Topic: CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 654. (Read 2347641 times)

I believe Probability (Genesis program doesn't use GPL code) < 0.01
Of course I can't prove it.
On the other hand, since they don't sell it and use "privately", I think they can do this

You are completely missing the point. Your example is the right way to do it. Presumably they did not use any open source,
Their code is all proprietary so they are not obliged to publish it.

SP is using open source code made by others adding a little bit of his own then not publishing the source.

It has nothing to do with charging for it, it's all about how you use someone else's code in your product. See the difference?

I agree.

Look at Gensis Mining?

Their coders wrote a Ethereum program called Enigma, which gives a significant boost to hash output - no-one is hate posting on their thread for them to opensource Enigma.

They've kept is completely secret; you can bet there are pieces of opensource coding in it.

SP_ was getting £16 donations in 2015, for a windows compiled version of ccminer, with tweaks and optimisations for small miners who cannot code and cannot compile from source.

People are being way to harsh on this thread to SP_

STOP BASHING SP_ --

For some spare change, sp_ has done good code work  It could use some careful editing, but I am mining with a faster miner.

What I would like to see:

          1) Finish the univeral miner and release it.

          2) Finish the dual Ethminer and release it.

          3) Release the private Ethminer!

If any code must be released, sp_ is allowed to charge a fee, hobby-sized, in order to keep up with expenses and fund further development.  Other devs are selling code to farms for big fees, and releasing nothing.  I am guessing that they started with public code.  These devs are not in the public forums, but they deserve the criticism that sp_ is getting.

--scryptr

What you are saying is ridiculous, because no-one would ever improve any piece of opensource software or offer their improvements to people who do not code or have the time to learn to code.

Then, Nvidia users would be stuck with rubbish versions of mining software, like what has happened with SGminer for AMD GPU owners.

And if I have written my own compiler, do I need to publish the compiler too?

Dickhead.

Semantics


Correct.


That's where you're wrong. If you used open source to build your binary you must publish all the source used.
It doesn't matter if your charge for the binary or distribute it only to donators or give it away free, if it contains
open source code then any code you add is also open source and must be published.

I will do it for 0.1BTC. Put the slow japanese (?) kernal into the free opensource official release. What do you say nicehash?

I am not selling anything. I gave away a faster lyra2v2 kernal  for free to my donators.
The 980ti is doing 40MHASH with this kernal. The sp-mod #6 it did 17MHASH. That is a +135% increase on the same clocks.

The Nichash opensource is +100%. When you opensource something you should expect that somebody else fork and optimize your work. If not you should spread the binary..

35% more money if you use my modded kernal. 0.1 BTC is nothing for some of you.

NO. What matters is he is selling binaries built from open source without publishing the source code, in violation of the GPL.
It would be a violation even if he was not charging for the binaries, but withholding the source code is a deliberate attempt
to keep his changes closed. Providing only binaries is not a service it is a restriction. With the exception of Spreadx11 he did
not develop many algos, he simply optimized them. Many of the optimizations he takes credit for were produced by other developpers.
He did a lot of good work and as long as he was keeping it open it didn't matter that he was taking more credit than he
deserved. It was only after he started trying to profit from other's work without publishing his code that people started
complaining.

And the latest stunt with Nicehash Miner is a perfect example.

If you know about changes, Please read http://askmona.org/4314.

Summary of 1.7.6-r5

• Support for Pascal, Kepler, and Fermi GPUs
• Optimized for Kepler and Fermi GPUs
• Support solo mining Monacoin
• Faster Cubehash
• Bugfix
• And so on...

A lot of people do not know how to compile from source and SP_ does offer a useful service to many small miners.

Secondly, he was the main coder for the spreadcoin Nvidia miner.

Thirdly, SP_ added a degree of competition, which helps to develop better coding mining software, which has consistently caused much better mining software on Nvidia GPU's.

Look at the situation on AMD GPU's.

The big farms, who hire coders have huge advantages over the official sgminer coding. Consequently, the people most likely to mine and hold those coins are put off.

This depresses the Alt Coin prices, because it takes many years to develop a successful Crypto currency and you need to have many wallet holders to popularise an Alt Coin.

Does it really matter that much, that SP_ made some BTC when BTC was at a very low Fiat Currency price, he sold compiled Window versions of the software to people who cannot compile from source, he developed the original mining software for many algos and when he get's drunk on this thread his posts are really funny.

Remember, he sold this software when BTC was £160 per coin in 2015 e.g. £16 for windows complied version of ccminer, which then got another 3-9 releases over the next year.

To be fair: I can spend £16 in one weekend on one bottle of Southern Comfort.

Now, it looks a little steep, because BTC is £400 a coin.

The only reason SP_ sales have down recently, is because he didn't cut his BTC price, when the BTC doubled from £160 to £400 over the last 16 months.

I think, everyone is being way to harsh on SP_, because he has all of these people on this thread asking him for more software releases on a daily basis.

For starters I presume one is based off SP the other TPruvot. I would also guess they are similarly optimized rfrom different bases.

This is great news considering SP is burned out.

A big thank you to Nanashi San and Nicehash.

I agree with Nicehash's disappointment with SP trying to sell free code. It is also technically a violation of the GPL
to sell tainted code without the source. Adding another 10% to a 100% free speedup then charging for it is pure and
simple greed.

Whats the difference between 1580r10 and 176r5 ?
Is there way to read about changes between versions?

well as you said it's an hobby, if you pretend to make million on your hobby, there is something wrong on your thinking

Don't be too hard: sp sent optimized version to his supporters, for free. That's not greedy. And author sold his miner to you?

Yeah, I know. But Nanashi Meiyo-Meijin's ccminer 1.7.6 is faster than nicehash one.
There are his ccminer binary and source.
Current version is 1.7.6-r5,and current source version is ccminer-windows 1.7.6-r5.
1.7.6-r5-fix is for Kepler and Fermi. Nanashi Meiyo-Meijin's ccminer 1.5.80-r10 is here too.
https://onedrive.live.com/?authkey=%21AHk1hss4HZtCU60&id=C7ABE390AB1575E7%21825&cid=C7ABE390AB1575E7
This ccminer needs the latest NVIDIA GeForce drivers.

Nanashi Meiyo-Meijin BTC: 1P2SbHB5DaGPymu9C8WQCsEbnrw3dEWcDe
please donate.

I don't think they care so much you're selling it, rather you aren't adding it to your public releases which a lot of people use and crediting it to them.


Any plans for a dual miner for Nvidia?

Tpruvot merged the code from the nicehash fork.

I made it 10% faster. Not greed. Evolution.