CCminer(SP-MOD) Modded GPU kernels.

[sp_]

Most of alexis work is based on my opensource optimized maxwell ccminer kernels. I have more than 1000 commits to github. Don't forget this... Anyway, the unreleased nist5 private is doing 55mhash. Ccminer 2.1 45 mhash and alexis 1.0.0 48mhash.
Gtx 1070 same clocks and launch config.

Lie!
Alexis was the first to implement shared mem technique that gives the significant boost

In nist5? nope.

Alexis improved AES by moving some of the precalc tables from shared mem to the level 1 cache. This give a significant boost in most of the X algos. (x11,x13,x15,x17)

The nist 5 isn't using AES.

[Q_R_V]

The nist 5 isn't using AES.

Can't argue with that.

[sp_]

You are the last person to tell that no gpu miner for nist5 exist.

Gpu miners exist, but most of the new coins you see today doesn't support mining with the  JSON-RPC  html protocol (getwork()) in the wallet. This meens that without a pool, you cannot use ccminer or sgminer to solo-mine. I linked 2 new coins and asked for pools, and it turned out that they had pools already.


Here is my original mesage.

These are new coins here are the announcement threads. New coins are a bit risky, but with abit of luck you can make $30 per card per day.

No pools , and no gpu miner (that I know of).

virta-unique-coin:

https://bitcointalk.org/index.php?topic=1966621

coimatic-2:

https://bitcointalk.org/index.php?topic=1961830.0

[abudfv2008]

You are the last person to tell that no gpu miner for nist5 exist.

Gpu miners exist, but most of the new coins you see today doesn't support mining with the  JSON-RPC  html protocol (getwork()) in the wallet. This meens that without a pool, you cannot use ccminer or sgminer to solo-mine. I linked 2 new coins and asked for pools, and it turned out that they had pools already.


Here is my original mesage.

These are new coins here are the announcement threads. New coins are a bit risky, but with abit of luck you can make $30 per card per day.

No pools , and no gpu miner (that I know of).

virta-unique-coin:

https://bitcointalk.org/index.php?topic=1966621

coimatic-2:

https://bitcointalk.org/index.php?topic=1961830.0

And that is what I pointed on. It is lie.
Because Gpu miners exist (for at least a year and you definitely knew that as far as you uses best of them as the basis), pools exist (CTIC2 had the pool from the beginning).

[sp_]

No pools , and no gpu miner (that I know of).

I didn't lie. I didn't find a pool, there are no pools on the announcement thread.

Check for yourself in the first post:

https://bitcointalk.org/index.php?topic=1966621

Gpu miners exist, but most of the new coins you see today doesn't support mining with the  JSON-RPC  html protocol (getwork()) in the wallet. This meens that without a pool, you cannot use ccminer or sgminer to solo-mine. I linked 2 new coins and asked for pools, and it turned out that they had pools already.

[vpixele]

Hi!
Who bought and use sp-mod for bitcore (btx) on 1060 6Gb? Can u write me PM your hashrate? Thx!

[LuisMafixD]

No pools , and no gpu miner (that I know of).

I didn't lie. I didn't find a pool, there are no pools on the announcement thread.

Check for yourself in the first post:

https://bitcointalk.org/index.php?topic=1966621

Gpu miners exist, but most of the new coins you see today doesn't support mining with the  JSON-RPC  html protocol (getwork()) in the wallet. This meens that without a pool, you cannot use ccminer or sgminer to solo-mine. I linked 2 new coins and asked for pools, and it turned out that they had pools already.

Price of Nigt5, I'm Interested! I can not see in your list of private miners.

[PVmining]

Most of alexis work is based on my opensource optimized maxwell ccminer kernels. bla bla bla...

based on your work? lol... your are still the funny reality twisting guy as usual...
https://github.com/alexis78/ccminer => Forked from tpruvot/ccminer

[zijieli]

Hi!
Who bought and use sp-mod for bitcore (btx) on 1060 6Gb? Can u write me PM your hashrate? Thx!

