Why N/A on p2pool's blocks?
It is nice to see the top 4 pools combined only have 48% of hashing power. Even top 5 is only 51%. Hopefully the trend continues and we see the top 5 combined fall under the 50% mark.
Topic: Weekly pool and network statistics - page 28. (Read 91252 times)
The luck index has changed since last week to better reflect the differences in luck, and is now:
Some highlights:
I didn't add your suggestion, SgtSpike, I'll try to do so next week
Weekly average pool hashrates, 15th May 2012:
Weekly average % of network hashrate to date:
Weekly luck index to date:
Code:
Luck index = log(Percentage of network blocks solved / Percentage of network hashrate)
Luck index = 0 is average luck
Luck index > 0 is good luck
Luck index < 0 is poor luck
Some highlights:
- Deepbit and Slush continue to lose hashrate as a percentage of network hashrate; Ozcoin getting quite close to Slush's hashrate now.
- Bitlc and Maxbtc continue their hashrate plummet.
- Itzod recovers some hashrate.
- Bitminter improves hashrate again, for the fourth week in a row.
- Itzod and Ozcoin continue their lucky streak.
I didn't add your suggestion, SgtSpike, I'll try to do so next week
Weekly average pool hashrates, 15th May 2012:
Weekly average % of network hashrate to date:
Weekly luck index to date:
Thank you! It took a long time to get there. This is a big milestone for BitMinter. 
Well, there's been quite a shuffle to the pools subforum top 10. Bitminter is in! Well done Dr. H. We also see the return of Mt. Red, first time since their heady proportional days.
On the other hand we say good bye to Bitclockers and Bitlc.net. Maybe they'll both switch to a fair payout structure soon and be back.
On the other hand we say good bye to Bitclockers and Bitlc.net. Maybe they'll both switch to a fair payout structure soon and be back.
That won't be as good a comparison as you might think - it would be more accurate to compare total submitted shares divided by difficulty to total blocks solved if you want to look at luck from a miner payout point of view rather than as a proportion of network luck. I'll look at adding something more like what you want next week.
Ah, yes, good point..!
That won't be as good a comparison as you might think - it would be more accurate to compare total submitted shares divided by difficulty to total blocks solved if you want to look at luck from a miner payout point of view rather than as a proportion of network luck. I'll look at adding something more like what you want next week.
This makes me even more interested to run my singles against each pool and see if my payouts match your calculated luck. Now that should be VERY telling!
What, 19% above expected? BitMinter did 42% above expected.
Watch us do it again next week!
Watch us do it again next week!
Nice pool luck there, Graeme. Ozcoin's lookin' good.
Fixed that!
<3
and thanks . things are looking good atm, more good stuff to come - stay tuned
Nice pool luck there, Graeme. Ozcoin's lookin' good.
Fixed that!
Nice pool luck there, Graeme. Ozcoin's lookin' good.
heh thanks , we had a few ugly days, but yesterday caught us up nicely * Graet tries not to anger the luck gods by talking about luck 
egluck has been excellent here recently
*knocks on wood*
Since you talked about it, you have automatically killed it. Never discuss luck because that breaks things.*knocks on wood*
Nice pool luck there, Graeme. Ozcoin's lookin' good.
good work again
thanks
thanks
Thanks for the pool 'luck' suggestion, Sgt Spike.
Weekly average pool hashrates, 8th May 2012:
Weekly average % of network hashrate to date:
Weekly luck index to date:
Weekly average pool hashrates, 8th May 2012:
Weekly average % of network hashrate to date:
Weekly luck index to date:
Also, it's easy to record the number of shares you submitted locally. If your valid shares minus your stale/invalid shares don't add up to the number of shares you're credited with, or if your stales (as recorded on the pool website) are much higher than you record locally, you'll move to another pool.
It's easier just to credit miners with shares correctly, and just fake round lengths. Or just take the whole reward for short blocks. Neither of these scams work on PPS pools of course - for PPS you just have to keep your eye on submitted vs accepted shares and on your stale/invalid share to valid share ratio.
Also won't work at Slush's pool since you don't get to know your accepted shares, and I'm not sure if you can back calculate from the score since it get normalised hourly - that is intraround sometimes. Lucky Slush is a trustworthy guy.
It's easier just to credit miners with shares correctly, and just fake round lengths. Or just take the whole reward for short blocks. Neither of these scams work on PPS pools of course - for PPS you just have to keep your eye on submitted vs accepted shares and on your stale/invalid share to valid share ratio.
Also won't work at Slush's pool since you don't get to know your accepted shares, and I'm not sure if you can back calculate from the score since it get normalised hourly - that is intraround sometimes. Lucky Slush is a trustworthy guy.
But isn't the pool hashrate estimated based on the number of shares? Couldn't the pool misrepresent the number of shares? Or the payout per block? Or etc, etc, etc?
I think there's a lot of wiggle room for pools to potentially skim or scrap coins off the top without anyone noticing. That's why I'd be interested in a direct comparison on the same hardware between 10 of the most popular pools.
