Bitcoin Forum

Awww yeah, it's like Christmas all over again.  Got 500 GH/s coming your way... at least for 42% of the time.

Awww yeah, it's like Christmas all over again.  Got 500 GH/s coming your way... at least for 42% of the time.

You are welcome :)

No offence man, but how much research did you do into payout systems before you chose the most exploitable one?

What do you mean?

Yeah, that 43% is mine! All mine!  God help you if you try to change the payout method in the middle of a block.

Grats on the fact that you pool didn't just explode when I pointed my farm at you.  I am impressed with that at least.

I'm even going to be awesomely nice and update all my rigs to 2.8.0 so I will hit you with GBT on all my rigs instead of killing your GW.

Inaba , after you find the first block tell me what i should add to web page? Some features...

https://eclipsemc.com/ is Inaba's site - maybe you can glean few clues on what he likes from there :)

Thanks , i`l not change it will remains with no fees.

Right now i`m planning to get more servers and i`m monitoring your speed to learn what i have to do next to be great :)

P.S i`m new on bitcoin and i see this is my first A+ :P

request:

Please add worker share numbers to api json.

Please add pool stats to api json

thank you

Greee, everybody is mocking you. Your payout strategy is highly flawed. This should be your #1 priority (all other features don't matter in comparison). Read the pdf linked above by Graet to learn more.

tl;dr: Don't use proportional payout. Ever. It has severe, crippling flaws.

PS. Sorry pool-hoppers  :-\

$estReward = $totalShares/$totalPoolShares;
$adminFee = '0';
$estReward = round((50*$estReward)-(50*($adminFee*.01)), 8);

Thanks for this , but as i have said we are new on bitcoin network and we try to improve our pool server to work flawless, we have the resources to create more servers and to develope existing or new web functions so first time we have no fees to make a profit.

We simply reward the miners exacly for their work.

This isn't about whether you as a pool make a profit, it's about your miners getting ripped off.


No, you pay a lot to pool hoppers and very little to 24/7 miners who support your pool.

Take a look at the bottom of this webpage: http://organofcorti.blogspot.no/2012/09/23rd-september-weekly-pool-statistics.html

The two bottom charts show the pool hopping going on at 5 pools with vulnerable reward systems, "proportional" and "scored proportional". See how the hashrate is much higher for the short blocks than the long ones that take a lot of work. This is pool hoppers ripping off the 24/7 miners on these pools.

There's no excuse for these pools to still be running these reward systems and letting their miners get ripped off. Please don't start another pool like that.

You would need about 10,000 BTC in reserve to properly operate a PPS shop.

You would need about 10,000 BTC in reserve to properly operate a PPS shop.

Thanks for this , but as i have said we are new on bitcoin network and we try to improve our pool server to work flawless, we have the resources to create more servers and to develope existing or new web functions so first time we have no fees to make a profit.

We simply reward the miners exacly for their work.

Reward formula:

No one has ever asked where I got the name Inaba...

No one has ever asked where I got the name Inaba...

Right, I've earned nothing and he'll get a scammer tag :)  It would be worth a few hours of hashing though.

He hasnt said anything about shutting down or changing payouts at this current time.. if he wants to change the payout system he will need to create a block and pay out accordingly.. otherwise he can just let the pool run and wait until it finds a block.. thats all.

Wow missing a few hours and there is a mess:). Thanks wknight , no one says about shutting down or scam.. i`m not a scamer and i will never be one. I`m thinking to prep more servers . more functions , ideas.. it will remain for the moment prop and no fees. And yes "hahahafr" we pay miners tax that ( 0.005 i think you refer ) = total 0% fees.

Waiting to find the first block and anyone who mined they will get theirs share/reward.

Thank you.

With all that's been said I have to say it's a good interface on this pool - if it had any anti-hopper setup at all (even a rudimentary one like deepbit has) then I could see people using it.

I`m waiting to find the first block and pay his share to Inaba :) then to deny Pool-hopping, if someones like Inaba leaves i`l split his reward with all active miners. Thinking...

If the round duration at the end of the round is greater than 1 hour, each user is looked at to see their 'window of participation', which is the time difference between their first and last submitted shares.

If the time difference between their shares is less than one half of the round duration, their share count will be reduced by 50% and the other half of their shares will be credited to an account setup by the pool operators.

We've monitored this over the past 10 rounds and seen no higher than $0.75 BTC worth of shares removed per round from pool hoppers, and on average, less than 15 accounts are affected per round.

