CCminer(SP-MOD) Modded GPU kernels.

[sp_]

I will create a NVIDIA gpu miner as well, perhaps tonight. And then point one 750ti(3MHASH) to recieve my 0.04BTC a day
0.04BTC/day with a cost of 40watt of power (less than 1KWh/day)

With the current difficulty my private bastion miner will make with one single 750ti card:

Profit per day: $11,68
cost of power per day: $0.1

if I pick up a 750ti on ebay for $100 I will ROI the card in less than 10 days.

This is because my kernal will be 60 times faster than a modern Intel pentium i7.

[chrysophylax]

hi all ...

will be back in 24-48hours and continue working on the projects - including thefarm ...

just reading up on the joincoin 'thing' ...

would really like to test it when i get it thefarm running again ...

sp - is there a linux version we could use for the optimized kernel ( pentablake is it? ) on the gigabyte 750ti oc lp cards? ...

wolf - have you run it on the test machine? do you want to run it on farmamd? ...

#crysx

For what? I own the hashrate for Bastion - more hash isn't going to make more profit, difficulty would just meet us...

for a higher return ...

but its no issue ... i actually thought it was the algo that was discussed earlier with us - but realized its not ... so all good ...

#crysx

[chrysophylax]

hi all ...
will be back in 24-48hours and continue working on the projects - including thefarm ...
just reading up on the joincoin 'thing' ...
would really like to test it when i get it thefarm running again ...
sp - is there a linux version we could use for the optimized kernel ( pentablake is it? ) on the gigabyte 750ti oc lp cards? ...
wolf - have you run it on the test machine? do you want to run it on farmamd? ...
#crysx

The Pentablake profit has decreased to 0.004BTC per gtx 970. 2 days ago it was 0.01btc per gtx 970 (per day).

quark@nicehash is around 0.002BTC. This meens that without the private pentablake miner, you earn more mining quark or lyra2v2

yes - but is there a linux version of your optimized miner that i can compile and use on thefarm? ...

even if its another private one ( or even profitshare ) - ill maintain the deal for as long as the agreement holds - if we come to one ...

#crysx

[chrysophylax]

I will create a NVIDIA gpu miner as well, perhaps tonight. And then point one 750ti(3MHASH) to recieve my 0.04BTC a day
0.04BTC/day with a cost of 40watt of power (less than 1KWh/day)

With the current difficulty my private bastion miner will make with one single 750ti card:

Profit per day: $11,68
cost of power per day: $0.1

if I pick up a 750ti on ebay for $100 I will ROI the card in less than 10 days.

This is because my kernal will be 60 times faster than a modern Intel pentium i7.

ok - so what will 40 of those cards bring in? ...

want to profit share from them? ...

#crysx

[sp_]

ok - so what will 40 of those cards bring in? ...
want to profit share from them? ...
#crysx

40 cards will only bring in 23$ a day. since the difficulty will skyrocket.

80 blocks/160 coins per hour so 3840 coins per day. But that's across all 13 algos which means roughly 295 coins per algo per day, hence my 0.082 figure. Unless I missed something.

bathrobehero, just wrote it.
With the current marketprices and the current joincoin block reward system, only 0.082BTC are handed out to the miners per day. (BASTION)
With 40 cards. you will get 100% of the coins. With 1 card you get 50% of the coins (with the current difficulty)

But if wolf0 turns on his furyx's you get less.

This is because his kernal is 300 times faster than a modern Intel pentium i7.


There is no public opensource gpu miner for the Bastion algo... The competion is the CPU miners, and their mining software only produce 80KHASH on the i7 3,5Gz

[sp_]

yes - but is there a linux version of your optimized miner that i can compile and use on thefarm? ...
even if its another private one ( or even profitshare ) - ill maintain the deal for as long as the agreement holds - if we come to one ...
#crysx

No linux from me.. But I might sell the sourcecode, and then you get Linux compability.

The price is set at 3 BTC.

Pentablake sp-mod 3 (with sourcecode)

[sp_]

yes - but is there a linux version of your optimized miner that i can compile and use on thefarm? ...
even if its another private one ( or even profitshare ) - ill maintain the deal for as long as the agreement holds - if we come to one ...
#crysx

No linux from me..

Maybe we should work together, sp_? I used to use Windows exclusively - I now detest it, but am DAMNED good at cross-compiling, as well as Windows development from my Windows days - in addition to Linux from now. Meaning, I could port your shit, if necessary.

I have sold 8 copies of the pentablake and made 2,4 BTC already. I think the marked is full since the profit of mining penta with the current difficulty is only at 200% more profitable than mining quark@nicehash.

But the sourcecode is for sale for 3 BTC.

The sourcecode should be bought by the NVIDIA, so they can create  a proper compiler. Pentablake sp-mod 3 is 125% faster than the opensource.