I expect the hashes per second displayed by a pool to be honest numbers.I think there's a lot of wiggle room for pools to potentially skim or scrap coins off the top without anyone noticing. That's why I'd be interested in a direct comparison on the same hardware between 10 of the most popular pools.
If a pool advertises 500 GH/s, I would expect 500 * 1 GH/s rigs are mining at that pool, for example.
Calculating hashing power any other way is deceptive because the value is altered to appear other than it is.
The easiest way to do something like you suggest that's also hard to detect is to simply keep round rewards for very short rounds.
Even if you're blatant about recording abnormal round lengths - like bitclockers, although I'm not saying they stole any round rewards, just faked rounds lengths - it will take a while to record enough data to prove it unless you're deepbit.
If you go to the Neighbourhood poolwatch blog (see my sig) there's more info about these types of problem.
Even if you're blatant about recording abnormal round lengths - like bitclockers, although I'm not saying they stole any round rewards, just faked rounds lengths - it will take a while to record enough data to prove it unless you're deepbit.
If you go to the Neighbourhood poolwatch blog (see my sig) there's more info about these types of problem.
It'd be interesting to somehow compare the average hashrate to the pool payouts. See if there are any pools that are consistently paying out less than the others with regards to a payout/hashrate ratio.
I suppose it would require a miner to be pointed at each pool, and assume all else being equal.
Doesn't require mining, only a calculator, provided the pools post enough stats to make calculations.I suppose it would require a miner to be pointed at each pool, and assume all else being equal.
Quote from: ssateneth
1 share ≈ 2^32 hashes or 4294.967296 megahashes
60*60*24 seconds in a day = 86,400 seconds/day
1010MH/sec = 87,264,000 Megahashes/day
Shares/day = (MHash/day) / (Megahashes/share)
Shares/day = 87264000 / 4294.967296
Shares/day ≈ 20317.73
Payout/day = 20317.73 * 0.00003 Bitcoins/share
Payout/day = 0.60953199 Bitcoins
Above, the Payout Per Share is .0000360*60*24 seconds in a day = 86,400 seconds/day
1010MH/sec = 87,264,000 Megahashes/day
Shares/day = (MHash/day) / (Megahashes/share)
Shares/day = 87264000 / 4294.967296
Shares/day ≈ 20317.73
Payout/day = 20317.73 * 0.00003 Bitcoins/share
Payout/day = 0.60953199 Bitcoins
Eclipse has a 12% better payout currently over Ozcoin, while Eligius doesn't publish, that I could find, average shares round.
{Pools total mega hashes} * 86,400 / 4294.967296 * nPPS / {Pools total Gigahashes} = ⊅BTC per GH/s per day
nPPS can be any reasonable arbitrary value, .00003 can be used. It won't represent actual payouts just differences between pools.
I think there's a lot of wiggle room for pools to potentially skim or scrap coins off the top without anyone noticing. That's why I'd be interested in a direct comparison on the same hardware between 10 of the most popular pools.
It'd be interesting to somehow compare the average hashrate to the pool payouts. See if there are any pools that are consistently paying out less than the others with regards to a payout/hashrate ratio.
I suppose it would require a miner to be pointed at each pool, and assume all else being equal.
Doesn't require mining, only a calculator, provided the pools post enough stats to make calculations.I suppose it would require a miner to be pointed at each pool, and assume all else being equal.
Quote from: ssateneth
1 share ≈ 2^32 hashes or 4294.967296 megahashes
60*60*24 seconds in a day = 86,400 seconds/day
1010MH/sec = 87,264,000 Megahashes/day
Shares/day = (MHash/day) / (Megahashes/share)
Shares/day = 87264000 / 4294.967296
Shares/day ≈ 20317.73
Payout/day = 20317.73 * 0.00003 Bitcoins/share
Payout/day = 0.60953199 Bitcoins
Above, the Payout Per Share is .0000360*60*24 seconds in a day = 86,400 seconds/day
1010MH/sec = 87,264,000 Megahashes/day
Shares/day = (MHash/day) / (Megahashes/share)
Shares/day = 87264000 / 4294.967296
Shares/day ≈ 20317.73
Payout/day = 20317.73 * 0.00003 Bitcoins/share
Payout/day = 0.60953199 Bitcoins
Eclipse has a 12% better payout currently over Ozcoin, while Eligius doesn't publish, that I could find, average shares round.
{Pools total mega hashes} * 86,400 / 4294.967296 * nPPS / {Pools total Gigahashes} = ⊅BTC per GH/s per day
nPPS can be any reasonable arbitrary value, .00003 can be used. It won't represent actual payouts just differences between pools.
If you ignore the losses due to strategic mining, this could be done as a percentage of network reward for the time period:
Pool blocks solved / Network blocks solved
Sort of a luck measurement. Good idea, Sgt Spike
Now, I don't suppose you'd mindparting with donating 10% of those BFL singles, would you?
I'll part with any of them if the price is right. Pool blocks solved / Network blocks solved
Sort of a luck measurement. Good idea, Sgt Spike
Now, I don't suppose you'd mind
Should have been more specific. Sorry.
Jump to: