Ripped off by a pool?

[gene]

Good. Would it be possible to discuss in some detail what solution you implemented? What exactly the problems were and how you overcame them? Us code monkeys would like to know. Any chance to see actual code? That would go a long way in restoring some confidence with those who care about such issues.

To those who are obsessed with the issue of the sums involved, I will only say that if one can't be trusted with small sums, how can one be trusted with large sums? Not that it is at all relevant, but the anonymous nature of this forum and bitcoin allow me to experiment with pools using whichever resources I have, especially those which are less productive. Hint: I live in a cold climate. Quite simply, I don't tolerate abuse or evasiveness.

We need a vigilant community. This need will only become more pressing as the currency gains prominence.

[1715244000]

1715244000

[1715244000]

1715244000

[1715244000]

1715244000

[1715244000]

1715244000

[kiba]

We need a vigilant community. This need will only become more pressing as the currency gains prominence.

I gotten very angry with people for bad customer services before on this forum but...I never throw "ripped off" or "scammer" around so easily.

[distant]

All this thread tells me is that gene is an Unappreciative Impatient Douche

[gene]

All this thread tells me is that gene is an Unappreciative Impatient Douche

Top class. But you should see me at the PTA meetings. I'm very popular there.

[davout]

Top class. But you should see me at the PTA meetings. I'm very popular there.

I'm not OK with the previous UID-whatever post and I don't associate myself to that.
However, I think you owe slush an apology. (Oh yes, you should also remove this thread from your signature, OR make a big fat clarification about slush's honesty)

[gene]

Top class. But you should see me at the PTA meetings. I'm very popular there.

I'm not OK with the previous UID-whatever post and I don't associate myself to that.
However, I think you owe slush an apology. (Oh yes, you should also remove this thread from your signature, OR make a big fat clarification about slush's honesty)

I appreciate your distaste for that comment. I'm going to leave my signature intact, as it will hopefully dissuade anyone from getting cute in the future. I cannot be compelled to apologize, nor can I vouch for slush's character beyond what has already been written by everyone.

And thanks to whoever sent me the 20BTC.

[davout]

nor can I vouch for slush's character beyond what has already been written by everyone.

You can still concur

[slush]

I have been very open about all problems with accounting. Everything was solved, nobody was shorted in his reward. I have nothing to say anymore. Nobody is forced to trust me, but I don't understand why anybody spend so much time to go against me. That's all. For me, this topic is closed.

[jib]

gene, can you update the first post in this thread to indicate that you've now been paid and the accusations of scamming were false?

[Mahkul]

Actually, all this whining from gene made me feel like joining this pool. I wouldn't be surprised if it turned out that slush paid him to make this little bit of advertising.

I may put one of my Radeons on it in a moment.

[davout]

just generated a block of my own, felt like joining the pool for a little while again too

[Mahkul]

just generated a block of my own, felt like joining the pool for a little while again too

I think we have similar GPU power (just I have two separate cards generating 290Mhash each). Do you think it makes more sense to have them connected to one Bitcoin server or would you rather have them separated? I wonder does it make any difference; I had them connected to the same server for a while, but the results weren't what I expected and now I have them running on two separate machines, but for too short period of time to tell the difference as yet.

Sorry for balancing off the topic, we will get back to ripping off in a minute.

[davout]

I think we have similar GPU power (just I have two separate cards generating 290Mhash each). Do you think it makes more sense to have them connected to one Bitcoin server or would you rather have them separated? I wonder does it make any difference; I had them connected to the same server for a while, but the results weren't what I expected and now I have them running on two separate machines, but for too short period of time to tell the difference as yet.

Theoretically, it should give the exact same results.

Sorry for balancing off the topic, we will get back to ripping off in a minute.

Yes, we'd better get back on topic or gene might lodge a formal complaint

[gene]

This seems to be the most appropriate place to discuss various proposed modes of pool cheating and possible workarounds.

When I first heard of pools, I (along with many other, I am sure) immediately thought of the possibility of "micro-skimming." These are old attacks, but for a recent example, read the following link:
http://www.wired.com/threatlevel/2010/06/ftc-sues-scammers

Some people seem to think that these attacks are a waste of time, because of the small amounts being stolen at a time -- and for a small number of transaction this is true. However, the attack scales well. In fact, it scales well in two ways: 1) the more people/transactions are involved, the more money you steal and 2) the easier it is to hide what you are doing since the skimmed amounts get "lost in the wash." As long as the attacker doesn't get too greedy, the attack is difficult to detect.

My suspicion was raised in part because payments were being made in some seemingly arbitrary way. In fact, the details of how this works in slush's pool remain unclear. I hope that by discussing various forms of the fraud discussed in the above linked article, we can perhaps develop some resistant method by which we can minimize or eliminate the possibility of such attacks. How would you design a pool to cheat people out of money?

