organofcorti
Donator
Legendary
Offline
Activity: 2058
Merit: 1007
Poor impulse control.
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:02:44 PM |
|
Well, it's not possible to mine forever. If nothing else, eventually people will stop making transactions, and eventually the block reward will go to zero. Maybe not forever, but if Bitcoin mining stops in our lifetimes, it means Bitcoin was a failure. Yeah, but the chance that people who have been continuously mining at Eligius since the start of CPPSRB will ever get 100% shares rewarded during our lifetimes is quite low. Ignoring transaction fees, it's probably immeasurably low. I'd like to see your analysis. I'm not sure you are correct.
|
|
|
|
wizkid057 (OP)
Legendary
Offline
Activity: 1223
Merit: 1006
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:04:34 PM |
|
One advantage of PPLNS is that you don't have to wait around potentially forever to find out how much you're going to make.
Yep, you immediately know if you made nothing at all. ![Wink](https://bitcointalk.org/Smileys/default/wink.gif) So that 4.72% (with transaction fees you'd get 102.72%) is solely due to orphans? That seems like a very high orphan rate.
I'm.... not quite sure what these numbers are exactly? ![Huh](https://bitcointalk.org/Smileys/default/huh.gif) You can see for yourself from the stats that just based on counts Eligius's long-term orphan rate is just over 2%. Factoring in the subsidy halving the earnings-based orphan rate is lower, at almost exactly 2%. (Our orphan rate decreased a bit after changing datacenters after the subsidy halving) This is pretty normal, and actually quite low. Look at Ghash.io's orphan rate... was up at almost 7% at one point. Well, if you add in transaction fees, which apparently you're doing, then 100% is quite possible.
Sure, txn fees *can* push it to 100%... but on average they do not, as there is less than a 2% increase in block reward due to txn fees on average.
|
|
|
|
anth0ny
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:05:41 PM |
|
For the expected value, I am not sure. I'll cede this one to WK or un_ordinateur. Over time it should be 100% PPS
Maybe if the block reward never went down, and the difficulty never changed... Neither of these things matter for calculating % shares rewarded in terms of maximum PPS, since the formula for 100% PPS at any given time is (block_reward/difficulty)*work_difficulty. Sure it matters. The people at the very end of the shelved share list have virtually no chance of ever getting paid for those shares, in large part because the block rewards keep going down and the difficulty keeps going up. The expected long term reward with any pool reward system, unless the pool is somehow subsidizing orphans (which means it probably has a fee in excess of the orphan % anyway...) is 100% PPS minus orphan % minus fees (if any).
Are you assuming the pool will last forever? Eventually Eligius will cease to exist. Probably before Bitcoin ceases to exist, and probably during our lifetimes for that matter. At that point, if there are any shelved shares, then Eligius has paid out less than 100% PPS (including transaction fees) minus orphan %. Those shelved shares are never paid.
|
|
|
|
wizkid057 (OP)
Legendary
Offline
Activity: 1223
Merit: 1006
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:08:33 PM |
|
For the expected value, I am not sure. I'll cede this one to WK or un_ordinateur. Over time it should be 100% PPS
Maybe if the block reward never went down, and the difficulty never changed... Neither of these things matter for calculating % shares rewarded in terms of maximum PPS, since the formula for 100% PPS at any given time is (block_reward/difficulty)*work_difficulty. Sure it matters. The people at the very end of the shelved share list have virtually no chance of ever getting paid for those shares, in large part because the block rewards keep going down and the difficulty keeps going up. The expected long term reward with any pool reward system, unless the pool is somehow subsidizing orphans (which means it probably has a fee in excess of the orphan % anyway...) is 100% PPS minus orphan % minus fees (if any).
Are you assuming the pool will last forever? Eventually Eligius will cease to exist. Probably before Bitcoin ceases to exist, and probably during our lifetimes for that matter. At that point, if there are any shelved shares, then Eligius has paid out less than 100% PPS (including transaction fees) minus orphan %. Those shelved shares are never paid. You apparently are not understanding why the shelved shares persist for any extended period in the first place. Those shares statistically are the "minus orphan %", not in addition to it, and it is to be expected that not all of them will ever be paid, barring some ridiculous amount of luck. This doesn't change that the long term (6+ months) variance in % shares rewarded wont average out to roughly 100% minus orphan %.
|
|
|
|
smooth
Legendary
Offline
Activity: 2968
Merit: 1198
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:10:12 PM |
|
The expected long term reward with any pool reward system, unless the pool is somehow subsidizing orphans (which means it probably has a fee in excess of the orphan % anyway...) is 100% PPS minus orphan % minus fees (if any).
