Continuum Mining Pool: No fees; Client uptime monitoring via twitter and email

[disq]

thanks!

[Meni Rosenfeld]

Ok, I think now we're allowed to ask... martok, are you absolutely sure there's no problem with the pool that would cause it to miss winning shares?

[relmeas]

I tried this pool, since it offers about 6% better PPS rate than deepbit.

However it turned out the rejected rate is about 10% so it's still less profitable...

I hope the long polling will be implemented soon (or whatever is required to have 0.6% rejected ratio I have at deepbit), then I'll give it another try.

[ADebelius]

My rejection rate averages .5% to 1% here at continuum, which is the same I get at deepbit. I'd look client side before blaming the server.

[martok]

Ok, I think now we're allowed to ask... martok, are you absolutely sure there's no problem with the pool that would cause it to miss winning shares?

It appears there is nothing amiss with shares.
The pps side has found 3 blocks since it went up, scoring has not found a block. I was curious as to whether there was a pool problem also so ran the software on the testnet for a bit and all works fine with finding blocks and recognizing whether they are PPS or proportional. I think we just need more hashing power on proportional side.

Note also that when the round does complete, a share history will be made available. Assuming your client logs its share submits, you will be able to compare that against the history to ensure all your shares were counted.

[martok]

I tried this pool, since it offers about 6% better PPS rate than deepbit.

However it turned out the rejected rate is about 10% so it's still less profitable...

I hope the long polling will be implemented soon (or whatever is required to have 0.6% rejected ratio I have at deepbit), then I'll give it another try.

If you're getting 10% rejected, there is a problem somewhere. If you can PM me your login address, I'll look up the reject reasons. The pool does support long polling so that isn't the cause.

[Meni Rosenfeld]

The pps side has found 3 blocks since it went up, scoring has not found a block. I was curious as to whether there was a pool problem also so ran the software on the testnet for a bit and all works fine with finding blocks and recognizing whether they are PPS or proportional. I think we just need more hashing power on proportional side.

The word "proportional" is usually reserved to the defective system of using the same score for all shares. What we have here is called score-based.

Maybe you can try being more aggressive with promoting the pool? All the other pools are more or less the same and this is the only one that offers a distinctive advantage, there's no shame in spreading the word.

[Sukrim]

I tried this pool, since it offers about 6% better PPS rate than deepbit.

(50-0.0000541647*877226.66666667)*2 = 4.97056154%

I thought deepbit charges 10% for PPS? This would mean this pool only offers 5% better rates - while not(?) paying for stale shares. In the long run this is not really desirable as a miner imho as you earn 5% less than expected. For pool operators it's also not desirable as miners can be easily patched to "leech" PPS pools (aka "not submitting winning shares attack") and such a pool can be killed by any rivalling bigger pool with just the missing 5% covered out of own pockets.

[twmz]

I tried this pool, since it offers about 6% better PPS rate than deepbit.

(50-0.0000541647*877226.66666667)*2 = 4.97056154%

I thought deepbit charges 10% for PPS? This would mean this pool only offers 5% better rates - while not(?) paying for stale shares. In the long run this is not really desirable as a miner imho as you earn 5% less than expected. For pool operators it's also not desirable as miners can be easily patched to "leech" PPS pools (aka "not submitting winning shares attack") and such a pool can be killed by any rivalling bigger pool with just the missing 5% covered out of own pockets.

Continuum pays 47.5 BTC / difficulty per share
Deepbit pays 45 BTC / difficulty per share

47.5 / 45 = 5.5555555% more at continuum.

Neither pay for stale shares.

[Sukrim]

Why would you mine with 5-10% "built-in" bad luck using a method that is easily exploitable by a known method?

To decrease variance you can just mine in several pools (even at once) and if they're score based you'd also get the expected rate (if they are prop. you can hop and get even more )

[relmeas]

For pool operators it's also not desirable as miners can be easily patched to "leech" PPS pools (aka "not submitting winning shares attack") and such a pool can be killed by any rivalling bigger pool with just the missing 5% covered out of own pockets.

If some pool would want to perform this kind of attack, I don't see why would it not be able to do the same thing with proportional system.

He would still be paid for each found block but would not submit winning ones he find... Same thing basically - taking but not giving. In the long run it would be paid the same (or little more) than if he would using PPS - just like the honest miners.

So what exactly makes PPS more desirable for such attack?

Sure, the pool might lose if the PPS rate is more than statistical worth of a single share. But then the pool owner can (and should) always change the PPS rate so it would reflect current situation much better. For example 7% deepbit declares seems fair to me since as you can see, the actual shares per block is statistically much more than it's supposed to be, for some mysterious reasons... perhaps it's those attacks going on?

[PoulGrym]

Attacking PPS then Prop in the sense is that PPS pays you every share you send in even tho no block is found. While Prop only pays once the block is found. Not finding a block on PPS would just take money from the Pool admin. Not finding a block in Prop will only make the pool to never pay you.

[gentakin]

So what exactly makes PPS more desirable for such attack?

PPS: Pool operator pays for losses when blocks are not submitted. (Since they still have to pay for every submitted share.) Effectively, by not submitting a block, the attacker steals 50BTC that should be rightfully belonging to the operator.
Proportional: When blocks are not submitted, the pool simply appears to have bad luck. So it's the miners that pay for the attack, not only the pool operator.

So attacking like this works much better on PPS pools, as it directly hits the operator and might quickly ruin them.

