CCminer(SP-MOD) Modded GPU kernels.

[Velgelm]

I think I can do a blake2s mod with a big boost... Anybody interested?

maybe, try and post results

[scryptr]

ONE MINER--

I think you should combine your efforts into ONE (1) deluxe miner.  I suspect that it would sell much better.  Really, much better sales.       --scryptr

[pallas]

ONE MINER--

I think you should combine your efforts into ONE (1) deluxe miner.  I suspect that it would sell much better.  Really, much better sales.       --scryptr

I agree.

[sp_]

I have done blake2s +7% so far. Not a big boost

[pallas]

blake's no fun ;-)

[sp_]

blake's no fun ;-)

hehe. No. Enough blake for a while. (I have already modded and optimized decred,vcash,blake2s,pentablake,blake256)

0.1 BTC for blake2s subscription if anyone interested. I will post some more details and snapshot of the speed later.

[sp_]

ONE MINER--
I think you should combine your efforts into ONE (1) deluxe miner.  I suspect that it would sell much better.  Really, much better sales.       --scryptr

I agree.

A refork has been planned for a long time.. The latest mod's (decred,vcash @ blake2s) are all based on the latest 1.7.4/1.7.5 fork) I could merge my work in the 1.5.78 private to the latest.


What do you think about a sp-mod 1.7.5 with increases in the following algos:

pentablake(+100%)
decred (+20%)
vcash(+13%)
lyra2v2(+10%)
blake2s(+7%)
quark(+7%)
x11(+3%)
x13(+3%)
x14(+3%)
x15(+3%)
x17(+3%)
nist5(+5%)

All in one miner.

[sp_]

My blake2s is doing 2080-2100MASH on the 980ti with oc.

There is a bug in original kernal. I keep getting share has already been sendt error messages.



bug:

[flipclip]

My blake2s is doing 2080-2100MASH on the 980ti with oc.

There is a bug in original kernal. I keep getting share has already been sendt error messages.

Isn't that due to checks_dups being automatically set to true for blake2s

[survaisija]

ONE MINER--
I think you should combine your efforts into ONE (1) deluxe miner.  I suspect that it would sell much better.  Really, much better sales.       --scryptr

I agree.

A refork has been planned for a long time.. The latest mod's (decred,vcash @ blake2s) are all based on the latest 1.7.4/1.7.5 fork) I could merge my work in the 1.5.78 private to the latest.


What do you think about a sp-mod 1.7.5 with increases in the following algos:

pentablake(+100%)
decred (+20%)
vcash(+13%)
lyra2v2(+10%)
blake2s(+7%)
quark(+7%)
x11(+3%)
x13(+3%)
x14(+3%)
x15(+3%)
x17(+3%)
nist5(+5%)

All in one miner.

sounds nice

[scryptr]

SP_PRO 2.0 --

Has a nice ring to it.  I will set aside 0.1 BTC for another donation.  Happy Easter!       --scryptr

[sp_]

Thanks scryptr. Decred #7 is very profitable at the moment.

[bensam1231]

ONE MINER--

I think you should combine your efforts into ONE (1) deluxe miner.  I suspect that it would sell much better.  Really, much better sales.       --scryptr

I agree.

I disagree... Lets say his deluxe miner includes all five of his private miners he sells for .1BTC a piece. Do you want to pay .5BTC for his deluxe miner or even .4BTC? I believe you guys are assuming he'll cut you a deal for the whole package, when it might not be the case.

Current smaller miners have much more options for people mining different coins, although bundling means he'll sell code no one is really interested in, like Vanilla and Cryptonote...

That being said, it definitely could be a bundle if there was a miner fee... As you aren't purchasing single parts of the whole system to cut down on cost and increase ROI... Once again, miner fee is the best system.


Curiously SP, why don't you try a miner fee in addition to a paid fee? As in you release a private miner for .1BTC or 2% miner fee. People who really want it will pay .1 and people who don't have the money will mine on 2%. It's safe to assume there is no exclusivity here as we aren't paying $500 for a miner, so people who really want it will buy it anyway as .1 is affordable.

