Bitcoin Forum
November 15, 2024, 02:12:20 AM *
News: Check out the artwork 1Dq created to commemorate this forum's 15th anniversary
 
   Home   Help Search Login Register More  
Pages: « 1 ... 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 [193] 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 ... 1240 »
  Print  
Author Topic: CCminer(SP-MOD) Modded GPU kernels.  (Read 2347580 times)
sp_ (OP)
Legendary
*
Offline Offline

Activity: 2954
Merit: 1087

Team Black developer


View Profile
July 02, 2015, 09:16:11 AM
Last edit: July 02, 2015, 09:33:59 AM by sp_
 #3841

Yes, the code needs to be changed in order to get a boost. The problem of the lyra algorithm is that it is memory hard, buffers are bigger than the level1 cache of the gpu, so the program constantly fetch memory to the cache. DJM has made a private version wich does 3.2MHASH on the 780ti. He has recoded everything.


The sp-mod is at least doing bether than the new Furyx (here is a 100$ 1 gb NVIDIA 750 card with overclocking and a modded bios). Powerusage 40 watt



AMD furyx lyra hashrates. (source cryptomining blog)

– Lyra2RE default: 287 KHS
– Lyra2RE Pallas Mod: 450 KHS

Team Black Miner (ETHB3 ETH ETC VTC KAWPOW FIROPOW EVRPROGPOW MEOWPOW + dual mining + tripple mining.. https://github.com/sp-hash/TeamBlackMiner
djm34
Legendary
*
Offline Offline

Activity: 1400
Merit: 1050


View Profile WWW
July 02, 2015, 11:05:26 AM
 #3842

Submitted another speedup in quark. More is comming.
I will improve the bandwidth of the hashes since half of the buffers are not used. Then the intensity can be increased..


UNUSED BUFFERS--

Do unused buffers have anything to do with the poor performance of GTX 960/970 relative to GTX 750ti on Lyra2?  Or is it the memory controller?  Or both?  My 4GB GTX 960 is slower than a 2GB 750ti mining Lyra2, and it just indicates improper coding to match the card's capacity.  Both cards are on a Win 7 x64 system.

DJM34, don't be afraid to chime in, you are the code master of Lyra2.       --scryptr
being in discussion with vtc team, can't really say anything  Grin

       some progress on neoscrypt:  ~850kH/s on gtx980
                                                ~500kh/s on gtx780ti
                                                  201kh/s on gtx750

djm34 facebook page
BTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze
Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
sp_ (OP)
Legendary
*
Offline Offline

Activity: 2954
Merit: 1087

Team Black developer


View Profile
July 02, 2015, 11:38:11 AM
 #3843

being in discussion with vtc team, can't really say anything  Grin

Even with a kernal that does 3MHASH@lyra2re on the gtx 980, mining quark@19MHASH is 230% more profitable with the current rates.

Team Black Miner (ETHB3 ETH ETC VTC KAWPOW FIROPOW EVRPROGPOW MEOWPOW + dual mining + tripple mining.. https://github.com/sp-hash/TeamBlackMiner
AliMan
Hero Member
*****
Offline Offline

Activity: 2198
Merit: 502


DGbet.fun - Crypto Sportsbook


View Profile
July 02, 2015, 12:06:07 PM
 #3844

Speaking of quark, thought I'd mention that the fastest is still just release 51 for 970/980's. With 3 970's and 2 980's, I'm getting 92.6-93 mh/s compared to a max of 92.5 mh/s on the newer releases.

myagui
Legendary
*
Offline Offline

Activity: 1154
Merit: 1001



View Profile
July 02, 2015, 02:26:10 PM
 #3845

@sp_, Epsylon3, djm34, tsiv & others:
Bounty for improved open source Monero GPU miner is opening up again, including partial bounty awards for progressive performance bumps.
https://bitcointalk.org/index.php?topic=656841.msg11772611#msg11772611

Given the performance of existing ccminer (including private version going by the figures earlier posted), and taking into account the effective load on the 750TI while hashing this, there is possibly shitloads of optimizations to be had. In my mind, one quick path to improved performance, is in the lines of shrinking the scratchpad and running extra threads (as mentioned the other day, something like -L in the traditional cudaminer).

djm34
Legendary
*
Offline Offline

Activity: 1400
Merit: 1050


View Profile WWW
July 02, 2015, 02:36:46 PM
Last edit: July 02, 2015, 02:47:40 PM by djm34
 #3846