[relmeas]

Attacking PPS then Prop in the sense is that PPS pays you every share you send in even tho no block is found. While Prop only pays once the block is found. Not finding a block on PPS would just take money from the Pool admin. Not finding a block in Prop will only make the pool to never pay you.

that's why PPS fees are supposed to be larger - for taking the risk. also, it's SUPPOSED to even up with time.

honestly I don't see what the problem is. if the pool can't afford PPS, just switch it off and don't advertise it.

And once again, it does not matter if there is PPS or there is no PPS. Attackers would attack anyway and prop miners would still get less and eventually leave.

Also, repeating myself, if PPS is draining too much BTC then just decrease per share payout. Easy enough! If it's still not paying out, well then maybe there is a problem with the pool that does not generates any blocks at all... why would anyone bother with it even in prop-like mode?

[Meni Rosenfeld]

that's why PPS fees are supposed to be larger - for taking the risk. also, it's SUPPOSED to even up with time.

PPS fees are higher to account for the operator's variance when all miners are honest. It can't hope to handle sabotage attacks.

honestly I don't see what the problem is. if the pool can't afford PPS, just switch it off and don't advertise it.

I agree with this. martok bit more than he can chew with the PPS. Which is very unfortunate, had he consulted with me I'd have explained the risks. The fact that he conditioned PPS payouts on its eventual finding of blocks means he doesn't get PPS at all.

And once again, it does not matter if there is PPS or there is no PPS. Attackers would attack anyway and prop miners would still get less and eventually leave.

Wrong, trying to sabotage a share/score-based pool is shooting oneself in the foot. If the attacker is 50% of the pool, then by withholding blocks he denies himself 50% of the reward. Thus sabotaging such pools is much more expensive. In PPS withholding blocks does not decrease the reward (beyond the normal PPS fee).

Also, repeating myself, if PPS is draining too much BTC then just decrease per share payout. Easy enough!

Not easy at all, the operator needs to keep hundreds or thousands of BTC in reserve to ensure stable PPS payouts. Which is what I don't think martok realized.

[relmeas]

that's why PPS fees are supposed to be larger - for taking the risk. also, it's SUPPOSED to even up with time.

PPS fees are higher to account for the operator's variance when all miners are honest. It can't hope to handle sabotage attacks.

honestly I don't see what the problem is. if the pool can't afford PPS, just switch it off and don't advertise it.

I agree with this. martok bit more than he can chew with the PPS. Which is very unfortunate, had he consulted with me I'd have explained the risks. The fact that he conditioned PPS payouts on its eventual finding of blocks means he doesn't get PPS at all.

And once again, it does not matter if there is PPS or there is no PPS. Attackers would attack anyway and prop miners would still get less and eventually leave.

Wrong, trying to sabotage a share/score-based pool is shooting oneself in the foot. If the attacker is 50% of the pool, then by withholding blocks he denies himself 50% of the reward. Thus sabotaging such pools is much more expensive. In PPS withholding blocks does not decrease the reward (beyond the normal PPS fee).

I agree that PPS is more a privilege of larger pools that get stable number of blocks per day and are harder to attack.

However, for the new pool with very small hashing power, this could be the main reason for people to even consider moving there. who wants to wait several days per block with the difficulty constantly raising, when you could just mine at a larger pool and get full advantage from current difficulty which is lower than it will become in just a few days? with PPS those miners would still get their reward but the pool would also grow.

also why are you only talking about pools being cheated? why not talk about the miners? what prevents a pool from simply redirecting some of the miners to the completely separate bitcoin client or a pool that would produce blocks not from a pool's name but from "other"? miners would get less reward (longer blocks) but would blame it on bad luck but the pool would get absolutely free 50 BTC per such block. with PPS the miner at least does not get less than he is supposed to in such theoretical occurrence (still needs to be checked of course but at least it's very simple to do - just multiply accepted by PPS rate and compare with received reward).

[Meni Rosenfeld]

who wants to wait several days per block with the difficulty constantly raising, when you could just mine at a larger pool and get full advantage from current difficulty which is lower than it will become in just a few days?

How about being the only pool around using a hopping-proof scoring method?

with PPS those miners would still get their reward but the pool would also grow.

martok's PPS and score-based are two separate pools. Drawing people to the PPS does not help the score-based grow. And opening up a new PPS pool makes no sense for someone small.

also why are you only talking about pools being cheated? why not talk about the miners?

Because that is a different problem which has not come up in this discussion.

The short answer is that the mining software knows when it finds a valid block. The user can monitor for this event and if he discovers that a block was found and the pool wasn't rewarded he can blow the whistle.

And yes, not having to worry about this issue is one of the selling points of PPS.

[martok]

So what exactly makes PPS more desirable for such attack?

PPS: Pool operator pays for losses when blocks are not submitted. (Since they still have to pay for every submitted share.) Effectively, by not submitting a block, the attacker steals 50BTC that should be rightfully belonging to the operator.
Proportional: When blocks are not submitted, the pool simply appears to have bad luck. So it's the miners that pay for the attack, not only the pool operator.

So attacking like this works much better on PPS pools, as it directly hits the operator and might quickly ruin them.

True, and as I have said elsewhere in this thread, if that happens then PPS will simply go away. So by attacking PPS, it will simply no longer be available to them.

Score-based continues to be my primary focus with this pool as I believe longterm, it is the way to go.

[PoulGrym]

I've also reported about PPS risk when I was on swepool.. I do still think it's good to have competition to deepbit. One solution could be the MaxPPS as Eligius uses I think...

[twmz]

Continuum apparently had issues starting an hour or two ago.  More recently, I am able to connect and get work, but have massive stale/rejected shares (close to 10%) that are clearly not because a new block has been found in the network. 

Let me know when the issues have been resolved, and I'll switch back from my backup pools.