[ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX]

[dga]

implementing fee is actually good so that not everybody can use. The rich get richer and the poor remain poor because they can't afford miner. Also prevent flooding of hashrate , good idea. Maybe in the future every new algo will need to pay fee. That is a really good idea. But I don't think Christian would want that. Right ? Christian is in for the love , not for greed. I bet relasing x11 would make a lot of people happy , partly because they have a lot of new coins to play with and they can participate in mainstream algo. Performance of algo is one thing , being able to mine the algo is another. I don't mind mining x11 with 1 mh on gtx 750 ti and later get performance boost, at least I can mine that algo and be happy about it.

No it isn't a good idea... you want to pay fee go mine xpm (aka mafia coin). 10% fee for mining where you are told to mine (and nowhere else).

The problem with the fee is not necessarily the fee itself, it is everything else actually:

* no open source code (meaning you cannot check what is really in); that blocks further development since only 1 persons has access to the code (more people means more optimization, new ideas etc...)

* you are limited on what you run, on the pool you run etc..

* defeat the purpose of crypto currency (both the decentralized stuff and the "corrupt banker" part)

  ok I am so fed up with this xpm crap (their mafia, their software, their pool, their unfriendly ddos to new pool... meaning all the crap they bring to cryptoworld) that I can't be for fee software...

There are many models.

yvg1900's fee-based software works on a lot of different pools.  It's closed source, but fairly open-ecosystem.

NaN's is similar, though it doesn't support quite as many protocols.

I tried a 10% dev fee on Riecoin decreasing every other day until 2%, and then open sourced it, and that worked out decently (but I wasn't trying to make a living off of this).

The prime stuff is currently a mess, but that's just a matter of time.  Once some of the miners reach performance parity and platform parity, the thing left to compete on is price and pool support.  At some point, the profit margin will get low enough that someone might take the nuclear option and open source their miner.

[gpuminer]

implementing fee is actually good so that not everybody can use. The rich get richer and the poor remain poor because they can't afford miner. Also prevent flooding of hashrate , good idea. Maybe in the future every new algo will need to pay fee. That is a really good idea. But I don't think Christian would want that. Right ? Christian is in for the love , not for greed. I bet relasing x11 would make a lot of people happy , partly because they have a lot of new coins to play with and they can participate in mainstream algo. Performance of algo is one thing , being able to mine the algo is another. I don't mind mining x11 with 1 mh on gtx 750 ti and later get performance boost, at least I can mine that algo and be happy about it.

No it isn't a good idea... you want to pay fee go mine xpm (aka mafia coin). 10% fee for mining where you are told to mine (and nowhere else).

The problem with the fee is not necessarily the fee itself, it is everything else actually:

* no open source code (meaning you cannot check what is really in); that blocks further development since only 1 persons has access to the code (more people means more optimization, new ideas etc...)

* you are limited on what you run, on the pool you run etc..

* defeat the purpose of crypto currency (both the decentralized stuff and the "corrupt banker" part)

  ok I am so fed up with this xpm crap (their mafia, their software, their pool, their unfriendly ddos to new pool... meaning all the crap they bring to cryptoworld) that I can't be for fee software...

that's what I am trying to say. Its for love and not for greed.

sorry.. you put fee at the beginning of your message, I didn't see the rest of the message  

yeah , I was trying to build up my message and deliver my point at the end 

[AizenSou]

There are many models.

yvg1900's fee-based software works on a lot of different pools.  It's closed source, but fairly open-ecosystem.

NaN's is similar, though it doesn't support quite as many protocols.

I tried a 10% dev fee on Riecoin decreasing every other day until 2%, and then open sourced it, and that worked out decently (but I wasn't trying to make a living off of this).

The prime stuff is currently a mess, but that's just a matter of time.  Once some of the miners reach performance parity and platform parity, the thing left to compete on is price and pool support.  At some point, the profit margin will get low enough that someone might take the nuclear option and open source their miner.

Prof. Andersen maybe I missed something here in the scenes. You have published a fast performance CPU-miner for Riecoin, right? I heard about the GPU-miner from you long ago, did you publish already somewhere?
I tried to find some infos about it but I only find about your CPU-miner. Could you guide me somehow?

Thanks in advance.