My code will transform a gtx 970 to a TitanX

[chrysophylax]

yes - but is there a linux version of your optimized miner that i can compile and use on thefarm? ...
even if its another private one ( or even profitshare ) - ill maintain the deal for as long as the agreement holds - if we come to one ...
#crysx

No linux from me..

Maybe we should work together, sp_? I used to use Windows exclusively - I now detest it, but am DAMNED good at cross-compiling, as well as Windows development from my Windows days - in addition to Linux from now. Meaning, I could port your shit, if necessary.

two devs working together to cross compile AND optimize? ...

am i dreaming here - or are my dreams coming to reality? ...

THAT would be awesome ...

as long as the cross-compiling works for the nvidia crads AND you could optimize them for amd cards as well - that would cover the full spectrum of the market ...

i would definitely be interested in helping where i could in this scenario ...

#crysx

[chrysophylax]

yes - but is there a linux version of your optimized miner that i can compile and use on thefarm? ...
even if its another private one ( or even profitshare ) - ill maintain the deal for as long as the agreement holds - if we come to one ...
#crysx

No linux from me..

Maybe we should work together, sp_? I used to use Windows exclusively - I now detest it, but am DAMNED good at cross-compiling, as well as Windows development from my Windows days - in addition to Linux from now. Meaning, I could port your shit, if necessary.

I have sold 8 copies of the pentablake and made 2,4 BTC already. I think the marked is full since the profit of mining penta with the current difficulty is only at 200% more profitable than mining quark@nicehash.

But the sourcecode is for sale for 3 BTC.

The sourcecode should be bought by the NVIDIA, so they can create  a proper compiler. Pentablake sp-mod 3 is 125% faster than the opensource.

My code will transform a gtx 970 to a TitanX

would the sourcecode readily compile in fedora 22 x64 cuda 7.5 ( f22x64c7.5 ) ? ...

#crysx

[sp_]

would the sourcecode readily compile in fedora 22 x64 cuda 7.5 ( f22x64c7.5 ) ? ...
#crysx

If you can compile the sp-mod release 72 you can compile the pentablake miner as well..
Tonight I will make a Bastion miner. Not sure what speed I can get yet..

[theotherme]

yes - but is there a linux version of your optimized miner that i can compile and use on thefarm? ...
even if its another private one ( or even profitshare ) - ill maintain the deal for as long as the agreement holds - if we come to one ...
#crysx

No linux from me..

Maybe we should work together, sp_? I used to use Windows exclusively - I now detest it, but am DAMNED good at cross-compiling, as well as Windows development from my Windows days - in addition to Linux from now. Meaning, I could port your shit, if necessary.

I have sold 8 copies of the pentablake and made 2,4 BTC already. I think the marked is full since the profit of mining penta with the current difficulty is only at 200% more profitable than mining quark@nicehash.

But the sourcecode is for sale for 3 BTC.

zzzzzzzzzzzzzzzzzzzzzzzzzzzz
(after all if it works...  )

[chrysophylax]

would the sourcecode readily compile in fedora 22 x64 cuda 7.5 ( f22x64c7.5 ) ? ...
#crysx

If you can compile the sp-mod release 72 you can compile the pentablake miner as well..
Tonight I will make a Bastion miner. Not sure what speed I can get yet..

ok - thats great ...

ill have a look more in depth when i get back ...

seems i have a great deal of work to do - so will be working flat out to get it all sorted ...

id be interested to see what the bastion miner can do ... and maybe discuss if the pentablake kernel itself would be a viable inclusion in what im getting done here with thefarm ...

im actually looking at a number of algos for granite now - as a replacement or an addition to the x11 algo its based on ...

tanx sp ...

#crysx

[myagui]

I have sold 8 copies of the pentablake and made 2,4 BTC already. I think the marked is full since the profit of mining penta with the current difficulty is only at 200% more profitable than mining quark@nicehash.

But the sourcecode is for sale for 3 BTC. [...]

I know that profits are never guaranteed, be that with private or public versions.
Market conditions and profitability can change in an instant.

That said, do you really think that if you sell the sources now, that there is any chance at all for those eight early buyers to ever land a profit?
With each new sell that you make, for such a tiny market/opportunity, the returns shrink further for those that bought in. It's not as if the pie is large enough for everybody, is it?

Do note, I do not give a flying f@%# for what you make or don't make on this miner, I have a big fat zero stake on this one. I'm only making this comment as it looks to me that you're approaching private sales in a way that will ultimately hurt your own profits as well as those of people buying stuff from you.

I'll be happy to see everyone involved in the green, and the more, the merrier.

[sp_]

I have sold 8 copies of the pentablake and made 2,4 BTC already. I think the marked is full since the profit of mining penta with the current difficulty is only at 200% more profitable than mining quark@nicehash.
But the sourcecode is for sale for 3 BTC. [...]

