Eligius Miners: [POLL] Proposed changes to Eligius Reward System

[Luke-Jr]

Honestly, I don't believe it would have been several months if we maintained the hash rate of the pool at the time the bad luck (orphans) started.  It was a run of orphans that was the cause of the bad luck, then half of the pool left which pretty much increases the time to pull out of the rut insanely.  CPPSBAM works against miners who leave the pool, plain and simple.  If you don't want to mine the pool and support it, then you don't get paid. Easy as that.  If you feel like supporting the pool later, then by all means "unlock" your EC.

-wk

Certainly you aren't saying that hash rate determines luck.  Otherwise, the little miners don't stand a chance.

It does play a part in how quickly the variance ("luck") swings from one extreme to the other.

[1714936097]

1714936097

[1714936097]

1714936097

[1714936097]

1714936097

[imsaguy]

Honestly, I don't believe it would have been several months if we maintained the hash rate of the pool at the time the bad luck (orphans) started.  It was a run of orphans that was the cause of the bad luck, then half of the pool left which pretty much increases the time to pull out of the rut insanely.  CPPSBAM works against miners who leave the pool, plain and simple.  If you don't want to mine the pool and support it, then you don't get paid. Easy as that.  If you feel like supporting the pool later, then by all means "unlock" your EC.

-wk

Certainly you aren't saying that hash rate determines luck.  Otherwise, the little miners don't stand a chance.

It does play a part in how quickly the variance ("luck") swings from one extreme to the other.

I agree with that statement.  Looking at http://eligius.st/~artefact2/, hashrate appears to have dropped by about 50GH on average over the last month.  Do you happen to have any longer term graphs tucked away anywhere?

[wizkid057]

Honestly, I don't believe it would have been several months if we maintained the hash rate of the pool at the time the bad luck (orphans) started.  It was a run of orphans that was the cause of the bad luck, then half of the pool left which pretty much increases the time to pull out of the rut insanely.  CPPSBAM works against miners who leave the pool, plain and simple.  If you don't want to mine the pool and support it, then you don't get paid. Easy as that.  If you feel like supporting the pool later, then by all means "unlock" your EC.

-wk

Certainly you aren't saying that hash rate determines luck.  Otherwise, the little miners don't stand a chance.

I didn't say it determined luck, I said hash rate determines how long it'll take to get out of the rut, on average.  If we maintained hashrate we could possibly have been past this stroke of bad luck by now.  Possibly not, but the odds would have been better.

-wk

Edit: Whoops, didn't see that Luke had already responded to this more concisely than I.

[wizkid057]

Honestly, I don't believe it would have been several months if we maintained the hash rate of the pool at the time the bad luck (orphans) started.  It was a run of orphans that was the cause of the bad luck, then half of the pool left which pretty much increases the time to pull out of the rut insanely.  CPPSBAM works against miners who leave the pool, plain and simple.  If you don't want to mine the pool and support it, then you don't get paid. Easy as that.  If you feel like supporting the pool later, then by all means "unlock" your EC.

-wk

Certainly you aren't saying that hash rate determines luck.  Otherwise, the little miners don't stand a chance.

It does play a part in how quickly the variance ("luck") swings from one extreme to the other.

I agree with that statement.  Looking at http://eligius.st/~artefact2/, hashrate appears to have dropped by about 50GH on average over the last month.  Do you happen to have any longer term graphs tucked away anywhere?

I don't have a graph anywhere, but I suppose I could generate one, technically.

In any case, at the point the buffer flipped into extra credit we had over 550GH IIRC, meaning we easily lost over half of the hash power we originally had due to SMPPS extra credit mode.

-wk

-wk

[imsaguy]

If we maintained hashrate we could possibly have been past this stroke of bad luck by now.  Possibly not, but the odds would have been better.

-wk

Actually, no.  If the pool had been lucky,  the EC could have been paid off sooner.  Maintaining the hashrate while the the pool was unlucky wouldn't get the pool out of the rut faster.

You could also say that since the pool continues to be unlucky, the loss of hash power has been a good thing because less EC has accrued as would have otherwise.

[wizkid057]

If we maintained hashrate we could possibly have been past this stroke of bad luck by now.  Possibly not, but the odds would have been better.

-wk

Actually, no.  If the pool had been lucky,  the EC could have been paid off sooner.  Maintaining the hashrate while the the pool was unlucky wouldn't get the pool out of the rut faster.

