Sukrim
Legendary
Offline
Activity: 2618
Merit: 1007
|
|
July 19, 2011, 11:11:12 PM |
|
With "decent" hashrate I mean a rate that is above a certain point to avoid hoppers having the 1Mh/s miner running just to be connected during the part after the 43.5% point.
This means you will screw all miners with hash rates below a "decent hash rate" + force hoppers to have a "decent hash rate"+1 MH/s. Also there is no measure of hash rates to a pool - it is only an average over a certain time. If you out of bad luck don't happen to find any shares in this period of time, you seem to have 0 MH/s to the pool, even if you in reality have 1 GH/s. Congratulations to be another pool stepping away from block limited proportional scoring!
|
|
|
|
1bitc0inplz (OP)
Member
Offline
Activity: 112
Merit: 10
|
|
July 20, 2011, 01:18:42 AM |
|
How about this: PROP but.
I do think that you will enjoy our new reward system. The great thing about it is that it will reduce your variance. There is nothing to not love about that. As for Prop., I'm afraid that was off the table for the moment we started discussing alternative reward systems. Proportional is just inherently flawed. The fact that it's payouts are predictably variant is enough to game it. I could continue to hide stats until I'm blue in the face, and they will be tit-for-tat with our changes. I've already wasted a week keeping them at bay, and in that week no new features have been coded. I'd rather be coding features than hacking up our API, leaderboard, etc.
|
|
|
|
muyoso
Member
Offline
Activity: 84
Merit: 10
|
|
July 20, 2011, 02:06:04 AM |
|
If I had to guess, I would say that people mining at your pool made more when the pool was being hopped than they did when it wasn't. Sure they may have gotten a slightly smaller payout per block, but this 50Ghash pool discovered 9 blocks in under 5 days. That is ridiculous for a 50Ghash pool. Almost 2 blocks a day. I don't have the statistics obviously, but I'd bet miners at this pool absolutely made more per 24 hour period while the pool was being hopped than they will/have without the hopping.
A person with 1Ghash would have made around 3.24 bitcoins during the same 5 day period if the pool wasn't hopped with an average round time of 37 hours for 50ghash/s at the difficulty of 1563027. That same person ACTUALLY made 3.75 bitcoins while the pool was being hopped over the 5 day period before you started messing with the API.
|
I drink it up!
|
|
|
1bitc0inplz (OP)
Member
Offline
Activity: 112
Merit: 10
|
|
July 20, 2011, 02:15:29 AM |
|
If I had to guess, I would say that people mining at your pool made more when the pool was being hopped than they did when it wasn't. Sure they may have gotten a slightly smaller payout per block, but this 50Ghash pool discovered 9 blocks in under 5 days. That is ridiculous for a 50Ghash pool. Almost 2 blocks a day. I don't have the statistics obviously, but I'd bet miners at this pool absolutely made more per 24 hour period while the pool was being hopped than they will/have without the hopping.
Actually, it's reduced our member's payouts to approximately 75% of what they would have been. Where else do you think the hopper's take their additional income from Here is a simulation using parameters in line with the hash rate flucturations our pool was receiving: === Prop ===
| Average | Median | Min | Max | Variance -------------------------+---------------+---------------+---------------+---------------+------------------- Honest Earnings | 720.8945 | 721.8807 | 690.2052 | 739.4789 | 78.5162 Hopper Earnings | 1848.3674 | 1842.9480 | 1783.1904 | 1947.1667 | 1126.3150 Honest Shares | 28326150.0621 | 28356810.6817 | 25249228.2421 | 30468502.4626 | 1278454932311.3887 Hopper Shares | 31161689.6574 | 31166729.5802 | 29928070.3722 | 32233605.3469 | 178450276113.6692 Honest Payout Percentage | 76.4419 | 76.4535 | 71.7608 | 82.0071 | 5.6908 Hopper Payout Percentage | 177.9985 | 177.1536 | 166.3651 | 195.1847 | 23.3973
|
|
|
|
muyoso
Member
Offline
Activity: 84
Merit: 10
|
|
July 20, 2011, 02:18:07 AM |
|
If I had to guess, I would say that people mining at your pool made more when the pool was being hopped than they did when it wasn't. Sure they may have gotten a slightly smaller payout per block, but this 50Ghash pool discovered 9 blocks in under 5 days. That is ridiculous for a 50Ghash pool. Almost 2 blocks a day. I don't have the statistics obviously, but I'd bet miners at this pool absolutely made more per 24 hour period while the pool was being hopped than they will/have without the hopping.