2017-08-08 22:23:47] [S/A/T]: 2/2090/2099, diff: 3.380, 77.35MH/s yes!
[2017-08-08 22:23:52] bitcore block 27340, diff 461.413
[2017-08-08 22:24:10] [S/A/T]: 2/2091/2100, diff: 21.077, 77.35MH/s yes!
[2017-08-08 22:24:13] [S/A/T]: 2/2092/2101, diff: 0.697, 77.35MH/s yes!
[2017-08-08 22:24:20] [S/A/T]: 2/2093/2102, diff: 0.278, 77.35MH/s yes!
[2017-08-08 22:24:22] GPU#0:Zotac GTX 1060 6GB, 12.75MH/s
[2017-08-08 22:24:33] [S/A/T]: 2/2094/2103, diff: 2.481, 77.34MH/s yes!
[2017-08-08 22:24:41] [S/A/T]: 2/2095/2104, diff: 0.236, 77.34MH/s yes!
[2017-08-08 22:24:49] GPU#1:Zotac GTX 1060 6GB, 12.85MH/s
[2017-08-08 22:24:54] bitcore block 27341, diff 461.413
[2017-08-08 22:25:04] [S/A/T]: 2/2096/2105, diff: 0.353, 77.33MH/s yes!
[2017-08-08 22:25:32] [S/A/T]: 2/2097/2106, diff: 0.258, 77.33MH/s yes!
[2017-08-08 22:25:34] Stratum difficulty set to 44.2676 (0.17292)
[2017-08-08 22:25:34] bitcore block 27342, diff 461.413
[2017-08-08 22:25:39] [S/A/T]: 2/2098/2107, diff: 0.721, 77.32MH/s yes!
[2017-08-08 22:25:51] [S/A/T]: 2/2099/2108, diff: 0.461, 77.32MH/s yes!
[2017-08-08 22:26:06] [S/A/T]: 2/2100/2109, diff: 0.204, 77.31MH/s yes!
[2017-08-08 22:26:09] [S/A/T]: 2/2101/2110, diff: 0.220, 77.31MH/s yes!
[2017-08-08 22:26:12] [S/A/T]: 2/2102/2111, diff: 0.392, 77.31MH/s yes!
[2017-08-08 22:26:27] [S/A/T]: 2/2103/2112, diff: 0.201, 77.31MH/s yes!
[2017-08-08 22:26:31] GPU#2:Zotac GTX 1060 6GB, 12.87MH/s
[2017-08-08 22:26:43] bitcore block 27343, diff 461.413
[2017-08-08 22:26:51] [S/A/T]: 2/2104/2113, diff: 0.463, 77.30MH/s yes!
[2017-08-08 22:27:14] [S/A/T]: 2/2105/2114, diff: 0.846, 77.30MH/s yes!
[2017-08-08 22:27:15] [S/A/T]: 2/2106/2115, diff: 0.846, 77.30MH/s yes!
[2017-08-08 22:27:18] [S/A/T]: 2/2107/2116, diff: 9.713, 77.30MH/s yes!
[2017-08-08 22:27:31] [S/A/T]: 2/2108/2117, diff: 0.332, 77.30MH/s yes!

[sp_]

Most of alexis work is based on my opensource optimized maxwell ccminer kernels. bla bla bla...

based on your work? lol... your are still the funny reality twisting guy as usual...
https://github.com/alexis78/ccminer => Forked from tpruvot/ccminer

The tpruvot fork is full of my kernel optimalizations, but instead of approving my pullrequests, he copied my work. We have had a paralell opensource fork for years. The tpruvot fork 2.1  has a bether API implementation, more stable stratum code, bether support for the NVSMI, and bugfixes. But slower kernels. I forked version 1.5 and have included some of the fixes.

[Zean]

sp_
check pm plz~

[sp_]

If you want to know some ccminer history github is a nice place to check.

Here is a commit back in nov 2014: (faster echo512)

https://github.com/sp-hash/ccminer/commit/3cc7e9db091b277d9357b20f2d9101846bee54dc

Here is another commit Jan 2015 (Faster groestl512)

https://github.com/sp-hash/ccminer/commit/1c61d0eebb1ee49bcdb137d5c7291fc0fb3e810d

Every algo have been optimized by me, and I have done it opensource.

If you see the latest alexis fork, he base his work on the same code.

So I did the work, Eplyson copied my work, and Alexis reforked my work. And then They bitch about it when I add some more speed and refuse to share my changes..

[HeroKung]

I am working on Skunk #6. Look's like 35MHASH is possible. ccminer 2.1 does 35MHASH on the gtx 1080ti, so my code will turn your cheap 1070 card into a 1080ti.

With proper settings your gtx 1070 should be able to do this with the released bins (sp-mod 5):

https://i.gyazo.com/dfaebe590821806de4869310efe544b4.png

Since my work has been stolen and spread, I will include a 2% fee in build #6

waiting  for free download with 1-2 % fee version

[mystere_miner]

So I did the work, Eplyson copied my work, and Alexis reforked my work. And then They bitch about it when I add some more speed and refuse to share my changes..

Well, that is because the work is GPL'd, so.. yeah.  That's the whole point of the GPL.  

If you want to fork your own code and take it private, that's fine, but if you base your work on others who do not agree, you have to re-implement their work from scratch or you are infringing their copyright.  Even if it's 1 line of code.

Do I really have to explain how the GPL works to you?