You could also say that since the pool continues to be unlucky, the loss of hash power has been a good thing because less EC has accrued as would have otherwise.

Well, between the time EC mode started and now, lets say there were X shares submitted.  If the hash rate were still doubled, there would be 2X shares submitted now instead of X shares.  That gives X extra shares which could have resulted in lucky rounds, but instead were never submitted.  So, since history shows that 1X shares did not get us out of the rut as of now, it is possible that 2X shares could have by now.

Make sense?  Luke's post summarizes this better I think anyway.

[Hash rate] does play a part in how quickly the variance ("luck") swings from one extreme to the other.

-wk

[imsaguy]

Well, between the time EC mode started and now, lets say there were X shares submitted.  If the hash rate were still doubled, there would be 2X shares submitted now instead of X shares.  That gives X extra shares which could have resulted in lucky rounds, but instead were never submitted.  So, since history shows that 1X shares did not get us out of the rut as of now, it is possible that 2X shares could have by now.

Flawed logic.  Specifically "That gives X extra shares which could have resulted in lucky rounds, but instead were never submitted." Its just as probable they could have been unlucky.

[wizkid057]

Well, between the time EC mode started and now, lets say there were X shares submitted.  If the hash rate were still doubled, there would be 2X shares submitted now instead of X shares.  That gives X extra shares which could have resulted in lucky rounds, but instead were never submitted.  So, since history shows that 1X shares did not get us out of the rut as of now, it is possible that 2X shares could have by now.

Flawed logic.  Specifically "That gives X extra shares which could have resulted in lucky rounds, but instead were never submitted." Its just as probable they could have been unlucky.

Its not flawed. What I said is 100% accurate, although it is from the optimistic view.  The shares could have resulted in lucky rounds.  You can not deny this.  Just because they also could have been unlucky does not make my statement flawed.

[imsaguy]

Well, between the time EC mode started and now, lets say there were X shares submitted.  If the hash rate were still doubled, there would be 2X shares submitted now instead of X shares.  That gives X extra shares which could have resulted in lucky rounds, but instead were never submitted.  So, since history shows that 1X shares did not get us out of the rut as of now, it is possible that 2X shares could have by now.

Flawed logic.  Specifically "That gives X extra shares which could have resulted in lucky rounds, but instead were never submitted." Its just as probable they could have been unlucky.

Its not flawed. What I said is 100% accurate, although it is from the optimistic view.  The shares could have resulted in lucky rounds.  You can not deny this.  Just because they also could have been unlucky does not make my statement flawed.

You're right, it could.  So why is a vague statement like this used as support for change?

[cablepair]

I'm going to review this tomorrow and I'll vote and also make my opinion public

but I just want to note right now - that whatever the best payout route is for eligius - I do not believe that a bitcointalk forum poll is the best system for determining who should have a say and how much of a say in what happens here

and I say that as someone who has (through the good and the bad) mined nearly exclusively with this pool for the better part of a year and always being in the top 5 for hashrate


I am not saying my vote should mean more than any other miner who has contributed extensively to this pool but I am saying that it should mean more than some random person who reads this thread

-Tom
http://eligius.st/~artefact2/7/1Cp8HBr3vtEAvQ1RqQhjbhGciYDapf3ATN

[wizkid057]

I am not saying my vote should mean more than any other miner who has contributed extensively to this pool but I am saying that it should mean more than some random person who reads this thread

While in general, you're probably right.  However, I think that due to the nature of Eligius's account-less service, tallying votes linked to individual miners (via message signing and such) would be troublesome for even some of the largest Eligius miners.  I had considered writing up a setup to do just this, then realized that it just was too bothersome.

I think overall that even if non-miners vote after actually reviewing the poll that the consensus will be valid.

I've argued the merits of CPPSBAM on IRC several times today and convinced several miners that its the obvious choice.  I believe that anyone with the pool's well being as a whole at heart when voting will see it the same way.  If not, please give me the opportunity to attempt to convince you before voting.

In my opinion, if someone decides that the EC payout limit to active miners is a bad thing, that just means they are either an inactive Eligius miner or someone with some other self-serving benefit to vote against it.
An honest miner who mines the pool faithfully will not notice any negative effects from CPPSBAM, and these are the miners that I would say the pool wishes to attract and retain.
A pool hopper or someone who just bails on the pool after a bit of bad luck obviously wouldn't desire the change because it effectively bars them from receiving their extra credit rewards unless they actually support the pool.  I honestly don't see how this is a bad thing.