@sp_, Epsylon3, djm34, tsiv & others:
Bounty for improved open source Monero GPU miner is opening up again, including partial bounty awards for progressive performance bumps.
https://bitcointalk.org/index.php?topic=656841.msg11772611#msg11772611
well "partial bounty"... they come up with their unrealistic number, we come with our hard work, they give you only a fraction of what they collected and keep the rest for themselve: NO THANKS  Grin

ps: remember they didn't even want to give the bounty to tsiv

ps2: proper way to do thing, you hire a dev  Grin

ps3: I give a bounty for a 300GH/s sha256 for 750ti, fractional bounty  Grin
if it is ok for you to work for free for 2 weeks, this challenge is for you

djm34 facebook page
BTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze
Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
myagui
Legendary
*
Offline Offline

Activity: 1154
Merit: 1001



View Profile
July 02, 2015, 03:07:34 PM
Last edit: July 02, 2015, 05:07:58 PM by myagui
 #3847

Hey djm34,  Roll Eyes

I understand your logic, but I really don't agree with your take on working for a bounty.

- If the bounty is too low for the possible measure of work, then the bounty is just too low, so it needs to be raised in order to attract developers. Different people will naturally value their time differently, for plenty of reasons and not just the obvious one (quality work = expensive work).

- There isn't someone keeping the bounty for themselves in case full performance is not delivered, those that funded the bounty will simply not have the full expense. It is a community organized service, paid with community funds. Being for an open source release, it would stand to benefit the Monero (and CN) community at large.

Hiring a dev is surely another way to approach the problem, but that also has it's own issues, which you're probably entirely familiar with, even though you'd probably sit on the receiving end of that deal  Wink
Personally, and funds permitting, I tend to prefer the hiring dev approach, but others might feel differently.

PS: I recall that there was some delay in getting the bounty over to TSIV, but it was never the case that the people with the bounty offer did not want to pay him. The discussion (and delay) around the time, concerned performance and support for various compute levels, etc...

Edit:
ps3: I give a bounty for a 300GH/s sha256 for 750ti, fractional bounty  Grin
if it is ok for you to work for free for 2 weeks, this challenge is for you

I wanted to take you up on this bounty, but your offer would only cover an estimate of 0 minutes of my talents, so, no go amigo.
I suggest you raise your offer, say perhaps, > 0?  Grin

bensam1231
Legendary
*
Offline Offline

Activity: 1764
Merit: 1024


View Profile
July 02, 2015, 04:27:11 PM
 #3848

being in discussion with vtc team, can't really say anything  Grin

Even with a kernal that does 3MHASH@lyra2re on the gtx 980, mining quark@19MHASH is 230% more profitable with the current rates.

Pretty much... market isn't fair even with improvements. Neoscrypt is quite lucrative right now as well.

Not sure why you guys sorta neglect Cryptonote as it has a lot of market volume every day (XMR). I know SP you said you have a private miner for it, but it's not much better then the public one except for power numbers.


AMD furyx lyra hashrates. (source cryptomining blog)

– Lyra2RE default: 287 KHS
– Lyra2RE Pallas Mod: 450 KHS

That's the way all new cards are till people figure out the architecture. Guessing none of the devs own one and no one is developing for it just yet. We probably wont see well designed kernels for it for a few months as there is no money to be made for designing for it either as no one owns them.

@sp_, Epsylon3, djm34, tsiv & others:
Bounty for improved open source Monero GPU miner is opening up again, including partial bounty awards for progressive performance bumps.
https://bitcointalk.org/index.php?topic=656841.msg11772611#msg11772611

Given the performance of existing ccminer (including private version going by the figures earlier posted), and taking into account the effective load on the 750TI while hashing this, there is possibly shitloads of optimizations to be had. In my mind, one quick path to improved performance, is in the lines of shrinking the scratchpad and running extra threads (as mentioned the other day, something like -L in the traditional cudaminer).

And literally just read this on the next page after typing out my response.

Also agree, the XMR market is huge... It's one of the biggest coins out there. I mined for quite awhile on Claymore with my AMD cards, it's well worth looking into as another algo to add to the Nvidia miner arsenal.

Hey djm34,  Roll Eyes

I understand your logic, but I really don't agree with your take on working for a bounty.