[sp_]

I just did a comparison between my buyable private miner (0.1BTC) and the latest 1.7.5 release.
I have written 0-10% faster, but I should have written 17-74% faster

Sp-Mod #5 default intensities hashrates in MHASH/s. (980ti 1290 coreclock/1650 memclock)

nist5: 43.80 (+17%)
quark: 27.85 (+61,91%)
lyra2v2 16.10 (+16.6%)
x11:   12.85 (+18%)
x13:   10.27 (+19.97)
x15:    8.83 (+18.52 )
neoscrypt: 0.99 (+73.68%)

1.7.5: (32bit windows build cuda 7.5) default intensities in MHASH/s (980ti 1290 coreclock/1650 memclock)

nist5: 37.45
quark: 17.20
lyra2v2: 13.81
x11: 10.88
x13: 8.56
x15: 7.45
neoscrypt: 0.57

1.7.5: (32bit windows build cuda 6.5) (980ti 1290 coreclock/1650 memclock)

nist5: 38.93
quark: 17.5
lyra2v2: 13.18
x11: 10.95
x13: 8.88
x15: 7.7
neoscrypt: 0.71

[sp_]

Curiously SP, why don't you try a miner fee in addition to a paid fee? As in you release a private miner for .1BTC or 2% miner fee. People who really want it will pay .1 and people who don't have the money will mine on 2%. It's safe to assume there is no exclusivity here as we aren't paying $500 for a miner, so people who really want it will buy it anyway as .1 is affordable.

I think I will do this, and include all the algos in one miner.

[bensam1231]

Curiously SP, why don't you try a miner fee in addition to a paid fee? As in you release a private miner for .1BTC or 2% miner fee. People who really want it will pay .1 and people who don't have the money will mine on 2%. It's safe to assume there is no exclusivity here as we aren't paying $500 for a miner, so people who really want it will buy it anyway as .1 is affordable.

I think I will do this, and include all the algos in one miner.

Something worth trying to see how much it makes you and whether or not people try to remove the fee. You always seem to be improving things, so you can just go back to a .1BTC private miner if it turns out horrible and improve it so it obsoletes the miner fee version.

[sp_]

I just did a comparison between my buyable private miner (0.1BTC) and the latest 1.7.5 release.
I have written 0-10% faster, but I should have written 17-74% faster

Sp-Mod #5 default intensities hashrates in MHASH/s. (980ti 1290 coreclock/1650 memclock)

nist5: 43.80 (+17%)
quark: 27.85 (+61,91%)
lyra2v2 16.10 (+16.6%)
x11:   12.85 (+18%)
x13:   10.27 (+19.97)
x15:    8.83 (+18.52 )
neoscrypt: 0.99 (+73.68%)

Current profit with my modded kernal on nicehash.com:(one single gtx 980ti, 24 hours mining)

neoscrypt:    0,00971250BTC ($4,0792500)
quark:   0,00907910BTC ($3,8132220)
x15:       0,00819512BTC ($3,4419504)
x13:       0,00796027BTC ($3,3433134)
x11:       0,00718058BTC ($3,0158436)
lyra2v2:   0,00663964BTC ($2,7886488)
nist5:     0,00161184BTC ($0,6769728)

[crysx]

I just did a comparison between my buyable private miner (0.1BTC) and the latest 1.7.5 release.
I have written 0-10% faster, but I should have written 17-74% faster

Sp-Mod #5 default intensities hashrates in MHASH/s. (980ti 1290 coreclock/1650 memclock)

nist5: 43.80 (+17%)
quark: 27.85 (+61,91%)
lyra2v2 16.10 (+16.6%)
x11:   12.85 (+18%)
x13:   10.27 (+19.97)
x15:    8.83 (+18.52 )
neoscrypt: 0.99 (+73.68%)

Current profit with my modded kernal on nicehash.com:(one single gtx 980ti, 24 hours mining)

neoscrypt:    0,00971250BTC ($4,0792500)
quark:   0,00907910BTC ($3,8132220)
x15:       0,00819512BTC ($3,4419504)
x13:       0,00796027BTC ($3,3433134)
x11:       0,00718058BTC ($3,0158436)
lyra2v2:   0,00663964BTC ($2,7886488)
nist5:     0,00161184BTC ($0,6769728)

does this mean you will cut out most of the other algos that are already contained in ccminer-tpruvot175 and only provide the above kernels? ...

or still have the other algos - but have the above modded - but still be able to mine the other algos ( non-modded ) ? ...

#crysx

[bensam1231]

Curiously is the Decred #7 you sent to .1 subs the same as vanilla+decred you sent out before? If not, what's the difference in speed?

[scryptr]

Curiously is the Decred #7 you sent to .1 subs the same as vanilla+decred you sent out before? If not, what's the difference in speed?

DELUXE EDITION--

When you release the number one top miner, SP_PRO v 2.0, please let us all know.  I will donate.

The hash rates seem real.  Just mark the release and I'll go for it.       --scryptr