We believe that this will become an effective deterrent for pool hopping, as it will remove the benefit of doing so, add incentive to stay participating in the pool, and at the same time, offer FairUser and Myself a minimal compensation for all of our hard work and late nights.

We considered our options and alternatives with this for quite a long time before deciding to implement it and we believe it is a fair method that allows an easy correctable course of action for each affected miner to take.

Edit: (Posted this below, but added it here too)

I wanted to clarify exactly what this is doing...

As an example, lets say the current round has been going for 2 hours when we solve a block. Our framed window for checking on pool hoppers will be 1 hour.

Each user will then have timestamps taken of their first and last share of the round to see if the difference between them is greater than or equal to 1 hour. If it is not, we will assume they're pool hopping and we will discount their shares by 50%.

It doesn't matter if the user joins 30 minutes into the round and leaves 30 minutes before it ends, as long as they were present for a minimum of 50% of the round duration, they will not be affected.

On average, this affects 5 - 10 users per round.

This does not make shares submitted later in the round have more value.
This does not favor GPU over CPU miners.
This does not affect any user if the round is shorter than 1 hour.
This is not a score-based system.

Whatever you do you need to make sure you make it very clear in advance.
What you're describing sounds remarkably similar to slush's score-proportional system, but you need to make it clear in advance.

Delay stats? Or better yet.. Do what Bitcoinpool does.

http://bitcoinpool.com/forum/viewtopic.php?f=1&t=103

And yes "hahahafr" we pay miners tax that ( 0.005 i think you refer ) = total 0% fees.
Thank you.

Thankfully it's still profitable to hop at 52%  ;)

No one has ever asked where I got the name Inaba...

Nope, not that Inaba.  It does involve hopping though, in a tangential sort of way.

因幡の白兎っていうの神話的作り話からの「いなば」でしょう

Delay stats? Or better yet.. Do what Bitcoinpool does.

Thankfully it's still profitable to hop at 52%  ;)

http://en.wikipedia.org/wiki/Hare_of_Inaba

Did I hit the nail? Hare, hop.

Nope. It can't be profitable after 43.45%, it's just impossible. Mostly the share value expectation is 1.0 at 23.6% or so, but as the pool hashrate increases to a larger proportion of D, the hop window stretches to a maximum of 0.4345xD - sort of like what happens with Slush, only more so.

Run a few simulations and you'll see I'm right. I'll get around to posting a derivation some day, and calculate how much fulltimers there lose (especially since only half of the hopper tax is returned to fulltimers). From what I see most hoppers are leaving a 23.6% regardless of hashrate and not profiting as much as they could.

...

@ Greee I know certain forum members will hate me ;) for repeating what Dr Haribo has said but there is no future in using a payout method which unfairly penalises fulltime miners.

First, read https://bitcoil.co.il/pool_summary.pdf and https://bitcoil.co.il/pool_analysis.pdf. From the first, you can pick and choose a reward method that suits you. From the second, you get more details on each reward method.

PPS: You'll see that you can calculate exactly how much risk you want to take with PPS. In Meni's example, the bitcoin block reward is 50 btc, the risk of bankruptcy you'll will to accept is 1/1000, and the fee is 5%, then you'll need to have a bankroll of 3454btc to start the pool - although if you wait until the block reward is halved, then the required bankroll to start is 1727 btc.

PPLNS: It's a little trickier than you might think to implement in a completely pool hopper invulnerable way. Just ask Dr. Haribo. It's intuitively easier to grasp than DGM, but variance and maturity time can't be controlled to the same extent.

DGM: Tricky, and you might want to ask Inaba or Graet (if Inaba doesn't want to lose a hoppable pool :)) for some help. But you can control a great deal about your miners' experience by varying it's variables.

That's it, Greee. There are no other options if you want to retain fulltime miners.

Good luck!

Any plan for Stratum mining ?

Any plan for Stratum mining ?

The pool already supports the standard GBT protocol. There's no need for Stratum, which just does basically the same thing (except in a way that is more harmful to Bitcoin), despite a couple of big pools throwing around their weight on it.

Well I hope you aren't a hypocrite calling Stratum harmful but considering including it in your SatanMiner?
That wouldn't show well for your character to do something like that ... ... ... oh wait ...

We have write a new reward formula/hopping prof based on latest miners speed logs and we will reveal it tomorrow after we run some simulations.