- If the bounty is too low for the possible measure of work, then the bounty is just too low, so it needs to be raised in order to attract developers. Different people will naturally value their time differently, for plenty of reasons and not just the obvious one (quality work = expensive work).

- There isn't someone keeping the bounty for themselves in case full performance is not delivered, those that funded the bounty will simply not have the full expense. It is a community organized service, paid with community funds. Being for an open source release, it would stand to benefit the Monery (and CN) community at large.

Hiring a dev is surely another way to approach the problem, but that also has it's own issues, which you're probably entirely familiar with, even though you'd probably sit on the receiving end of that deal  Wink
Personally, and funds permitting, I tend to prefer the hiring dev approach, but others might feel differently.

Also agree with this. I don't think there is anything wrong with offering bounties say for percentage gains on current miners. If you don't think it's worth your time or it's not cost effective you simply don't do it. Having milestone bounties for development encourages prolonged development which is really no different then receiving a paycheck for your time. Although depending on the algo, gains may come a lot easier in some then others. First percents will yield the most amount of money for the least amount of work as optimizing code isn't linear.

This is actually approaching what we like to call a 'business model'. It's where the rest of the professionals work in the real world. There really isn't enough competition to be had in the crypto kernel development to actually make people want to try, so they don't. As I mentioned a company could really shake this up, unfortunately I don't know enough about the software to take on something like that. There is definitely plenty of money to be made here. Large chinaminers probably already have something like that.

I buy private Nvidia miners. Send information and/or inquiries to my PM box.
djm34
Legendary
*
Offline Offline

Activity: 1400
Merit: 1050


View Profile WWW
July 02, 2015, 09:31:27 PM
Last edit: July 02, 2015, 10:42:35 PM by djm34
 #3849


Also agree with this. I don't think there is anything wrong with offering bounties say for percentage gains on current miners. If you don't think it's worth your time or it's not cost effective you simply don't do it. Having milestone bounties for development encourages prolonged development which is really no different then receiving a paycheck for your time. Although depending on the algo, gains may come a lot easier in some then others. First percents will yield the most amount of money for the least amount of work as optimizing code isn't linear.

This is actually approaching what we like to call a 'business model'. It's where the rest of the professionals work in the real world. There really isn't enough competition to be had in the crypto kernel development to actually make people want to try, so they don't. As I mentioned a company could really shake this up, unfortunately I don't know enough about the software to take on something like that. There is definitely plenty of money to be made here. Large chinaminers probably already have something like that.
no sure where or how you work... but if you work 15days for 10% of your monthly paycheck then it means you are getting screwed.  Grin
(but yes for someone who makes quote based on last year btc price, or make up numbers... I am not entirely surprised you find that attractive  Grin)
As you say development isn't linear, meaning to get to the first 10% hashrate increase it will cost you a lot more time and effort and this won't be rewarded...
and also we don't know if the target for the full reward is realistic either... (that's why I was saying in the example I was giving, asking for 300GH/s sha256 of a 750ti... )

So yes this isn't a pretty good model and not worth my time (unless I had some time to lose which isn't the case at the moment).


djm34 facebook page
BTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze
Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
GingerAle
Legendary
*
Offline Offline

Activity: 1260
Merit: 1008


View Profile WWW
July 03, 2015, 05:14:09 AM
 #3850

Then what is a good model? I kind of understand the downside of the bounty model - made up target numbers by peeps (like me) that have no idea wtf goes into the work. But the "hiring a dev" also has its potential problem of the dev not being able to deliver (for whatever reason). The Monero community is still working through one such instance. And this all, of course, is wrapped up in the fact that the cryptonight bounty proposed is for an improved open source GPU miner, because Monero people want to bump the hashrate in order to secure the network.... pocketing more coin for themselves is not the goal, because optimized mining software that is publicly released won't provide any financial advantage to the miners. OK, perhaps for a week.

I was trying to make the 2X-hashrate-increase-on-existing-hardware bounty model such that lots of individuals would put into the bounty, creating a ridiculous bounty for the 2X target. Something crazy, like 15,000 XMR. Because I figure at some bounty level, it does become an attractive incentive, no matter how flawed the funding model. But, unfortunately, we're not there yet.

I mean, should we just contact tsiv and try hiring him? He seems to have working knowledge of cryptonight and maxwell architecture.