Are you assuming the pool will last forever? No, I simply made a mathematical statement of fact. At that point, if there are any shelved shares, then Eligius has paid out less than 100% PPS (including transaction fees) minus orphan %. Those shelved shares are never paid.
As I said.
|
|
|
|
wizkid057 (OP)
Legendary
Offline
Activity: 1223
Merit: 1006
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:11:54 PM |
|
The expected long term reward with any pool reward system, unless the pool is somehow subsidizing orphans (which means it probably has a fee in excess of the orphan % anyway...) is 100% PPS minus orphan % minus fees (if any).
Are you assuming the pool will last forever? No, I simply made a mathematical statement of fact. At that point, if there are any shelved shares, then Eligius has paid out less than 100% PPS (including transaction fees) minus orphan %. Those shelved shares are never paid.
As I said. I want to point out that I read anth0ny's " paid out less than 100% PPS (including transaction fees) minus orphan %. Those shelved shares are never paid." statement as the pool paying less than 100% already somehow and in addition subtracting orphan %, which isn't the case.
|
|
|
|
anth0ny
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:13:03 PM |
|
One advantage of PPLNS is that you don't have to wait around potentially forever to find out how much you're going to make.
Yep, you immediately know if you made nothing at all. ![Wink](https://bitcointalk.org/Smileys/default/wink.gif) And can abandon the address and stop monitoring it instead of keeping it around forever because you might get a few satoshi some time between now and the end of the universe. So that 4.72% (with transaction fees you'd get 102.72%) is solely due to orphans? That seems like a very high orphan rate.
I'm.... not quite sure what these numbers are exactly? ![Huh](https://bitcointalk.org/Smileys/default/huh.gif) You can see for yourself from the stats that just based on counts Eligius's long-term orphan rate is just over 2%. Factoring in the subsidy halving the earnings-based orphan rate is lower, at almost exactly 2%. (Our orphan rate decreased a bit after changing datacenters after the subsidy halving) This is pretty normal, and actually quite low. Look at Ghash.io's orphan rate... was up at almost 7% at one point. Well, if you add in transaction fees, which apparently you're doing, then 100% is quite possible.
Sure, txn fees *can* push it to 100%... but on average they do not, as there is less than a 2% increase in block reward due to txn fees on average. My bad. I misread https://blockchain.info/stats .
|
|
|
|
anth0ny
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:15:53 PM |
|
For the expected value, I am not sure. I'll cede this one to WK or un_ordinateur. Over time it should be 100% PPS
Maybe if the block reward never went down, and the difficulty never changed... Neither of these things matter for calculating % shares rewarded in terms of maximum PPS, since the formula for 100% PPS at any given time is (block_reward/difficulty)*work_difficulty. Sure it matters. The people at the very end of the shelved share list have virtually no chance of ever getting paid for those shares, in large part because the block rewards keep going down and the difficulty keeps going up. The expected long term reward with any pool reward system, unless the pool is somehow subsidizing orphans (which means it probably has a fee in excess of the orphan % anyway...) is 100% PPS minus orphan % minus fees (if any).
Are you assuming the pool will last forever? Eventually Eligius will cease to exist. Probably before Bitcoin ceases to exist, and probably during our lifetimes for that matter. At that point, if there are any shelved shares, then Eligius has paid out less than 100% PPS (including transaction fees) minus orphan %. Those shelved shares are never paid. You apparently are not understanding why the shelved shares persist for any extended period in the first place. Those shares statistically are the "minus orphan %", not in addition to it, and it is to be expected that not all of them will ever be paid, barring some ridiculous amount of luck. This doesn't change that the long term (6+ months) variance in % shares rewarded wont average out to roughly 100% minus orphan %. The shelved shares are orphan %, plus bad luck %, plus block withholding %, minus transaction fee %... Right?
|
|
|
|
smooth
Legendary
Offline
Activity: 2968
Merit: 1198
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:16:40 PM |
|
Keep in mind that 100% PPS is not statistically possible long term with *any* reward system mainly due to orphaned blocks.