We hope to be a revelation :-))

Sounds....intriguing.

We have write a new reward formula/hopping prof based on latest miners speed logs and we will reveal it tomorrow after we run some simulations.

We hope to be a revelation :-))

Please, don't. If it's not provably fair anyone will be able to write an exploit for it within a day, and your pool hoppers will still profit only a little less so. There's no reason to create a new reward method.

Well this is your opinion , i wish and i`m sure that it cant be exploited , it will have at the base miner speed calculated every 10 minutes.

Let`s say there are 50 miners with medium speed of 10 GH/s each someone will come with 500 GH/s ( like Inaba :-) ) he mine 2 hours then he leave then the system will read his latest 10 min speed and will see that drop to 0 from 500 so 2 hours = 12 speed update rounds , so will apply the formula $inaba_speed = 500+500+500...12 times and he continues +0+0+0+0 ( $estimated_inaba_reward = $inaba_speed / $total_pool_speed * 50; ) so his reward will drop down every 10 minutes and will go up to full online miners.

Anyway there are some theory its a little more complicated i`l present the full formula after i`l implement it and its extra hopping prof.

sounds similar to the point system... without the points.

I believe slush's pool uses point system:

https://bitcointalk.org/index.php?topic=1976.msg50002#msg50002

Btw from your description it looks like you tried to invent a wheel, but you actualy invented square.

grabs popcorn

Well this is your opinion , i wish and i`m sure that it cant be exploited , it will have at the base miner speed calculated every 10 minutes.

......


Anyway there are some theory its a little more complicated i`l present the full formula after i`l implement it and its extra hopping prof.

How confident are you? Are you willing to pay btc if someone can find a way to hop your method and report it to you? Let me know and I'll start warming my keyboard.

:) Your est reward will drop down seriously if u disconnect from pool , i`l let u know when u can start.

I don't mine. I don't need to mine in order to find a way to exploit a reward method. I'm asking if you're willing to pay me to find out if your reward method is exploitable and report any exploits I find. I'm also asking you to offer sum of btc payable if I do find an exploit.

How much you offer me to do this will be a indication of how confident you are in your reward method, and how confident full time miners should be that they are being paid fairly.

I`m confident because all simulations go well , it`s use same logic as slush`s pool but with other formula , we will see it in action in a few hours and yes full time miners will be paid 101% fairly.

So that's a "No", then? You're confident enough to go online with your reward method but not confident enough to see if someone can break it?


Full time miners please take note.


edit: because I misplaced the damn quotes!

edit2: You do know slush's reward method is exploitable, and he wants to change to a non-exploitable reward method? You know this, right?

Are you argue with me?

We or i`m not a 12 years old kid, i like to create/build not to debate something... So please calm down , i`m confident that will be non-exploitable u need 100% answer?

Miners , answer only for "organofcorti": YES will be non-exploitable.

Please again let the miners decide that after will be live.

Thank you.

sup.

oh, disconnect = massive reward drop? psh, that's easy.
...

Are you argue with me?

We or i`m not a 12 years old kid, i like to create/build not to debate something... So please calm down , i`m confident that will be non-exploitable u need 100% answer?

Miners , answer only for "organofcorti": YES will be non-exploitable.

Please again let the miners decide that after will be live.

Thank you.

Hate to argue with you but, miners could just connect right when it records your hashrate, then disconnect after its recorded. Then reconnect 10 minutes later, then disconnect... etc etc. Full payout while only mining say 1 minute every 10.

Well actually, that depends on if your using the average hashrate over the last 10 minutes, or if you are using instantaneous hashrate.

We have created a hooping proof formula, feel free to test it.

I will explain it in a few moments :)

... weird pm your ip to check some routes...

We have created this new hooping-proof formula so all non-active miners loses their shares.

I had bitarena blocked at my router  :)

We have created a hooping proof formula, feel free to test it.

I will explain it in a few moments :)

!!!  this is unacceptable, you will not get anybody to mine at your pool if you 'take' their shares.  what if my miners crash? i lose my shares?

So you want people to mine on your pool by stealing their hashing power?  Good luck with that.  Your pool is 100% fail from the start.

Theft of resources is unacceptable under any circumstance.  Anyone mining on this pool will know that at any time you, the operator, will take their shares and BTC at will.  You are basically admitting to being a thief now.