Actually, it's reduced our member's payouts to approximately 75% of what they would have been. We're else do you think the hopper's take their additional income from And its more than doubled the NUMBER of payouts. Average time to generate a block with 50Ghash would be 37 hours. In the 5 days where your pool found 9 blocks, you regularly would have only found ~3.24 blocks. Its your pool and I respect whatever you want to do with it. It seems counter-intuitive that hopping actually gains more BTC for your users, but in many cases especially for smaller pools it happens. Also, let me just say that your site is EXCEPTIONALLY good. I really like the layout and the design. Simple and elegant. Fantastic work. It would be awesome if you would add on the Round History page the amount earned per solved block next to the status column and a total at the bottom.
|
I drink it up!
|
|
|
1bitc0inplz (OP)
Member
Offline
Activity: 112
Merit: 10
|
|
July 20, 2011, 02:29:24 AM |
|
Also, let me just say that your site is EXCEPTIONALLY good. I really like the layout and the design. Simple and elegant. Fantastic work. It would be awesome if you would add on the Round History page the amount earned per solved block next to the status column and a total at the bottom.
Thank you, we appreciate the compliment. Disagreement aside, those changes you are talking about are the type of thing I'd love to get back to coding Honestly, I'd love to see the round history show who solved it, how many shares it too, the "luck", and (like you said) perhaps some payout stats.
|
|
|
|
TheSeven
|
|
July 20, 2011, 08:25:16 AM |
|
It seems counter-intuitive that hopping actually gains more BTC for your users, but in many cases especially for smaller pools it happens.
This was purely one thing: luck.
|
My tip jar: 13kwqR7B4WcSAJCYJH1eXQcxG5vVUwKAqY
|
|
|
organofcorti
Donator
Legendary
Offline
Activity: 2058
Merit: 1007
Poor impulse control.
|
|
July 20, 2011, 08:41:09 AM |
|
It seems counter-intuitive that hopping actually gains more BTC for your users, but in many cases especially for smaller pools it happens.
This was purely one thing: luck. There is one other thing though - you're getting a 100ghps boost for part of the round which must increase the average hashrate for the round. This in turn significantly will decrease the average time to solev a block and so on average decrease variance. I think this is what muyoso meant. You could probably work out what the average pool hashrate increase was - for a short round it would be significant (which makes short rounds take even less time) and for long rounds it would have less of an effect. But even though rounds are being solved more quickly there are fewer shares per miner as opposed to no 'hopping and so less coinage. That's too much to pay for an increased average total hashrate and the low cost easy to calculate nature of proportional. 'Non hoppers still lose out.
|
|
|
|
paraipan
In memoriam
Legendary
Offline
Activity: 924
Merit: 1004
Firstbits: 1pirata
|
|
July 20, 2011, 10:37:52 AM |
|
yep, just like I thought, it's difficult to see if the hopping is good or bad for pools or hoppers... too many parameters in the equation. hurray for bitp.it changing their payout system, will have to see if theirs is better altough
|
BTCitcoin: An Idea Worth Saving - Q&A with bitcoins on rugatu.com - Check my rep
|
|
|
organofcorti
Donator
Legendary
Offline
Activity: 2058
Merit: 1007
Poor impulse control.
|
|
July 20, 2011, 10:59:17 AM |
|
yep, just like I thought, it's difficult to see if the hopping is good or bad for pools or hoppers... too many parameters in the equation. hurray for bitp.it changing their payout system, will have to see if theirs is better altough
No, hopping is provably good for hoppers and pools and provably bad for non-hoppers. And yes there are a bunch of parameters but luckily people much smarter than me laid the groundwork!
|
|
|
|
luffy
|
|
July 20, 2011, 02:03:01 PM |
|
the base of economy and profit is how you can outsmart other people! Thus, it is only natural to try to find ways of maximizing the profit and let others "die"
|
|
|
|
hugolp
Legendary
Offline
Activity: 1148
Merit: 1001
Radix-The Decentralized Finance Protocol
|
|
July 20, 2011, 04:47:54 PM |
|
the base of economy and profit is how you can outsmart other people in a way that develops society! Thus, it is only natural to try to find ways of maximizing the profit and let others "die" Corrected it for you.
|
|
|
|
muyoso
Member
Offline
Activity: 84
Merit: 10
|
|
July 21, 2011, 02:35:39 AM Last edit: July 21, 2011, 10:12:01 PM by muyoso |
|
It seems counter-intuitive that hopping actually gains more BTC for your users, but in many cases especially for smaller pools it happens.
This was purely one thing: luck. Amazing how that luck ended on the first round after the pool started defending against hopping. From 9 blocks in 143 hours WITH hopping to what will eventually be 1 block in now 70 hours and counting WITHOUT hopping. Edit: 90 hours and counting now without a block. At this point with hopping bitp.it had solved 6 blocks. 6 blocks with slightly reduced payment seems better than 1 block with full payment to me.
|
I drink it up!
|
|
|
organofcorti
Donator
Legendary
Offline
Activity: 2058
Merit: 1007
Poor impulse control.
|
|
July 21, 2011, 02:48:30 AM |
|
It seems counter-intuitive that hopping actually gains more BTC for your users, but in many cases especially for smaller pools it happens.