As of now, it seems easy to lie about the number of total work units required to solve a block. There is nothing to prevent the pool operator to open a few accounts and attributing some modest number of work to those accounts, increasing the number of total work units required, and dividing the 50BTC proportionally amongst all accounts, including his dummy accounts. With enough blocks, he can snag quite a sum of money. Honest clients will see that their work units are being properly accounted, but they can never see that their effort is being diluted by the operator.

As a bonus, the operator can also appeal to people's sense of gratitude by asking for donations for running his "generous and free" service.

I'll wager that if I have thought of this, somebody else has already implemented it.

How can this be defeated? Comments? Suggestions?

[jib]

I have an idea. Instead of paying by share of the block, the pool could pay a constant amount per share.

e.g slush's pool gives out work with difficulty 1 (I think), so on average there should be 12252.03471156 (current difficulty) shares per block. Each share is thus worth about 0.00408095 BTC, right?

So if the pool had enough BTC in reserve to cover statistical variation, he could just pay that much per share for every share, and you could verify that you're getting paid the right amount.

If this idea were implemented, theoretically on average the pool operator shouldn't make or lose any money (other than rounding errors). I'd be concerned about the chance of the pool possibly running out of money due to random variation, though.

[slush]

I have an idea. Instead of paying by share of the block, the pool could pay a constant amount per share.

e.g slush's pool gives out work with difficulty 1 (I think), so on average there should be 12252.03471156 (current difficulty) shares per block. Each share is thus worth about 0.00408095 BTC, right?

I cannot do this. Mining is not exact and I cannot pay out you forward, before we find a block. EDIT: There is only some chance that we will find block on every x shares. I cannot handle this risk and pay out per share.

For all - there is absolutely no other way than to trust me. Even opening sources will not solve this problem. If you don't trust me, you can leave the pool anytime. But you are welcome to join.

There is only one check, what everybody can do by himself. You can sum your reward from pool for long term (say few months) and sum your potential reward if you mine standalone. But if you make this calculation properly and will take into also rising difficulty, I'm pretty sure your reward from pooled mining will be higher, because you will receive more cents before difficulty goes higher every week.

[Raulo]

I'm pretty sure your reward from pooled mining will be higher, because you will receive more cents before difficulty goes higher every week.

No. In long term, there is no benefit of pooled mining. The only difference is reduced variance in the income stream. Yes, you can get, say, 5 BTC before difficulty gets higher with pooled mining but you have equivalently 10% chance of solving complete 50 BTC block before difficulty gets higher in standalone mining.

[BitLex]

No. In long term, there is no benefit of pooled mining.

depends on your processing-power (except for a very very long term, like decades at least).

if you try mining on a common CPU, you probably won't be able to contribute that much power to the pool, to earn you 5BTC within 2weeks,
most CPUs already need a year or longer on average to generate 1 block, at current difficulty, which will rise within that year.
so most CPU-miners won't generate any bitcoins on their own, ever!

within the pool those miners will not get 5BTC in a week or two, maybe a few cents, or only sub-cents, which isn't much, but far more than they could ever generate on their own.

if you really think, you could do it on your own, give it try, i bet you'll give up in a few weeks, or months without getting anything.

the/a pool is your only chance to get at least something back for contributing your CPU to the network, even if it's just a few bitcents, it's still better than nothing.
you won't make profit anyway, except for the unlikely case that you get your energy for free.


for high-end ATI GPUs you might be right, at least while those are on their own able to generate a block a day, or every couple of days, but that will change too.

for CPUs it's either a few cents from the pool, or nothing at all.

[Raulo]

depends on your processing-power (except for a very very long term, like decades at least).

Probability distribution changes with pooled mining but the average expected mining income does not. No matter what your mining speed is and what your time frame is. If you have a slow CPU miner from which you can get only a 1 BTC a month from pooled mining (and less when the difficulty increases), you have 2% chance of getting 50 BTC in the same time frame (actually you have an average payment of 1 BTC, you have a slightly less than 2% chance of getting 50 BTC but also some chance of getting 100, 150 BTC, etc. )

The "benefit" depends on what you like. Some people would rather have a 1 BTC than a 2% chance of getting 50 BTC. But others would rather have a chance of winning more than get pennies. Millions of people spend a few dollars to get a tiny chance of winning a million in lotteries (even though the expected payment is negative). 

most CPU-miners won't generate any bitcoins on their own, ever!

That's true. But there will be some CPU-miners who will win 50 BTC on their own, an amount they would never get from pooled mining. And the average of winnings of those lucky and unlucky is the same as the average income from pooled mining.

[BitLex]

a tiny chance of winning

that's the point and the big difference between pooled- and standalone-mining.

pools == get payed for the work you do. you will get payed, guaranteed
standalone == get a tiny chance to get payed