded GPU kernels.

[Velgelm]

just compare EQM 1.0.0b VS SP #1
1070 +200 core -500Mem
960 +200 core -500 Mem
Core i3 2100
cd 0 1
-t 4
EQM


SP

[1714971392]

1714971392

[1714971392]

1714971392

[bathrobehero]

There's a miner with up to 2.5 million sol/s who joined the party:



https://www2.coinmine.pl/zec/index.php?page=statistics&action=pool

[ol92]

Hello,
spmod with GTX 1080: the limiting factor is probably memory latency.
Oveclocking the core and memory (real gpu freq around 2088, mem around 11500) : around 110-120 sol/s.
Underclocking memory: less than 90s/s
Underclocking gpu @1500  and overclok mem @ 11500 : around 110S/S (almost the same than both overclocked).

This is a very different story than GTX 1070.

With core +100, powerlimit at the minimum (around 40% I think) (which leads to real core freq arount 1500mhz , mem@11500) : around 110 sol/s, power usage (hwinfo) decreased to 88W.

To be compared to the fury x @1100/500 which gives 220 s/s @180W : we have similar hashrate/power ratio. 

[NiceHashSupport]

just compare EQM 1.0.0b VS SP #1
1070 +200 core -500Mem
960 +200 core -500 Mem
Core i3 2100
cd 0 1
-t 4
EQM


SP

For EQM, you will get much higher speed with mem OVERCLOCK (not underclock). But too much overclock can cause mem corruptions that lead to crashes, so make sure your mem overclock is stable. We will try to make next version less prone to data corruptions due to overclocks.

[Nikolaj]

MINING WITH NICEHASHMINER 1.7.3.4--

OK, it is not sp_'s miner, but I was interested because NiceHash had rolled their proprietary EQM Equihash miner into the package, and because it auto-switches.  It auto-switched to Dagger-Hashimoto a few minutes after launching, that is a good sign that Zcash profitability is down.
 
But, the craziest thing!  I came home from the grocery store, and it was mining Dagger-Hashimoto and Equihash at the same time!  There are no claims on the NiceHash blog about this, but the miner is doing it.  There are two console windows on my Win 7 x 64 box, one mining  Equihash at about 30-40 Sols/s, the other mining Dagger-Hashimoto at about 17-20MH/s.   Both hash rates fluctuate, and the Dagger-Hashimoto console window is always closing and loading a new DAG file.

When mining only Zcash, my 980ti + i7 2600 CPU get about 130Sols/s with the NiceHash EQM miner.  When mining only ETH, the 980ti gets about 20MH/s.  It is as if the Equihash is a secondary coin in this (dual?) system.

It will run overnight.  I'll check in the morning.  NiceHash statistics say that the rig is earning more.       --scryptr

EDIT:  Here is a screenshot--


Two mining consoles open simultaneously.  980ti + i7 2600 CPU mining.

Is it a dual miner?       --scryptr

well, not

you're triple mining, with low efficiency.

CPU mining on equihash, worthless (1 window, in common with the vga)
VGA mining on equihash, worthless with such speeds (window in common with the CPU)
VGA mining on dagger hashimoro, low efficiency due CPU loading and cohoperative mining with equihash

It's not a bug, the window of equihash remains open due cpu mining. If you disable the cpu you won't have any. You should simply disable CPU

[duboisi]

https://z.cash/blog/founders-reward-transfers.html

Profit while you can before market is flooded with ZEC after after November 28th, 2016.

[scryptr]

MINING WITH NICEHASHMINER 1.7.3.4--

OK, it is not sp_'s miner, but I was interested because NiceHash had rolled their proprietary EQM Equihash miner into the package, and because it auto-switches.  It auto-switched to Dagger-Hashimoto a few minutes after launching, that is a good sign that Zcash profitability is down.
 
But, the craziest thing!  I came home from the grocery store, and it was mining Dagger-Hashimoto and Equihash at the same time!  There are no claims on the NiceHash blog about this, but the miner is doing it.  There are two console windows on my Win 7 x 64 box, one mining  Equihash at about 30-40 Sols/s, the other mining Dagger-Hashimoto at about 17-20MH/s.   Both hash rates fluctuate, and the Dagger-Hashimoto console window is always closing and loading a new DAG file.

When mining only Zcash, my 980ti + i7 2600 CPU get about 130Sols/s with the NiceHash EQM miner.  When mining only ETH, the 980ti gets about 20MH/s.  It is as if the Equihash is a secondary coin in this (dual?) system.

It will run overnight.  I'll check in the morning.  NiceHash statistics say that the rig is earning more.       --scryptr

EDIT:  Here is a screenshot--


Two mining consoles open simultaneously.  980ti + i7 2600 CPU mining.

Is it a dual miner?       --scryptr

well, not

you're triple mining, with low efficiency.