< Track your bitcoins! > < Track them again! > <<< [url=https://www.reddit.com/r/Bitcoin/comments/1qomqt/what_a_landmark_legal_case_from_mid1700s_scotland/] What is fungibility? >>> 46P88uZ4edEgsk7iKQUGu2FUDYcdHm2HtLFiGLp1inG4e4f9PTb4mbHWYWFZGYUeQidJ8hFym2WUmWc p34X8HHmFS2LXJkf <<< Free subdomains at moneroworld.com!! >>> <<< If you don't want to run your own node, point your wallet to node.moneroworld.com, and get connected to a random node! @@@@ FUCK ALL THE PROFITEERS! PROOF OF WORK OR ITS A SCAM !!! @@@@
bensam1231
Legendary
*
Offline Offline

Activity: 1764
Merit: 1024


View Profile
July 03, 2015, 07:19:57 AM
 #3851

no sure where or how you work... but if you work 15days for 10% of your monthly paycheck then it means you are getting screwed.  Grin
(but yes for someone who makes quote based on last year btc price, or make up numbers... I am not entirely surprised you find that attractive  Grin)
As you say development isn't linear, meaning to get to the first 10% hashrate increase it will cost you a lot more time and effort and this won't be rewarded...
and also we don't know if the target for the full reward is realistic either... (that's why I was saying in the example I was giving, asking for 300GH/s sha256 of a 750ti... )

I don't think you understand how commissions work. You get paid for 10% of the job. The first 10% for instance will be the easiest as often times optimizing has diminishing returns when it comes to software(more time you put in, less you get back). That aside, you are essentially responsible for your own time. No one is forcing you to take the commission nor do you need to double the hashrate of the miner, you just need to enhance it by a certain percentage and it's pretty easy to do that without doing a full rewrite. If you were aiming for the full bounty and you thunk 'I can do it better' then maybe you'd have to go in balls deep, but at that point you'll get the whole reward anyway if you reach the goal.

You're essentially working 10% for 10% of 'your paycheck' because you never worked for the full paycheck, you just worked for a portion of it. It's a percentage, and the easiest percentage is at the beginning (which they didn't take into account).

I was short on sleep, I used 500 as a baseline instead of 200 on accident, I already explained this. You can eat a dick if you don't believe that, you don't even understand percentages so I don't expect you to understand that either. This is like explaining business to a teen. A clerical error in a rant on BCT isn't the same thing as not being able to understand the fundamentals of business.

Quote
So yes this isn't a pretty good model and not worth my time (unless I had some time to lose which isn't the case at the moment).

Good for you, apparently $1500 isn't worth a 40 hour work week, probably not even that if you're as good as you make yourself out to be. There are plenty of programmers that get paid less, especially in other countries outside the US. Also why I mentioned kernel development here needs more competition. Some of you are pretty laid back... in a multitude of areas and want to get paid premium dollars for that. Meanwhile SP releases his work open source and you just build on top of it.

This is exactly why you guys don't advertise your services. You wouldn't want anyone to catch on just how lucrative this work is and there is plenty of programmers that are qualified in C++/Cuda/OCL, just not familiar with BCT or mining.

Then what is a good model? I kind of understand the downside of the bounty model - made up target numbers by peeps (like me) that have no idea wtf goes into the work. But the "hiring a dev" also has its potential problem of the dev not being able to deliver (for whatever reason). The Monero community is still working through one such instance. And this all, of course, is wrapped up in the fact that the cryptonight bounty proposed is for an improved open source GPU miner, because Monero people want to bump the hashrate in order to secure the network.... pocketing more coin for themselves is not the goal, because optimized mining software that is publicly released won't provide any financial advantage to the miners. OK, perhaps for a week.

I was trying to make the 2X-hashrate-increase-on-existing-hardware bounty model such that lots of individuals would put into the bounty, creating a ridiculous bounty for the 2X target. Something crazy, like 15,000 XMR. Because I figure at some bounty level, it does become an attractive incentive, no matter how flawed the funding model. But, unfortunately, we're not there yet.

I mean, should we just contact tsiv and try hiring him? He seems to have working knowledge of cryptonight and maxwell architecture.

Their model is 'I work on whatever I feel like it, whenever I feel like it, then I sell it to the highest bidder behind the scenes when no one is looking. If you don't have enough money or I don't like you enough (which go hand in hand), tough shit and I'll run you out of town cause you can't compete'.