You, Greee, have now admitted to stealing 43% of the shares submitted to your pool. (Well, you've actually stolen 100% of the shares, but we won't quibble)

Gree, you can't just go and reset the stats system mid-round. You need to pay Inaba for those shares.

Yeah, that 43% is mine! All mine!  God help you if you try to change the payout method in the middle of a block.

Grats on the fact that you pool didn't just explode when I pointed my farm at you.  I am impressed with that at least.

I'm sorry, no, it doesn't work like that.  I mined on a prop pool, not a hop proof pool.  I told you from the start:


You just did it.  You can not be trusted.  Plain and simple.

You made the drama for yourself by stealing shares.  You were told not to run a prop pool from the start, but you went ahead and did it anyway because you thought you knew better.

Now, when you finally realize you have no idea what you're doing, you switch mid round and steal all the work because of a mistake YOU made, which people tried repeatedly to enlighten you about and you refused to listen.  That's two mistakes you made and you are trying to blame the users?  Really?

Good luck man, your pool will never succeed if you steal from your users like that. 

No, you can not make changes in the middle of a round if you expect people to trust you.  You offered a pool with a set of parameters which I agreed to mine on.  If you want to change your parameters, you wait until AFTER the round is over.  You don't change it after I've already given you my hashes.

What's to stop you from doing that to others in the future?  Nothing, because you have no integrity.  You've demonstrated the fact that you can and will take away users shares for any reason you see fit and not pay them for it.   You have zero integrity at this point and you continue to compound your problem.  You will never have a successful pool with that sort of mind set.

Please get lost Inaba. I have no fees and i have no reason to steel or something like that i just love this , i have money / servers i just love to work and create so please get lost with your ideas , i will protect full time miners not hoopers like you.

I will repeat myself i will GLADLY restore your shares if you will continue mining.

There is no way I am mining on a pool who's operator is an admitted thief, you have got to be kidding, right?

YOU ARE A POOL HOOPER MAN, get a life.

IL REPEAT IL RESTORE YOUR SHARES with a simple command IF YOU WILL CONTINUE MINING. Our hooping proof system remove hoopers shares. If they come the system will restore them .... i`l present the full formula in a few moments.

Please respond to the following use case:


I mine at your pool with my 4 gig rig
my 4 gigs is 10% of pool
we mine for 4 days and I have about 100k shares and 10% of the pool shares
my hard drive crashes, rig is down.


what happens to all my work?


edit:  or better yet

I mine with all of my 54gig
i mine for 4 days and i am 50% of pool (jtfr 260 usd of electricity)
my internet goes offline for extended period of time


what happens to my work?

So your pool is all about the blackmail?

New proposal:

Greee must either:

-Pay Inaba the expected value of the shares he submitted (no, I don't mean PPS, I'm sure organofcorti can provide us with the average expected value of said shares when mined on a prop pool).
Or:
-Switch back to the old payout system and restore Inaba's sharecount, then switch to the new payout method at the end of the round.
Or:
-Get a scammer tag.

please respond to the above scenarios

Shadow383 is the same payout system and prop, but he have some hoopers proof rules, that i will fully explain in a few moments.

i`m not stupid to steal or something lol , i just install some hooping proofs.

jjiimm_64 i have all users shares backup and i can restore it in a few sec , i just want to explain the hooping prof formula so you can all understand it before i restore your shares and what are the risks if you not mine anymore.

Thank you.

http://www.youtube.com/watch?v=l1YmS_VDvMY

Hmm, I'm thinking scammer tag might be warranted here.  I wasn't really going to advocate for that, but given the ops reluctance honor the agreement he made with his miners when he started the pool and now his inability or unwillingness to provide any sort of "proof" for this hopper proof method smells like a giant scam at this point.  I wonder if we solved a block and he pocketed the 50 coins now.  I didn't bother looking at the blockchain around that time... may be worth investigating.

In either case, he's scammed me out of 1.3 millions shares at this point, possibly even ~50 BTC.

jjiimm_64 after il present the formula you can let me know if there is a bug or something or what i need to fix , what was my first post reply?

"New hooping prof fell free to test it" i never said Pool reset or something.

This formula will protect full time miners and they can earn extra from pool hoopers.
Percents can be adjusted waiting for feedback to make it near perfect.

Fast Description.