CPU mining on equihash, worthless (1 window, in common with the vga)
VGA mining on equihash, worthless with such speeds (window in common with the CPU)
VGA mining on dagger hashimoro, low efficiency due CPU loading and cohoperative mining with equihash

It's not a bug, the window of equihash remains open due cpu mining. If you disable the cpu you won't have any. You should simply disable CPU ;)t  

IT IS A BUG--

Waking up, one window was crashed (the Dagger-Hashimoto miner) and the equihash miner was running alone.  I may have lost $1 or so in the process.  The equihash miner was reporting a higher pay rate mining alone than when the system was mining both algorithms, about 1.80/day vs $0.50/day.  I shut the switching miner off.       --scryptr

[fenomenhaa]

Sp1 vers.
1070  +850mem +30 core 115 sol
970  +350mem  72 Sol

[Amph]

There's a miner with up to 2.5 million sol/s who joined the party:



https://www2.coinmine.pl/zec/index.php?page=statistics&action=pool

assuming that is a amd rig, which can do easily 1000 sol, we have 2000 rig there(now it's showing 2M sol), probably a consortium or a private pool, not a sinlge guy with all those rigs, i can't believe that

[jeanjean15]

Hello,
spmod with GTX 1080: the limiting factor is probably memory latency.
Oveclocking the core and memory (real gpu freq around 2088, mem around 11500) : around 110-120 sol/s.
Underclocking memory: less than 90s/s
Underclocking gpu @1500  and overclok mem @ 11500 : around 110S/S (almost the same than both overclocked).

This is a very different story than GTX 1070.

With core +100, powerlimit at the minimum (around 40% I think) (which leads to real core freq arount 1500mhz , mem@11500) : around 110 sol/s, power usage (hwinfo) decreased to 88W.

To be compared to the fury x @1100/500 which gives 220 s/s @180W : we have similar hashrate/power ratio. 

Thanks for your feedback .

It helped me to save 100 watts with my two GTX 1080 .

[tbearhere]

SP is there an intensity command in that build?  thx   

[Velgelm]

my lastest comparsion updated:
eqm 1.0.0.b
GTX 960 1478 core 3505 Mem -t 0
48 Ss


1070 +100 core +850 Mem -t 0
132 Ss

[giagge]

I dont understand , is possible mining ETH and Zcash together ?

If yes , with zcash-spmod1 and claymore 7.3 ? .

I have gtx 1070 on windows x64 .

[Amph]

I dont understand , is possible mining ETH and Zcash together ?

If yes , with zcash-spmod1 and claymore 7.3 ? .

I have gtx 1070 on windows x64 .

yeah just run two instances, one clymore and one spmod

[giagge]

I dont understand , is possible mining ETH and Zcash together ?

If yes , with zcash-spmod1 and claymore 7.3 ? .

I have gtx 1070 on windows x64 .

yeah just run two instances, one clymore and one spmod

Thanks forever amph , i try it now this is my result with only 2x gtx 1070 :

Power Limit 50% , clock gpu + 150 , clock memory +200 .



.

[mstang83]

here is the build

Zcash sp-mod #1

https://github.com/sp-hash/ccminer/releases/tag/zcash-spmod1

sp_, you're the man!! Thanks!!!

What rates are you getting?

After 12+ hours, looks like the software is reporting stable 114-121 Sols with 4 -Asus GTX 750ti GPU's standard clocked on this rig. Even with this, it's still more than double I was getting before. No support for AVX or AVX2 on the CPU. I think 8GB of RAM on this system. Pleased so far. Thank you!!

[tbearhere]

nhewminer is known to be unstable if you don't run with -t 0 (no cpu mining)

As a workaround. Run this

https://github.com/chriswier/zcash-genoil-loop/blob/master/disable-windows-errorui.bat

And put the miner in a loop in the batfile.

:loop
nheqminer -t 0 -cv 1 -cd 0 1 2 3 4 -cs -l eu1-zcash.flypool.org:3333 -u t1X4ghuK74TfmJeij9YxxEm6QQuJq5sEmsE.test
goto loop

For me I ran no -t at all last night and worked fine. Now I noticed a slight increase using 75% of my total cores, see if that stays stable. I have had no crashes yet. Lets see what the results are with 75% cpu. 

[bensam1231]

There's a miner with up to 2.5 million sol/s who joined the party:



https://www2.coinmine.pl/zec/index.php?page=statistics&action=pool

It's not a shift in network hashrate from Ethereum which means these are probably not GPUs. It would've already been present under one of the algos.

Heavily botneted algos like Cryptonote haven't seen a huge drop in difficulty, so this is new.

Guess this is AWS instances on highly optimized CPU miners. Someone figured out how to make that profitable, that's why it changes so much. Looks like it scales up and down with network profitability. The miner is on coinmine.

[mstang83]

There's a miner with up to 2.5 million sol/s who joined the party:



https://www2.coinmine.pl/zec/index.php?page=statistics&action=pool

It's not a shift in network hashrate from Ethereum which means these are probably not GPUs. It would've already been present under one of the algos.

Heavily botneted algos like Cryptonote haven't seen a huge drop in difficulty, so this is new.

Guess this is AWS instances on highly optimized CPU miners. Someone figured out how to make that profitable, that's why it changes so much. Looks like it scales up and down with network profitability. The miner is on coinmine.

Everyone updating to more optimized mining software could easily double the difficulty. Don't you think?

[bensam1231]

There's a miner with up to 2.5 million sol/s who joined the party:



https://www2.coinmine.pl/zec/index.php?page=statistics&action=pool

It's not a shift in network hashrate from Ethereum which means these are probably not GPUs. It would've already been present under one of the algos.

Heavily botneted algos like Cryptonote haven't seen a huge drop in difficulty, so this is new.

Guess this is AWS instances on highly optimized CPU miners. Someone figured out how to make that profitable, that's why it changes so much. Looks like it scales up and down with network profitability. The miner is on coinmine.

Everyone updating to more optimized mining software could easily double the difficulty. Don't you think?

It's one miner, it's on coinmine.pl.