Regards,

[he-man]

Is there anyone here who makes a living out of mining with cuda cards only? no ASIC.

[bigjme]

Is there anyone here who makes a living out of mining with cuda cards only? no ASIC.

Christian?
I dont think so, with things being so new and the development of coins and cheap but powerfull nvidia cards only recently being released. The amount of investment needed to earn enough to live with would be huge.

[dga]

There are many models.

yvg1900's fee-based software works on a lot of different pools.  It's closed source, but fairly open-ecosystem.

NaN's is similar, though it doesn't support quite as many protocols.

I tried a 10% dev fee on Riecoin decreasing every other day until 2%, and then open sourced it, and that worked out decently (but I wasn't trying to make a living off of this).

The prime stuff is currently a mess, but that's just a matter of time.  Once some of the miners reach performance parity and platform parity, the thing left to compete on is price and pool support.  At some point, the profit margin will get low enough that someone might take the nuclear option and open source their miner.

Prof. Andersen maybe I missed something here in the scenes. You have published a fast performance CPU-miner for Riecoin, right? I heard about the GPU-miner from you long ago, did you publish already somewhere?
I tried to find some infos about it but I only find about your CPU-miner. Could you guide me somehow?

Thanks in advance.

Regards,

scrypt optimizations for nvidia - released open source after I mined privately on Amazon's GPUs for a week. :-)  (now in cudaminer)
  my blog post about it - http://da-data.blogspot.com/2013/12/inside-better-cuda-based-scrypt-miner.html

Protoshares GPU miner for nvidia:  Released open source by sponsorship from Invictus Innovations
   https://github.com/dave-andersen/cudapts
and recently ported to ypool:  https://github.com/dave-andersen/cudaptsx

Protoshares CPU optimizations - Invictus paid to have the algorithmic part incorporated into Bitshares.  I never got the assembly-optimized parts cleaned enough to think it was a good idea to put into the source tree -- it made the build too ugly for something that is designed to be able to run on any platform.

Riecoin, open-sourced after a week of 10% dev fee decreasing every other day:
  https://github.com/dave-andersen/fastrie
  Remains open source with a 2% dev fee that the user can disable at the command line.

Riecoin's an interesting one - my miner is pretty much the only one in use, but I don't earn 2% of all riecoin, I earn more like 0.2%.  But I'm experimenting with all of these to understand better what the alt-coin economy is like.

I've never tried making a pool, because it sounds like really hard, thankless work.  I'm watching the XPM miner experiment with great interest because of that.

I think it's amazing that Christian and Christian so consistently release *really good* toys to the community.  And I think it's equally important therefore to find a way to continue to make it worth their while to do so, because most other stuff that comes out targets the AMD crowd.  I don't know if that's a dev fee mechanism, voluntary contributions, or what, but it's worth thinking about more.

[djm34]

Is there anyone here who makes a living out of mining with cuda cards only? no ASIC.

Christian?
I dont think so, with things being so new and the development of coins and cheap but powerfull nvidia cards only recently being released. The amount of investment needed to earn enough to live with would be huge.

I think there are a few "big" Nvidia farm (not sure though if they already made a return over investment) but since the new algos are taking off together with the asic moving to scrypt . It could become profitable (if nvidia keep the price on the reasonable side... )

[cryptodragon]

Is there anyone here who makes a living out of mining with cuda cards only? no ASIC.

Christian?
I dont think so, with things being so new and the development of coins and cheap but powerfull nvidia cards only recently being released. The amount of investment needed to earn enough to live with would be huge.

I think there are a few "big" Nvidia farm (not sure though if they already made a return over investment) but since the new algos are taking off together with the asic moving to scrypt . It could become profitable (if nvidia keep the price on the reasonable side... )

Look at 1gh's HVC mining power in the first week, I remembered someone had 3.4Gh/s farms there. I believe at least part of that power came from nVidia side.  Look at other 'cuda coins' pool, i believe someone may have 100-200 750Ti size cards farm there. 

For established coins (Bitcoin, scrypt-based coins), I do believe ASIC is the future, like it or not.  I got my foot wet by purchasing a Antminer S2 batch 3 just weeks ago, and now battle with some minor hardware issue, but I think it is the right way to go....