Well as I've occasionally said before, when orphan a block you should perhaps just dump 25 BTC of shares, not put them back on the share log. That keeps the expected value of shares going on to the log equal to the shares coming off. This would mean the pool empties the share log more often, and I don't know the specified behavior under that condition. Theoretically this can already happen with a run of very good luck (but is unlikely). Perhaps this would satisfy people who say that the share log will never empty and therefore expected value is lower than PPLNS. Given the cumulative effect of orphans they are correct in finite time and possibly even in infinite time. It would increase variance a bit, because orphaned shares would not be paid, ever.
|
|
|
|
aetaf
Newbie
Offline
Activity: 3
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:16:57 PM |
|
The force 1000TH is jumping up and down, income decreases a lot? ![Huh](https://bitcointalk.org/Smileys/default/huh.gif) ?
|
|
|
|
anth0ny
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:18:30 PM |
|
The expected long term reward with any pool reward system, unless the pool is somehow subsidizing orphans (which means it probably has a fee in excess of the orphan % anyway...) is 100% PPS minus orphan % minus fees (if any).
Are you assuming the pool will last forever? No, I simply made a mathematical statement of fact. So, what's your definition of 100% PPS?
|
|
|
|
wizkid057 (OP)
Legendary
Offline
Activity: 1223
Merit: 1006
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:19:30 PM |
|
The shelved shares are orphan %, plus bad luck %, plus block withholding %, minus transaction fee %... Right?
Not quite. If you must simplify it to something like this you must also include "minus good luck %". --- Keep in mind that 100% PPS is not statistically possible long term with *any* reward system mainly due to orphaned blocks.
Well as I've occasionally said before, when orphan a block you should perhaps just dump 25 BTC of shares, not put them back on the share log. That keeps the expected value of shares going on to the log equal to the shares coming off. This would mean the pool empties the share long more often, and I don't know the specified behavior under that condition. Theoretically this can already happen with a run of very good luck (but is unlikely). Perhaps this would satisfy people who say that the share log will never empty and therefore expected value is lower than PPLNS. Given the cumulative effect of orphans they are correct in finite time and possibly even in infinite time. It would increase variance a bit, because orphaned shares would not be paid, ever. I'd much rather keep a mathematically sound reward system in operation rather than break it completely in an attempt to cater to those who do not fully understand it, honestly.
|
|
|
|
wizkid057 (OP)
Legendary
Offline
Activity: 1223
Merit: 1006
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:20:30 PM |
|
The force 1000TH is jumping up and down, income decreases a lot? ![Huh](https://bitcointalk.org/Smileys/default/huh.gif) ? The effect of the larger miners on the pool tinkering with their hash power has no effect on your earnings...
|
|
|
|
anth0ny
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:21:20 PM |
|
This would mean the pool empties the share long more often, and I don't know the specified behavior under that condition.
In the case of Eligius, it would go toward the extra credit under the old system. That's why it's not 100% expected value. If you get unlucky, you get unlucky. If you get lucky, the reward goes to the people who mined before you.
|
|
|
|
smooth
Legendary
Offline
Activity: 2968
Merit: 1198
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:22:39 PM |
|
Well as I've occasionally said before, when orphan a block you should perhaps just dump 25 BTC of shares, not put them back on the share log. That keeps the expected value of shares going on to the log equal to the shares coming off.
This would mean the pool empties the share long more often, and I don't know the specified behavior under that condition. Theoretically this can already happen with a run of very good luck (but is unlikely). Perhaps this would satisfy people who say that the share log will never empty and therefore expected value is lower than PPLNS. Given the cumulative effect of orphans they are correct in finite time and possibly even in infinite time. It would increase variance a bit, because orphaned shares would not be paid, ever.
I'd much rather keep a mathematically sound reward system in operation rather than break it completely in an attempt to cater to those who do not fully understand it, honestly. The system I described is sound, as far as I know. I'm not actually sure what your definition of sound is though.
|
|
|
|
wizkid057 (OP)
Legendary
Offline
Activity: 1223
Merit: 1006
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:24:15 PM |
|
So, what's your definition of 100% PPS?
100% PPS = (block_reward/network_difficulty) This would mean the pool empties the share long more often, and I don't know the specified behavior under that condition.