I'm not judging, per se.  Yeah, it's basically unenforceable for most people.  But ethically speaking, it is unethical.  Then you complain that people steal your code and re-sell it, which is something that is both legally and ethically allowed by the GPL.

Maybe you should write your own miner from scratch if you want full ethical control over your code.

[brownfly]

He-he, yeah, GPL is a bitch for proprietary products. It was a reason why Apple took FreeBSD for their new OS X. sorry for offtopic.

[pallas]

He-he, yeah, GPL is a bitch for proprietary products. It was a reason why Apple took FreeBSD for their new OS X. sorry for offtopic.

[Singman33]

Most of alexis work is based on my opensource optimized maxwell ccminer kernels. bla bla bla...

based on your work? lol... your are still the funny reality twisting guy as usual...
https://github.com/alexis78/ccminer => Forked from tpruvot/ccminer

The tpruvot fork is full of my kernel optimalizations, but instead of approving my pullrequests, he copied my work. We have had a paralell opensource fork for years. The tpruvot fork 2.1  has a bether API implementation, more stable stratum code, bether support for the NVSMI, and bugfixes. But slower kernels. I forked version 1.5 and have included some of the fixes.

I should agree with SP_ on that. I'm following and using many forks of ccminers since the start of cryptos and I always saw tpruvot and sp_ "swapping" their patches.
Now, SP_ choose to be closed source (something I could understand when you see Claymore's success and wallets....) but I'm still using open sourced versions for x11,x13,keccak,x15, noescrypt and so on...
If you run his miner, you will see this message :

  Based on pooler cpuminer 2.3.2 and the tpruvot@github fork
  CUDA support by Christian Buchner, Christian H. and DJM34
  Includes optimizations implemented by sp, klaust, tpruvot, tsiv and pallas.

[brownfly]

He-he, yeah, GPL is a bitch for proprietary products. It was a reason why Apple took FreeBSD for their new OS X. sorry for offtopic.

https://developer.apple.com/library/content/documentation/Darwin/Conceptual/KernelProgramming/BSD/BSD.html

The BSD portion of the OS X kernel is derived primarily from FreeBSD, a version of 4.4BSD that offers advanced networking, performance, security, and compatibility features.

blame Apple, not me.

[mystere_miner]

Hey all you mining Signatum - Only 85 Minutes left to get those 2500 coin block rewards.  After that, only 1250.

[bensam1231]

Higher value cards depreciate more then lower value cards. They don't retain their price better. That's the way halo products work as they aren't a good deal when you buy them and they're even less of a good deal when you're buying them used as there are new products on the market which supersede them for cheaper. Sweet spot is always midrange for new and used from a consumer standpoint.

As far as mining goes, 1080/tis don't scale properly in a lot of algorithms and are crippled in memory hard algos such as Dagger from GDDR5 X. They have heat dissipation issues due to the size of the chips and will throttle a lot faster then a 1060/70 on algos they do work properly on. The size of the heatsinks required to cool them properly are in the three slot variety. They also consume more power which means you need to have a beefier power distribution system for each computer, which gets even more expensive the bigger the PSU.

Did I mention they can't really mine dagger? That's a pretty big deal. It wouldn't surprise me if we're back dual mining Eth again on Nvidia hardware. That already happened earlier this year. Also why AMD hardware is just outright better as far as investment goes for mining (if we could get it remotely around MSRP). For comparison a 580 is making $2.51 dual mining and a 1070 is making $2.35 mining Zcash. Actually looking at prices right now, you can get 580s for $350... So it still is a better deal. As hardware values come down from their bloated levels it'll become a even better deal.

It blows my mind people still can't compare Nvidia and AMD properly. Guess they just see the price tag for a 'mid-low end' card and they can't put it into context when it comes to mining, just playing games.

Other things in the crypto scene being more beneficial to do is outside the scope of buying hardware. We can discuss how much money Claymore is making and the fact he doesn't have competition or that we still don't have the equivalent of Claymore for Nvidia.

LUL AMD cards. With Nvidia you can mine a hole slew of other algorithms that AMD can't. It doesn't even matter if ETH isn't the best to mine.
You mine what's most profitable and convert it to whatever coin you want. SP is our Claymore equivalent for Nvidia. SP is even better than Claymore.
With his mods you have higher profitability than ETH or any of those coin listed on Whattomine.

Just like your picture you're a dumbass. What's the point of mining coins that have 10k in volume? Do you have any idea how fast low volume coins get saturated and destroyed by hashrate? Nvidia hashrate is already bloated, hence why we got all of maybe 3 days of good SIGT mining before that was done. Also why all the 'niche' algos are now equalizing around Equihash algos. Also why people in this thread are now complaining about ROI on their .05BTC investment in the miner. It doesn't matter how many algos you can mine if none of them have the volume to support the hardware mining it.