Case 1 >> User has stop mining.
Miner total shares = 1000s say
Old shares lose value
Every minute he will lose 10% of his shares. ( in 10 minutes he will lose all his shares )
Lets say miner will come back in 5 minutes , he lost 500 shares and 500 remains so the system will return to him 80% of 500 ( 500+500/80% of 500 )

Case 2 >> User drop down his speed ( ex: from 100 GH to 10 GH , 90%)
Miner total shares = 1000
Old shares lose value
User will lose 50% of his shares. Speed check every 10 minutes.

Case 3 >> User come back and start mining in maximum 30 minutes ( He got power off / Net was down / Hooping / etc... )
User will get shares restore 80% of lost shares.

if this is what you have chosen as a paymethod going forward, that is your choice. I wont waste my hashing power on your pool.

This method does not take into account user that has extended internet blackout.  I could spend hundreds of dollars on electricity and get paid nothing... no thanks

I`l gladly respect that , if you have internet problems or power... i can`t recommend you to mine its for full time miners , my rig mine for last full month without any interruptions so i`m stable.

LOL,  yes and i have not been under 50gig for more then a few minutes in 6 months.. not counting the power blackout that lasted 16 hours.... 

good luck then if your going to be that stubborn.  you can pay me pps for the shares i mined already.

It`s a logic system i`l restore the users shares but you will lose all your shares if you don`t mine.

It doesn't work this way. They mined with the contract of a proportional reward. Changing the contract now is fraud.

You seem to be under the impression that pool hopping is an exploit. It's really not - it's just ratoinal miner behaviour with proportional contracts.
The problem isn't the hopping, it's the reward system being unfair. Proportional means the pool pays above market rates for blocks found in the first 50% shares, and pays below market rates for blocks found in the last 50% shares.
For an analogy, this would be like a fast food restaurant paying you much more during on-peak hours, and paying less than minimum wage during off-peak hours. What you're doing is saying "yes, you worked on-peak when we agreed to pay $50/hr; but now nobody wants to work off-peak for 50 cents per hour, so we're just not going to pay you".

Greee, still waiting on reward method details -  "use cases" don't really allow miners to predict what they should be paid.

Any ETA on the details?

I have describe it here ( https://bitcointalk.org/index.php?topic=112243.msg1253931#msg1253931 )

I have restore the system to old format without this hooping-proof formula... a lot of ppl make noise so i use the prop system without this formula.

System and shares , restored.

This formula will protect full time miners and they can earn extra from pool hoopers.
Percents can be adjusted waiting for feedback to make it near perfect.

Fast Description.

Case 1 >> User has stop mining.
Miner total shares = 1000s say
Old shares lose value
Every minute he will lose 10% of his shares. ( in 10 minutes he will lose all his shares )
Lets say miner will come back in 5 minutes , he lost 500 shares and 500 remains so the system will return to him 80% of 500 ( 500+500/80% of 500 )

Case 2 >> User drop down his speed ( ex: from 100 GH to 10 GH , 90%)
Miner total shares = 1000
Old shares lose value
User will lose 50% of his shares. Speed check every 10 minutes.

Case 3 >> User come back and start mining in maximum 30 minutes ( He got power off / Net was down / Hooping / etc... )
User will get shares restore 80% of lost shares.

Congrats - here's hoping you find it sooner rather than later  ;)

Just so you know, this method is basically slush's method, still hoppable, just not as hoppable as straight proportional.

Ok its still hoppable but its a minimum , a little risky for hoppers they lose , i`m thinking to perfect it and check theirs internet connection to see if it`s UP or DOWN so then i will know what will be the reason why they left.

I have describe it here ( https://bitcointalk.org/index.php?topic=112243.msg1253931#msg1253931 )

I have restore the system to old format without this hooping-proof formula... a lot of ppl make noise so i use the prop system without this formula.

Lets say miner will come back in 5 minutes , he lost 500 shares and 500 remains so the system will return to him 80% of 500 ( 500+500/80% of 500 )

I did read that Greee, I just think it's a bit light on details.

1. Miner hashrate reduces to zero: Shares are devalued linearly so that every minute a miner is disconnected 0.1*(original shares) are lost so that after ten minutes the shares have devalued completely. After five minutes only 50% of the shares are left. However, you add:


If you have 1000 shares and you lose 500, surely that's only 50% returned? Unless you're saying that if a miner returns to mine more, more shares will be returned?

If it's the latter case, can you post the parameters? What percentage is returned after how long?

So you're penalising miners who leave at any time of the round?