Pages:
Author

Topic: [1050 TH] BitMinter.com [1% PPLNS,Pays TxFees +MergedMining,Stratum,GBT,vardiff] - page 22. (Read 837167 times)

newbie
Activity: 7
Merit: 0
How do you get luck 47.47% ?

Over 9 days even luck would give us 14.6 blocks. During these specific 9 days we ended up finding 12 (+1 stale) blocks. That's bad luck, but calling it 47% seems a bit extreme.


What would perfectly average mining look like? In a perfectly average world we would be finding 1.x something blocks per day and the CDF would always be 50%, right? And the only thing changing that number would be a change in our pool Thps or a change in the mining difficulty.

CDF is definitely not linear. Seems you can get really close to the limits of the scale but never quite reach them. We can mine the snot out of blocks, seemingly forever, before finding one and be somewhere in the 99.x% CDF. It's close but never quite 100%. A pool (not this one:) could mine for a month before finding a block and the CDF would still only be 99.x% only a little closer to 100%.

And what about the other end of the scale? If you find the next block on the very first hash you try, would this equate to exactly 0% CDF? Thanks
legendary
Activity: 4326
Merit: 8950
'The right to privacy matters'
I am adding a pair of s-3's  to the doc h team.  seems to be really nice gear so far.

Stock clock is freq 218  gives me 440 gh

I under clocked to freq 212  gives me 426 gh  for 1 machine   watts used with an evga 1300 psu are  640 for the 2 machines with a total gh of around 840

I also will be adding a pair of S3's next week.


I just switched my psu from the evga to the plat seasonic   power dropped to 620-630  watts.

 my meter does jumps of 10 watts.   so 620 watts for 846 hd = .736 watts a gh  630 watts for 846 gh = .744 watts a gh

the two are staying at freq 212.5 I seem to get the best results at 212.5  not the stock  218.625 or the 225 overclock


diff is moving up http://bitcoincharts.com/  has 17336 to 18597  that is 7.27 percent


https://bitcoinwisdom.com/bitcoin/difficulty  has 17336 to 19208  that is 10.80 percent
member
Activity: 296
Merit: 10
I am adding a pair of s-3's  to the doc h team.  seems to be really nice gear so far.

Stock clock is freq 218  gives me 440 gh

I under clocked to freq 212  gives me 426 gh  for 1 machine   watts used with an evga 1300 psu are  640 for the 2 machines with a total gh of around 840

I also will be adding a pair of S3's next week.
legendary
Activity: 4326
Merit: 8950
'The right to privacy matters'
I am adding a pair of s-3's  to the doc h team.  seems to be really nice gear so far.

Stock clock is freq 218  gives me 440 gh

I under clocked to freq 212  gives me 426 gh  for 1 machine   watts used with an evga 1300 psu are  640 for the 2 machines with a total gh of around 840
legendary
Activity: 2478
Merit: 1020
Be A Digital Miner
Good to know.  I've never gotten into the analytics of mining, so I was just going off what was listed on the site.
With that said, it still can be inferred that luck decreases as difficulty goes up.  And I think we have seen that in the recent weeks.  More shares are needed to keep the blocks coming in at a steady rate
"luck" does not change as difficulty rises.   Your expected outcome will be lower as difficulty rises if you hash rate does not increase in proportion with the rise in difficulty.
newbie
Activity: 20
Merit: 0
Good to know.  I've never gotten into the analytics of mining, so I was just going off what was listed on the site.
With that said, it still can be inferred that luck decreases as difficulty goes up.  And I think we have seen that in the recent weeks.  More shares are needed to keep the blocks coming in at a steady rate
hero member
Activity: 938
Merit: 502
Maintenance went fine. Looks like everyone is mining again. Just give the live hashrate stats a moment to catch up. Smiley

4-leaf clovers work  Grin

Yes, rabbit's foot is also supposed to work.
red underwear seems to work also  Grin

Personally, I wear my 4 leaf clover boxers every time Notre Dame plays or our CDF hits over 33.33%.  According to Anheuser-Busch, its only weird if it doesn't work.
legendary
Activity: 1750
Merit: 1007
This is an average of the luck percentages listed on the block timeline.
If you go back to May 15,the luck goes to 46.81 and average work per block goes to 9,886,480,396
Isn't lower better with regards to that number

I may be mistaken, but I'm fairly certain you can't simply take the average of CDFs to result in average luck.  If you want a number that is easier to compare with other pools (which report luck as earnings compared to expectation):  Average Difficulty / Shares Submitted.  That gives you the percentage expressed as a decimal (1.00 = 100%, 1.1 = 110%, 0.9 = 90%), and is how much was earned vs how much was expected in that period.  You should be including orphan+stale blocks in this figure.  Higher is better.
newbie
Activity: 20
Merit: 0
This is an average of the luck percentages listed on the block timeline.
If you go back to May 15,the luck goes to 46.81 and average work per block goes to 9,886,480,396
Isn't lower better with regards to that number
legendary
Activity: 2730
Merit: 1034
Needs more jiggawatts
How do you get luck 47.47% ?