No shit. I'm the one that started the 'mine what's most profitable, invest in what you want to keep' rhetoric you're spouting after I got tired of people mindlessly mining crap coins as 'supporting the coin'. You can look through this thread to find that. Also in no way, shape, or form related to good hardware purchases or what we're talking about.

SP is not the equivalent of Claymore. He does what's easy and then doesn't give a shit when it gets hard. He doesn't offer improvements outside of basic miner functionality and some speed improvements. That's why a lot of people don't like SP as he only 'optimizes' and nothing else. He doesn't make algos from the ground up either, only if something already exists.

I've supported SP for years, but saying something like that is completely wrong. I understand you came in with the Ethereum boat and you thought Nvidia was a great 'investment', cause it's the underdog, but that would be incorrect. There is a reason like 80% of the hashrate in the GPU Crypto scene is AMD. Mining is all about long term, not short term 'oh I made $2000 in 3 days!' it's about what you make for the other 361 days in the year.

Nvidia is ahead of amd already almost everywhere. Even on eth, AMD began to lose speed, unlike nvidia. amd Soon amd will die.

Ahead 'everywhere', how about some sources for that... Like coins and earnings. Yup, and AMD announced they're fixing it in a driver along with Claymore also working on fixes for the 480/580 hashrate bug. Completely blown out of proportion. You sure know what you're talking about.

Wow rude. SP has done great work for us. He has mods with significant improvements for almost every algorithm.
Sigt and Spreadcoin have been the most profitable coins to mine for almost a month now, but hey keep mining your ETH.
Nvidia gpus are still the best, they beat AMD in both the Ethash (32MH/s 1070) and Equihash (740 H/s 1080 Ti) algorithms.
Until we see the performance of RX Vega, Nvidia is still king.

Rude in what way? I've supported SP for years. I'm happy he increases speed as I've invested a lot of money in Nvidia, but saying he's a Claymore is just completely wrong. Claymore offers professional products with support. SP you don't know if you're going to even get the next version you paid for or he'll add a fee or he wont even release another version or he wont fix bugs. That's completely putting aside that he doesn't work on things he doesn't want to, like features, hence why when another dev bitches about third party code he'll just stop producing whatever product he's working on instead of either making a system that takes into account GPL licensing or rewriting parts of the code he needs to.

Why do you think we're still stuck with a paid version instead of a fee? Because he doesn't spend any amount of time securing his code against people tearing the dev fee out... cause he doesn't want to. A hobby dev is a far cry from a professional that does this for a living, like Claymore.

You're way too new to be lecturing me on how to treat devs. I've donated, I've bought, and I've supported. GTFO if you aren't even going to talk about the original subject matter and instead are turning this into a emotional debate on 'supporting the devs'. Remember we were talking about 1060/70s vs 1080/tis? Then also they're all a bad investment compared to AMD.

Also you aren't getting 32Mhs on a 1070. At +1000 mem which isn't remotely realistic you'll hit about 30Mh. A good solid memory OC for eth will be +500. Also why would you randomly throw a 1080ti in there when you're comparing a 1070? 'Hey look this card gets Y amount of hashrate at X dollars and this card gets Z amount at D dollars'. Good comparison. Lets add in that Nvidia did not get the assembly optimizations for dual mining for Claymore, which means it really doesn't matter what the 'raw' numbers are as any minor worth his salt should be dual mining... And AMD hardware gets about 2x the amount of speed on secondary algos.

Sigt is the flavor of the week. It's dead as Skein is... or as profitable as Equihash. Spreadcoin is semiprofitable, but whenever SP pushes his custom miner that equalizes or dips below Equihash. Niche coins for Nvidia are only as profitable as their obscurity. Hence why I talk about Equihash, because that's a good, solid, long running coin, much like all the breadwinners for AMD. Putting that aside AMD also has it's own niche coins as well. Weird people still don't know how that works. They tell their friends about a niche coin > their friends steal their earnings > their friends tell other friends who also steal their earnings > the coin now makes as much as the lowest common strong supporting algo (Equihash) and it doesn't matter anymore.

Nvidia isn't the king of anything bro, it's pretty easy to see that. That's just your gamer bias talking. If it were, AMD wouldn't be dominating the mining scene both in numbers, but also long term profitability and support. Like I said, it's not about what you earn 3 days out of the year, it's what you earn the other 362.

Oh and nice assuming I'm a AMD miner. I was here far before you and was invested in Nvidia before it became the 'cool thing cause screw AMD'. I'm about 90% Nvidia and 10% AMD right now. Weird that I can maintain perspective on things while being invested in it, no?