In the case of Eligius, it would go toward the extra credit under the old system. That's why it's not 100% expected value. If you get unlucky, you get unlucky. If you get lucky, the reward goes to the people who mined before you. Any new miner basically starts with a fresh slate, and their shares enter the top of the share log with an estimated reward of 100% immediately. Older miners have zero effect on this, ever. You don't understand how the system works, please make sure you understand properly before making FUD-ish statements like this.
|
|
|
|
smooth
Legendary
Offline
Activity: 2968
Merit: 1198
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:25:20 PM |
|
If you get unlucky, you get unlucky. If you get lucky, the reward goes to the people who mined before you.
This is true under PPLNS as well, so I'm not sure your point. Let's say someone mined yesterday (N>1 day) and then stopped. You are mining today and you get lucky. Some of that luck (potentially a lot of it) shifts to the person who mined yesterday. You have the same opportunity to get lucky tomorrow though, as does someone under the CPPSRB.
|
|
|
|
smooth
Legendary
Offline
Activity: 2968
Merit: 1198
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:27:17 PM |
|
Any new miner basically starts with a fresh slate, and their shares enter the top of the share log with an estimated reward of 100% immediately.
(Ignoring orhpans.) Their estimated reward is not really 100% though, for any finite time. As long as there is a possibility it will be <100% and no possibility it will be >100% then the expected value is <100%
|
|
|
|
wizkid057 (OP)
Legendary
Offline
Activity: 1223
Merit: 1006
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:28:10 PM |
|
Well as I've occasionally said before, when orphan a block you should perhaps just dump 25 BTC of shares, not put them back on the share log. That keeps the expected value of shares going on to the log equal to the shares coming off.
This would mean the pool empties the share long more often, and I don't know the specified behavior under that condition. Theoretically this can already happen with a run of very good luck (but is unlikely). Perhaps this would satisfy people who say that the share log will never empty and therefore expected value is lower than PPLNS. Given the cumulative effect of orphans they are correct in finite time and possibly even in infinite time. It would increase variance a bit, because orphaned shares would not be paid, ever.
I'd much rather keep a mathematically sound reward system in operation rather than break it completely in an attempt to cater to those who do not fully understand it, honestly. The system I described is sound, as far as I know. I'm not actually sure what your definition of sound is though. By definition under CPPSRB there can be no "orphaned shares." Shares are paid when the pool gets income, and are left alone when it doesn't. What happens if the pool gets several quick successive orphans in a row with far less than 25 BTC worth of mining (at 100% PPS) done to find them, but much more of the share log is dumped? Doesn't make much sense... Any new miner basically starts with a fresh slate, and their shares enter the top of the share log with an estimated reward of 100% immediately.
(Ignoring orhpans.) Their estimated reward is not really 100% though, for any finite time. As long as there is a possibility it will be <100% and no possibility it will be >100% then the expected value is <100% I was specifically referring to the shares they submit immediately upon starting to mine... they will be at the top of the share log with estimated reward of 100%. The shares before theirs in the share log have no affect on their own earnings.
|
|
|
|
anth0ny
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
February 24, 2014, 11:33:47 PM |
|
This would mean the pool empties the share long more often, and I don't know the specified behavior under that condition.
In the case of Eligius, it would go toward the extra credit under the old system. That's why it's not 100% expected value. If you get unlucky, you get unlucky. If you get lucky, the reward goes to the people who mined before you. Any new miner basically starts with a fresh slate, and their shares enter the top of the share log with an estimated reward of 100% immediately. Older miners have zero effect on this, ever. Start out as a new miner. Mine 16 shares. With average luck, you'll get paid for those 16 shares. With above average luck, you'll get paid for those 16 shares, and some people who mined before you also get paid for their shares. With bad luck, you get nothing. Expected value is less than 100%, even if there are no orphans, unless you assume that Eligius lasts forever. The expected long term reward with any pool reward system, unless the pool is somehow subsidizing orphans (which means it probably has a fee in excess of the orphan % anyway...) is 100% PPS minus orphan % minus fees (if any).
So, this is saying that it is expected for Eligius to last forever. 100% minus 0% minus 0% is not less than 100%. The average expected value of everyone is 100%, but that doesn't mean the average expected value of each person is 100%. You don't understand how the system works, please make sure you understand properly before making FUD-ish statements like this.
What was wrong about what I said above?
|
|
|
|
|