Someone really needs to make a company for this. This is very comparable to what happened with ASICs only using software as hardware so there aren't any startup costs. Literally no one would need to spend seed money on this besides hiring some devs for maybe a month till they can spit out a product that can be sold or leased. Since a lot of the backend work is already done (working miners, lots of open sources, good examples and documentation, infrastructure is there), it could roll out in probably two weeks with a 10 person staff structured around a couple high rolling algorithms. Probably targeting for x11, Cryptonote, and Quark based on popularity would get the biggest amount of miners on board. Add a 5% variable fee which could be walked down. A lot of basic DRM could be done that's a no brainer to keep people on board, regular releases and features would keep people on board anyway.

There IS NO COMPETITION here! Literally anyone could come in and swipe this out, but most people don't even know this is a fundamental part of mining yet. They just download whatever miner pops up and think they're set because these guys don't advertise. There is a ton of potential hashrate sitting there to be used for profit. The risk is quite minimal as well compared to getting into mining. There will always be people mining, regardless of what they're mining on and what BTC is at. So you basically need CUDA and OCL programmers, then someone to write a front end for it (or you simply ask someone like Nwolls on that already has one done).


Honestly it shouldn't matter who 'reaps the most from this'. DJM seems to be stuck on this note of people profiting off his work, even though he's selling it. It's two different things. Do gas stations get pissed off because oil tycoons or truck companies make so much money? No. They're a gas station, that's not their MO. If they had the resources to be a oil tycoon they probably wouldn't be a gas station anymore... or the oil tycoons simply buy up the gas stations.

I buy private Nvidia miners. Send information and/or inquiries to my PM box.
sp_ (OP)
Legendary
*
Offline Offline

Activity: 2954
Merit: 1087

Team Black developer


View Profile
July 03, 2015, 07:24:10 AM
Last edit: July 03, 2015, 07:34:34 AM by sp_
 #3852

My private cryptonight kernal is around 10% faster on windows(windows executable). It can be yours for a 0.2 BTC donation. Then you also get my private spreadcoin miner with sourcecode (linux compatible).. (8.75-10% faster)


Team Black Miner (ETHB3 ETH ETC VTC KAWPOW FIROPOW EVRPROGPOW MEOWPOW + dual mining + tripple mining.. https://github.com/sp-hash/TeamBlackMiner
djm34
Legendary
*
Offline Offline

Activity: 1400
Merit: 1050


View Profile WWW
July 03, 2015, 08:07:04 AM
Last edit: July 03, 2015, 09:03:23 AM by djm34
 #3853

no sure where or how you work... but if you work 15days for 10% of your monthly paycheck then it means you are getting screwed.  Grin
(but yes for someone who makes quote based on last year btc price, or make up numbers... I am not entirely surprised you find that attractive  Grin)
As you say development isn't linear, meaning to get to the first 10% hashrate increase it will cost you a lot more time and effort and this won't be rewarded...
and also we don't know if the target for the full reward is realistic either... (that's why I was saying in the example I was giving, asking for 300GH/s sha256 of a 750ti... )

I don't think you understand how commissions work. You get paid for 10% of the job. The first 10% for instance will be the easiest as often times optimizing has diminishing returns when it comes to software(more time you put in, less you get back). That aside, you are essentially responsible for your own time. No one is forcing you to take the commission nor do you need to double the hashrate of the miner, you just need to enhance it by a certain percentage and it's pretty easy to do that without doing a full rewrite. If you were aiming for the full bounty and you thunk 'I can do it better' then maybe you'd have to go in balls deep, but at that point you'll get the whole reward anyway if you reach the goal.

You're essentially working 10% for 10% of 'your paycheck' because you never worked for the full paycheck, you just worked for a portion of it. It's a percentage, and the easiest percentage is at the beginning (which they didn't take into account).

I was short on sleep, I used 500 as a baseline instead of 200 on accident, I already explained this. You can eat a dick if you don't believe that, you don't even understand percentages so I don't expect you to understand that either. This is like explaining business to a teen. A clerical error in a rant on BCT isn't the same thing as not being able to understand the fundamentals of business.

Quote
So yes this isn't a pretty good model and not worth my time (unless I had some time to lose which isn't the case at the moment).

Good for you, apparently $1500 isn't worth a 40 hour work week, probably not even that if you're as good as you make yourself out to be. There are plenty of programmers that get paid less, especially in other countries outside the US. Also why I mentioned kernel development here needs more competition. Some of you are pretty laid back... in a multitude of areas and want to get paid premium dollars for that. Meanwhile SP releases his work open source and you just build on top of it.