This was purely one thing: luck. Amazing how that luck ended on the first round after the pool started defending against hopping. From 9 blocks in 143 hours WITH hopping to what will eventually be 1 block in now 70 hours and counting WITHOUT hopping. Yes, but each share is worth more than when hoppers are present. The increased hashrate that helps solve blocks quicker means that even with more blocks you non-hoppers end up with less cash than they would if the blocks had been solved without the hashrate boost. Also keep in mind that the pool profits more with the hopper boost than without, so they're changing scoring for the pool member's benefit more than their own direct benefit. Finally, all you can say about the variance of total shares in a block is: long block is long.
|
|
|
|
organofcorti
Donator
Legendary
Offline
Activity: 2058
Merit: 1007
Poor impulse control.
|
|
July 21, 2011, 02:50:34 AM |
|
If I had to guess, I would say that people mining at your pool made more when the pool was being hopped than they did when it wasn't. Sure they may have gotten a slightly smaller payout per block, but this 50Ghash pool discovered 9 blocks in under 5 days. That is ridiculous for a 50Ghash pool. Almost 2 blocks a day. I don't have the statistics obviously, but I'd bet miners at this pool absolutely made more per 24 hour period while the pool was being hopped than they will/have without the hopping.
Actually, it's reduced our member's payouts to approximately 75% of what they would have been. Where else do you think the hopper's take their additional income from Here is a simulation using parameters in line with the hash rate flucturations our pool was receiving: === Prop ===
| Average | Median | Min | Max | Variance -------------------------+---------------+---------------+---------------+---------------+------------------- Honest Earnings | 720.8945 | 721.8807 | 690.2052 | 739.4789 | 78.5162 Hopper Earnings | 1848.3674 | 1842.9480 | 1783.1904 | 1947.1667 | 1126.3150 Honest Shares | 28326150.0621 | 28356810.6817 | 25249228.2421 | 30468502.4626 | 1278454932311.3887 Hopper Shares | 31161689.6574 | 31166729.5802 | 29928070.3722 | 32233605.3469 | 178450276113.6692 Honest Payout Percentage | 76.4419 | 76.4535 | 71.7608 | 82.0071 | 5.6908 Hopper Payout Percentage | 177.9985 | 177.1536 | 166.3651 | 195.1847 | 23.3973
Yep, that's about what I get at most pools. With variance it's usually 140% to 200%.
|
|
|
|
1bitc0inplz (OP)
Member
Offline
Activity: 112
Merit: 10
|
|
July 22, 2011, 05:04:17 AM |
|
Just a quick update for everyone. The implementation of ESMPPS is coming along nicely. We hope to have this rolled out "soon". With it, we will also be re-enabling stats and APIs. Also, we are going ahead and supporting 8-digit payouts. The new behavior will be full 8-digit payouts, with a minimum payout of 0.01 BTC. I thought I'd share some good news with everyone
|
|
|
|
phelix
Legendary
Offline
Activity: 1708
Merit: 1020
|
|
July 22, 2011, 06:21:25 AM |
|
Just a quick update for everyone. The implementation of ESMPPS is coming along nicely. We hope to have this rolled out "soon". With it, we will also be re-enabling stats and APIs. Also, we are going ahead and supporting 8-digit payouts. The new behavior will be full 8-digit payouts, with a minimum payout of 0.01 BTC. I thought I'd share some good news with everyone yippieh
|
|
|
|
rearwheels
|
|
July 22, 2011, 10:01:31 AM |
|
Hi,
In the JSON API for user stats, "estimated_reward" is spelt as "estimated_reword".
Regards, Koo
|
|
|
|
organofcorti
Donator
Legendary
Offline
Activity: 2058
Merit: 1007
Poor impulse control.
|
|
July 22, 2011, 10:07:38 AM |
|
Hi,
In the JSON API for user stats, "estimated_reward" is spelt as "estimated_reword".
Regards, Koo
Yes, you'll need to rew ord it. A HA HA HA HA HAAAAA! *sigh*
|
|
|
|
2weiX
Legendary
Offline
Activity: 2058
Merit: 1005
this space intentionally left blank
|
|
July 22, 2011, 10:19:25 AM |
|
love your pool, second everything about the interface and the setup of workers. i'd love to have automated payout, tho. i mine to three different pools and yours is the one with manual payout - if you would please incorporate the OPTION to be payed every _____ BTC? if possible, plz add email alert.
also, if you could get David Greene of Miner Status for Android to incorporate bitpit, that'd be swell.
|
|
|
|
|