I know that profits are never guaranteed, be that with private or public versions.
Market conditions and profitability can change in an instant.
That said, do you really think that if you sell the sources now, that there is any chance at all for those eight early buyers to ever land a profit?
With each new sell that you make, for such a tiny market/opportunity, the returns shrink further for those that bought in. It's not as if the pie is large enough for everybody, is it?
Do note, I do not give a flying f@%# for what you make or don't make on this miner, I have a big fat zero stake on this one. I'm only making this comment as it looks to me that you're approaching private sales in a way that will ultimately hurt your own profits as well as those of people buying stuff from you.
I'll be happy to see everyone involved in the green, and the more, the merrier.

For linux compability I can include the sourcecode for the right buyer.
After 8 sales of the pentablake, the public miner is not profitable anymore. I will continue with other algos.

-quark
-lyra2v2
-x11
-x13
-x15
-whirlpool

etc. etc..

I believe I can do a 50% in most of the algos. With 8 lucky buyes in each of the algos I get 10-15BTC Next up is

The Bastion algo...

Aiming for 6000% faster than a modern pentum i7 with the same wattage (750ti)

[theotherme]

I have sold 8 copies of the pentablake and made 2,4 BTC already. I think the marked is full since the profit of mining penta with the current difficulty is only at 200% more profitable than mining quark@nicehash.
But the sourcecode is for sale for 3 BTC. [...]

I know that profits are never guaranteed, be that with private or public versions.
Market conditions and profitability can change in an instant.
That said, do you really think that if you sell the sources now, that there is any chance at all for those eight early buyers to ever land a profit?
With each new sell that you make, for such a tiny market/opportunity, the returns shrink further for those that bought in. It's not as if the pie is large enough for everybody, is it?
Do note, I do not give a flying f@%# for what you make or don't make on this miner, I have a big fat zero stake on this one. I'm only making this comment as it looks to me that you're approaching private sales in a way that will ultimately hurt your own profits as well as those of people buying stuff from you.
I'll be happy to see everyone involved in the green, and the more, the merrier.

For linux compability I can include the sourcecode for the right buyer.
After 8 sales of the pentablake, the public miner is not profitable anymore. I will continue with other algos.

I don't understand your "new" sales pitch (considering you have been advocating for open-source for almost a year), I mean I don't even understand your sale pitch at all  
Your sale pitch, let me speechless

and I find that a bit pathetic... (I guess you noticed that already  )

(Also I totally agree with what myagui said...  )

If your interest is just getting btc, make a coin (a POS one  ) and ask for an IPO on yobits (rinse and repeat every week...  )

Sure we want to make money, but also, we don't want that people buying our (private) miner to not make any (that's one of the reason I am a little reluctant about selling my miners to small miners... knowing that I have as well to make a profit, and cover myself from the risk of leak...), if you are just in the "you take your responsibility" then you are probably closer in mentality to cloud mining scammers...  

[Grim]

wow ETH nice spike ...

any fix for the limited random access vram buffer on maxwell gpus?

anything above 1GB is useless for compute 5.0 architecture  

and anything above 2GB is useless for compute 5.2 architecture which still is large enough for the current dagger file

[sp_]

I have added bastion without gpu kernals@github. But you can only mine with the cpu. My lowend celeron is making 26KHASH

I get low difficulty at the pool, so you have to experiment with the --diff parameter.My cpu is too slow, but perhaps some of you can find the correct diff..

example of bat:

ccminer -D --diff 0.05 -a bastion  -o stratum+tcp://stratum.coinspool.cu.cc:3046 -u <joincoinadress> -p x

[sp_]

I don't understand your "new" sales pitch (considering you have been advocating for open-source for almost a year), I mean I don't even understand your sale pitch at all  
Your sale pitch, let me speechless

I have sold private kernals for more than one year. I have opensourced the bastion cpu miner. Don't you like it?

The gpu miner will be 60 times faster but closed source. You can buy it if you want..

[djm34]

I don't understand your "new" sales pitch (considering you have been advocating for open-source for almost a year), I mean I don't even understand your sale pitch at all  
Your sale pitch, let me speechless

I have sold private kernals for more than one year. I have opensourced the bastion cpu miner. Don't you like it?

The gpu miner will be 60 times faster but closed source. You can buy it if you want..

lol, not sure you will sell a lot of miner for bastion, it doesn't work, according the image you posted...
(it clearly doesn't validate wherever the solution is sent...   it is a bit like your unfinished axiom miner... )

[Genoil]

wow ETH nice spike ...

any fix for the limited random access vram buffer on maxwell gpus?

anything above 1GB is useless for compute 5.0 architecture  

and anything above 2GB is useless for compute 5.2 architecture which still is large enough for the current dagger file

Tried everything in my power to get to a solution but if even the big guys in (ArchaeaSoftware and njuffa are among the initial creators of CUDA) don't have an answer...there's a tiny spark of hope in the last comment of that thread, that apparently the issue has raised some attention within Nvidia, but I don't know what lies in their abilities.