This is exactly why you guys don't advertise your services. You wouldn't want anyone to catch on just how lucrative this work is and there is plenty of programmers that are qualified in C++/Cuda/OCL, just not familiar with BCT or mining.

Then what is a good model? I kind of understand the downside of the bounty model - made up target numbers by peeps (like me) that have no idea wtf goes into the work. But the "hiring a dev" also has its potential problem of the dev not being able to deliver (for whatever reason). The Monero community is still working through one such instance. And this all, of course, is wrapped up in the fact that the cryptonight bounty proposed is for an improved open source GPU miner, because Monero people want to bump the hashrate in order to secure the network.... pocketing more coin for themselves is not the goal, because optimized mining software that is publicly released won't provide any financial advantage to the miners. OK, perhaps for a week.

I was trying to make the 2X-hashrate-increase-on-existing-hardware bounty model such that lots of individuals would put into the bounty, creating a ridiculous bounty for the 2X target. Something crazy, like 15,000 XMR. Because I figure at some bounty level, it does become an attractive incentive, no matter how flawed the funding model. But, unfortunately, we're not there yet.

I mean, should we just contact tsiv and try hiring him? He seems to have working knowledge of cryptonight and maxwell architecture.

Their model is 'I work on whatever I feel like it, whenever I feel like it, then I sell it to the highest bidder behind the scenes when no one is looking. If you don't have enough money or I don't like you enough (which go hand in hand), tough shit and I'll run you out of town cause you can't compete'.

Someone really needs to make a company for this. This is very comparable to what happened with ASICs only using software as hardware so there aren't any startup costs. Literally no one would need to spend seed money on this besides hiring some devs for maybe a month till they can spit out a product that can be sold or leased. Since a lot of the backend work is already done (working miners, lots of open sources, good examples and documentation, infrastructure is there), it could roll out in probably two weeks with a 10 person staff structured around a couple high rolling algorithms. Probably targeting for x11, Cryptonote, and Quark based on popularity would get the biggest amount of miners on board. Add a 5% variable fee which could be walked down. A lot of basic DRM could be done that's a no brainer to keep people on board, regular releases and features would keep people on board anyway.

There IS NO COMPETITION here! Literally anyone could come in and swipe this out, but most people don't even know this is a fundamental part of mining yet. They just download whatever miner pops up and think they're set because these guys don't advertise. There is a ton of potential hashrate sitting there to be used for profit. The risk is quite minimal as well compared to getting into mining. There will always be people mining, regardless of what they're mining on and what BTC is at. So you basically need CUDA and OCL programmers, then someone to write a front end for it (or you simply ask someone like Nwolls on that already has one done).


Honestly it shouldn't matter who 'reaps the most from this'. DJM seems to be stuck on this note of people profiting off his work, even though he's selling it. It's two different things. Do gas stations get pissed off because oil tycoons or truck companies make so much money? No. They're a gas station, that's not their MO. If they had the resources to be a oil tycoon they probably wouldn't be a gas station anymore... or the oil tycoons simply buy up the gas stations.
stop your insane one page spamming...
oh and again your numbers, as usual, do not make sense (where did you see 1500$ for a week ? 40 hours... ) you are clearly not awake... still dreaming  
so it isn't last time, it is all the time  Grin
10% of the bounty is most likely 0.2-0.4btc, ok you are still in unicorn world where monero team proposed a 15k $ bounty (not sure what you used that time for btc either...) may-be one day, you will make a post when you are awake...

and for your info because you seem to invent really a lot of stuff this is how I work:
someone wants to get some work done on something,  he contacts me (or I contact him) we agree on a price and I work on it... that's it...

anyhow, might have a look it, but not now, because I don't have time and the bounty system obliged me (since I am not hired) to do that only on my free time... so long for your business model (as usual).

ps: Have you consider making of your trolling a business model ?
Type page long posts and being paid on a word basis... you will make 10k $ per post (I might not be awake, but that's ok  Grin)

djm34 facebook page
BTC: 1NENYmxwZGHsKFmyjTc5WferTn5VTFb7Ze
Pledge for neoscrypt ccminer to that address: 16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw
CapnBDL
Hero Member
*****
Offline Offline

Activity: 1064
Merit: 500


MOBU


View Profile WWW
July 03, 2015, 08:28:25 AM
 #3854

I must be bored, just read the whole previous post. Their fingers musta been tired after all that keyboarding.    Grin