Most of us are just geeks with the hope of rising cryptocurrencies, but unless you are a big farm owner or pool operator, it would be too risky to bid everything and make a living out of alter coin mining.  I bet someone may make more from trading than mining. Or the most profitable way is releasing some scam coin and just do pump and dump.  Just started mining back to Jan because of Doge ( I love that community), but I learned so much since then. Now I just follow Chris' summon on new cuda coins, and ignore all other scam coins. I think that's the "true power" of Chris, whenever he recommend something coins, followers like us will devote our hash power to at least give it a couple of days of try.

BTW, if Chris release a scam coin, I will follow.... LOL.

[Bombadil]

scrypt optimizations for nvidia - released open source after I mined privately on Amazon's GPUs for a week. :-)  (now in cudaminer)
  my blog post about it - http://da-data.blogspot.com/2013/12/inside-better-cuda-based-scrypt-miner.html

Protoshares GPU miner for nvidia:  Released open source by sponsorship from Invictus Innovations
   https://github.com/dave-andersen/cudapts
and recently ported to ypool:  https://github.com/dave-andersen/cudaptsx

Protoshares CPU optimizations - Invictus paid to have the algorithmic part incorporated into Bitshares.  I never got the assembly-optimized parts cleaned enough to think it was a good idea to put into the source tree -- it made the build too ugly for something that is designed to be able to run on any platform.

You know, Prof., when you published your DOT papers, I was all excited. I was a 15yrs-old into P2P filesharing those days, member of the Shareaza team (other nick). Though I was still unable to get my own implementation of DOT working fast enough at my old lappy. I might try it again with modern days standards (like an SSD! =D)
Anyway, you had a fan before, and now I know you're into crypto, you're the best professor ever! =D I'm hoping to see more from you

[mordecai779]

DRK is through the roof!

[cbuchner1]

DRK is through the roof!

15% is nice, but the roof is at 200%.

[noobminerd]

Make a found rally to relase the ccminer  x11 a lot of x11 now and nvidia it's out put a amount we donate until reach.

[bitcoinvideos]

Make a found rally to relase the ccminer  x11 a lot of x11 now and nvidia it's out put a amount we donate until reach.

I just don't think it's done yet that's all...not a matter of paying off C&C hashfactory....

[djm34]

right now, I am more interested in killer groestl... than in X11... DRK is out of reach in anyway

[ltcnim]

Beside of x11, I'd love to see the 1x to 16x riser boost. I think Christian said, it might improve hashing speed up to 30%, and that goes for every single algo implemented :>

[ivanlabrie]

Let's raise some funds for that then...I'm sure many of us can contribute with a few cents, eh?

EDIT: I manage a 6 gpu gtx 750 ti equipped rig, and it pays a cheap rent here in Argentina. Not half bad, I'd say.
Sips power, and is super quiet.

[cryptomines]

Let's raise some funds for that then...I'm sure many of us can contribute with a few cents, eh?

EDIT: I manage a 6 gpu gtx 750 ti equipped rig, and it pays a cheap rent here in Argentina. Not half bad, I'd say.
Sips power, and is super quiet.

I''ll gladly contribute some funds.

What are you mining with that  rig?

[cbuchner1]

Beside of x11, I'd love to see the 1x to 16x riser boost. I think Christian said, it might improve hashing speed up to 30%, and that goes for every single algo implemented :>

errr, no. only applies to cudaminer which needessly transfers computed hashes over PCIe

Christian

[dga]

The prime stuff is currently a mess, but that's just a matter of time.  Once some of the miners reach performance parity and platform parity, the thing left to compete on is price and pool support.  At some point, the profit margin will get low enough that someone might take the nuclear option and open source their miner.

Well - that took less time than I expected.  MadMax just took that option and open-sourced his.

https://bitcointalk.org/index.php?topic=602292.0

See?  These things work out in the long run. :-)  Or the not-so-long run.  Crypto seems to go at hyperspeed in that regard.

[ltcnim]

Beside of x11, I'd love to see the 1x to 16x riser boost. I think Christian said, it might improve hashing speed up to 30%, and that goes for every single algo implemented :>

errr, no. only applies to cudaminer which needessly transfers computed hashes over PCIe

Christian

really? cool, then it's really just x11. But since I'm currently still hunting the next jackpot @ JPC, I'm not in a hurry ^^