Basically, if you're an honest, steady miner, CPPSBAM is the way to go.

-wk

[eldentyrell]

several months isn't 'a little bit of bad luck'.

It wasn't; the evidence points to Eligius having been the target of a block withholding attack.  Luke-Jr posted this to bitcoin-dev about a week after Eligius started falling behind very badly:

  http://article.gmane.org/gmane.comp.bitcoin.devel/1112

IMHO none of these payout tweaks does anything to fix the block withholding attack vulnerability; they just redistribute the pain.  You should consider giving a 1% bonus to the submitter of the share-that-finds-a-block (deducted from everybody else's reward, of course) like P2Pool does.  If you jack it up to 3% it means that a share-withholder only gets 97% of their submitted hashrate, which should at least make idiot jackasses think twice.  The only penalty is that small-time miners only get PPS-level variance on 97% of their payout; the other 3% will have solo-miner-level variance.

I do not believe that a bitcointalk forum poll is the best system for determining who should have a say and how much of a say in what happens here…  and I say that as someone who has (through the good and the bad) mined nearly exclusively with this pool for the better part of a year and always being in the top 5 for hashrate

Agreed, from someone who was always in the top 3 for six months (and the top continuous/non-bursty miner for several months) until switching Eligius to "backup" status in response to the payout trainwreck.  But I guess you guys see people like me as "part of the problem".

[Luke-Jr]

several months isn't 'a little bit of bad luck'.

It wasn't; the evidence points to Eligius having been the target of a block withholding attack.  Luke-Jr posted this to bitcoin-dev about a week after Eligius started falling behind very badly:

  http://article.gmane.org/gmane.comp.bitcoin.devel/1112

No, there is actually no evidence of that (as far as I know). My post was related to brainstorming the withholding attack which might (or might not) have come from a discussion on the possibility of it that wouldn't have occurred if we had a nice big buffer, but it was certainly not directly correlated with any kind of actual in-practice problem to my knowledge.

[ehmdjii]

sorry for the noob question, but what does EC actually stand for?

[Luke-Jr]

sorry for the noob question, but what does EC actually stand for?

Extra Credit. http://eligius.st/wiki/index.php/Shared_Maximum_PPS

[muyuu]

I'm going to review this tomorrow and I'll vote and also make my opinion public

but I just want to note right now - that whatever the best payout route is for eligius - I do not believe that a bitcointalk forum poll is the best system for determining who should have a say and how much of a say in what happens here

and I say that as someone who has (through the good and the bad) mined nearly exclusively with this pool for the better part of a year and always being in the top 5 for hashrate


I am not saying my vote should mean more than any other miner who has contributed extensively to this pool but I am saying that it should mean more than some random person who reads this thread

-Tom
http://eligius.st/~artefact2/7/1Cp8HBr3vtEAvQ1RqQhjbhGciYDapf3ATN

Agreed. I haven't read enough about it to make an informed judgement, but in any case a poll here doesn't seem like the proper way to make a decision.

I also don't think that consensus matters much in a technical problem like this. The decision should be taken by those with a better understanding of the problem, and I guess Luke is the main man. Then probably we can discuss it here and agree something, but democracy is not the way to solve engineering problems.

[Exposed]

It would be better if you made an page dedicated to this new system suggestion and a public poll / comment system on Eligius site itself rather then here as active users are more likely to participate their then sign up on this forum just to voice 1 opinion

[Raize]

It would be better if you made an page dedicated to this new system suggestion and a public poll / comment system on Eligius site itself rather then here as active users are more likely to participate their then sign up on this forum just to voice 1 opinion

Eligius payouts are psuedo-anonymous, so there isn't really a way to do this on their site. We don't have typical "miner accounts".

[imsaguy]

It would be better if you made an page dedicated to this new system suggestion and a public poll / comment system on Eligius site itself rather then here as active users are more likely to participate their then sign up on this forum just to voice 1 opinion

Eligius payouts are psuedo-anonymous, so there isn't really a way to do this on their site. We don't have typical "miner accounts".

Not true. Address signing. Just like you do to get namecoins.

[Raize]

Not true. Address signing. Just like you do to get namecoins.

I was making the point that Luke would specifically have to code a voting system in. Seems like extra work for something the forums can do without trouble. Only real problem is that pool hoppers could game the vote, but I doubt any of them care enough to even read the topic.