           ▄██▄
████▄    ▄██████▄    ▄████
██████▄ ▀████████▀ ▄██████
▐███████▄ ▀████▀ ▄███████▌
▐█████████▄ ▀▀ ▄█████████▌
 ████ ▀█████▄▄█████▀ ████
 ████ ▄ ▀████████▀ ▄ ████
 ▐███ ██▄ ▀████▀ ▄██ ███▌
 ▐███ ████▄ ▀▀ ▄████ ███▌
  ███ ██████▄▄██████ ███
  ███ ██████████████ ███
  ▐██ ██████████████ ██▌
  ▐██ ██████████████ ██▌
   ██ ██████████████ ██
       ▀██████████▀
         ▀██████▀
           ▀██▀
.M O B U.███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████
   The Investment Bank of the Future  
The Security Token Protocol and Licensed Security Token Exchange
█▀




█▄
Facebook Medium
Whitepaper ANN
LinkedIn    Reddit
▀█




▄█

         ▄▄▄█████▄▄▄
      ▄███████████████▄
    ▄███████████████████▄
   ███████████████████████
  ████ ▀███████▀    ▀██████
 █████▌  ▀▀███        ▄█████
▐██████▄             ▐██████▌
▐█████▄               ██████▌
▐███████▄            ███████▌
 ███████▄          ▄████████
  ████████       ▄█████████
   █████▀   ▄▄▄███████████
    ▀███████████████████▀
      ▀███████████████▀
         ▀▀▀█████▀▀▀

         ▄▄▄█████▄▄▄
      ▄███████████████▄
    ▄███████████████████▄
   ███████████████████████
  █████▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀█████
 █████     ▄▄          █████
▐█████     ████▄▄      █████▌
▐█████     ████████    █████▌
▐█████     ████▀▀      █████▌
 █████     ▀▀          █████
  █████▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄█████
   ███████████████████████
    ▀███████████████████▀
      ▀███████████████▀
         ▀▀▀█████▀▀▀

         ▄▄▄█████▄▄▄
      ▄███████████████▄
    ▄███████████████████▄
   ███████████████████████
  ████████████████▀▀▀ ▐████
 ███████████▀▀▀  ▄█▀  ██████
▐█████▀▀▀      ▄█▀    ██████▌
▐███▄▄▄    ▄▄██▀     ███████▌
▐████████ ███▀       ███████▌
 ████████▌█          ███████
  ████████ ▄▄██▄    ███████
   ███████████████▄ ██████
    ▀███████████████████▀
      ▀███████████████▀
         ▀▀▀█████▀▀▀
jpouza
Legendary
*
Offline Offline

Activity: 2870
Merit: 1122


View Profile
July 03, 2015, 10:57:31 AM
 #3855

I must be bored, just read the whole previous post. Their fingers musta been tired after all that keyboarding.    Grin

heheh same here, just watching, mining with free electricity only when available  Cool
GingerAle
Legendary
*
Offline Offline

Activity: 1260
Merit: 1008


View Profile WWW
July 03, 2015, 12:24:11 PM
 #3856

My private cryptonight kernal is around 10% faster on windows(windows executable). It can be yours for a 0.2 BTC donation. Then you also get my private spreadcoin miner with sourcecode (linux compatible).. (8.75-10% faster)



Linux?

< Track your bitcoins! > < Track them again! > <<< [url=https://www.reddit.com/r/Bitcoin/comments/1qomqt/what_a_landmark_legal_case_from_mid1700s_scotland/] What is fungibility? >>> 46P88uZ4edEgsk7iKQUGu2FUDYcdHm2HtLFiGLp1inG4e4f9PTb4mbHWYWFZGYUeQidJ8hFym2WUmWc p34X8HHmFS2LXJkf <<< Free subdomains at moneroworld.com!! >>> <<< If you don't want to run your own node, point your wallet to node.moneroworld.com, and get connected to a random node! @@@@ FUCK ALL THE PROFITEERS! PROOF OF WORK OR ITS A SCAM !!! @@@@
sp_ (OP)
Legendary
*
Offline Offline

Activity: 2954
Merit: 1087

Team Black developer


View Profile
July 03, 2015, 12:27:46 PM
 #3857

Linux?

Works on linux, but you don't get the sourcecode and I won't make a build. Windows for now.

The miner needs some more work. As you can see the performance on 960 is worse than on the 750ti. The gtx 980 is above 500H/s

Team Black Miner (ETHB3 ETH ETC VTC KAWPOW FIROPOW EVRPROGPOW MEOWPOW + dual mining + tripple mining.. https://github.com/sp-hash/TeamBlackMiner
pokeytex
Legendary
*
Offline Offline

Activity: 1504
Merit: 1002



View Profile
July 03, 2015, 01:17:04 PM
 #3858

Linux?

Works on linux, but you don't get the sourcecode and I won't make a build. Windows for now.

The miner needs some more work. As you can see the performance on 960 is worse than on the 750ti. The gtx 980 is above 500H/s

Sp - can you show a screenshot of the spreadcoin miner please?

dga
Hero Member
*****
Offline Offline

Activity: 737
Merit: 511


View Profile WWW
July 03, 2015, 04:50:05 PM
 #3859

Good for you, apparently $1500 isn't worth a 40 hour work week, probably not even that if you're as good as you make yourself out to be. There are plenty of programmers that get paid less, especially in other countries outside the US. Also why I mentioned kernel development here needs more competition. Some of you are pretty laid back... in a multitude of areas and want to get paid premium dollars for that. Meanwhile SP releases his work open source and you just build on top of it.

This is exactly why you guys don't advertise your services. You wouldn't want anyone to catch on just how lucrative this work is and there is plenty of programmers that are qualified in C++/Cuda/OCL, just not familiar with BCT or mining.

There IS NO COMPETITION here! Literally anyone could come in and swipe this out, but most people don't even know this is a fundamental part of mining yet. They just download whatever miner pops up and think they're set because these guys don't advertise. There is a ton of potential hashrate sitting there to be used for profit. The risk is quite minimal as well compared to getting into mining. There will always be people mining, regardless of what they're mining on and what BTC is at. So you basically need CUDA and OCL programmers, then someone to write a front end for it (or you simply ask someone like Nwolls on that already has one done).


Honestly it shouldn't matter who 'reaps the most from this'. DJM seems to be stuck on this note of people profiting off his work, even though he's selling it. It's two different things. Do gas stations get pissed off because oil tycoons or truck companies make so much money? No. They're a gas station, that's not their MO. If they had the resources to be a oil tycoon they probably wouldn't be a gas station anymore... or the oil tycoons simply buy up the gas stations.

(a)  I think you're missing some of djm's points:  In a bounty, the developer takes on all of the risk.  There's a decent chance of failing to get a 20% speedup after a month of work.  That's kinda sucky, particularly when contract programming work pays better anyway.

(b)  $1500/week is low for a good US programmer.  http://money.usnews.com/careers/best-jobs/software-developer/salary
2013 average software dev salary was over $92k/year ($1769/week), and the ones who can crank out a 50% improvement on the Monero miner are closer to that top 10% figure -- think $150k/year and up.  The bounty also doesn't include benefits and the other nice things offered by a standard employer.

I base this upon knowing personally a few of the people who ship some of the highest performance mining code.  Many are not US based, but I know what their day jobs would pay if they were here.  (It's probably more than $150k/year.)

Freelancing generally needs to quote double the hourly rate of a salaried position when you take everything into account, which means that a US programmer earning $150k per year needs to charge about $150/hour -- or about $6000 for a full 40 hour week, not the $1500 you mentioned.

I ran the numbers on doing this as a business a while ago, and it's not clear that it's worth it given the other things you could do with the software talent you need.

  -Dave

myagui
Legendary
*
Offline Offline

Activity: 1154
Merit: 1001



View Profile
July 03, 2015, 05:14:12 PM
 #3860

[...] (b)  $1500/week is low for a good US programmer.  http://money.usnews.com/careers/best-jobs/software-developer/salary
2013 average software dev salary was over $92k/year ($1769/week), and the ones who can crank out a 50% improvement on the Monero miner are closer to that top 10% figure -- think $150k/year and up. [...]
Quote edited for brevity.

Which is why so much of the US tech industry has outsourced to other nations, India being one notable example. The best ones do of course get transplanted to whatever location makes their work most efficient, but all in all, the US is a rather extreme reference.

I do agree with your general points however, but for the fact that comparing to any traditional employment reference, will not always make sense, as the bounty taker might be unemployed, still a student, in other words, that employment reference might be out of reach anyhow.

Pages: « 1 ... 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 [193] 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 ... 1240 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!