Over 9 days even luck would give us 14.6 blocks. During these specific 9 days we ended up finding 12 (+1 stale) blocks. That's bad luck, but calling it 47% seems a bit extreme.
newbie
Activity: 20
Merit: 0
for those interested, I have done some mild number crunching based on the data from the block timeline on the bitminter site.

back to June 10th, give or take a day, here are the averages.  (this does not include the two orphaned blocks, or the stale one, as I did not know how to factor that in.)

average work per block found  |  average difficulty    |  average luck  |  average THps
13,122,789,965.68                 |    14,313,410,979    |         47.47%  |       1,329.96

I couldn't get time because excel is a pain.

EDIT: If anyone would like the file, I can get it to you somehow.
legendary
Activity: 2478
Merit: 1020
Be A Digital Miner
I see the pool hopping posts a lot.   Maybe someone should bring a roulette wheel to the next miner conference and only allow bets on red and black and let people see how much money can be made by waiting until three red/blacks come in a row and then betting the opposite.   I am surprised there are no books written on this strategy (I am sure wynn/adelson would be happy to fund the printing of them).
newbie
Activity: 46
Merit: 0
Maintenance went fine. Looks like everyone is mining again. Just give the live hashrate stats a moment to catch up. Smiley

4-leaf clovers work  Grin

Yes, rabbit's foot is also supposed to work.
red underwear seems to work also  Grin
legendary
Activity: 2730
Merit: 1034
Needs more jiggawatts
Maintenance went fine. Looks like everyone is mining again. Just give the live hashrate stats a moment to catch up. Smiley

4-leaf clovers work  Grin

Yes, rabbit's foot is also supposed to work.
legendary
Activity: 2730
Merit: 1034
Needs more jiggawatts
Reminder: Short maintenance in 10 minutes. Expected mining downtime 1 to 5 minutes.
member
Activity: 296
Merit: 10
Moving around trying to always be at the luckiest pool doesn't work. You can't move to a luckier pool when our luck is down, and back here when our luck is up.

Why?

Because you don't know how lucky a pool will be in the future. You don't know the future.

Crystal balls don't work. Tarot cards don't work. Voodoo doesn't work. Predicting today's luck based on yesterday's luck doesn't work. Nothing works. It's all superstition. The reality is that you can't tell the future.


4-leaf clovers work  Grin
legendary
Activity: 2730
Merit: 1034
Needs more jiggawatts
Moving around trying to always be at the luckiest pool doesn't work. You can't move to a luckier pool when our luck is down, and back here when our luck is up.

Why?

Because you don't know how lucky a pool will be in the future. You don't know the future.

Crystal balls don't work. Tarot cards don't work. Voodoo doesn't work. Predicting today's luck based on yesterday's luck doesn't work. Nothing works. It's all superstition. The reality is that you can't tell the future.
newbie
Activity: 28
Merit: 0

If I am still logging into Bitminter for mining, but can choose which pool my miner points at via the very nice Bitminter GUI ( so I can choose between Bitminter, Eligius, or GHASH.io for example ), then the good Doctor still gets paid.  Luck graphs can be compared, and you know some people will enjoy the ducking and diving bit.  Having it all on one website ( run by Bitminter ) would be a nice product for miners to buy into.


This would mean you are paying fees to both Bitminter and whatever pool your hash power is being directed to, which would be a bad idea.
full member
Activity: 120
Merit: 100
Yo Doc!  I just found this:

https://github.com/cdhowie/Bitcoin-mining-proxy

Allows you to have multiple blahs on a thingy.  By the looks of it you can have a bunch of miners on it, and then point it where you wish.   You have the worker on this server, and can point each worker at a different pool if need be.

Which would help with this luck issue.

IMO, it's time to drop the whole "luck" discussion.

We solved 5 BTClocks in less than 2 days.

Our luck looks just fine.

Not trying to be snarky, but nothing is preventing you from going to another pool if you are not happy with BitMinter.




This is precisely my point.  If I need to jump to another pool for a bit the good Doctor will lose his cut for a bit.  And I will lose my excellent GUI ( the Bitminter setup is very nice after all ).

If I am still logging into Bitminter for mining, but can choose which pool my miner points at via the very nice Bitminter GUI ( so I can choose between Bitminter, Eligius, or GHASH.io for example ), then the good Doctor still gets paid.  Luck graphs can be compared, and you know some people will enjoy the ducking and diving bit.  Having it all on one website ( run by Bitminter ) would be a nice product for miners to buy into.

Hence why I looked for some tech for the good Doctor ( to help him ).  The mining proxy link I put up.


But yes, I think I need to stop making this sales pitch now and let people have a think.


FYI:  I'll never leave Bitminter, but I shift some of my power to other pools if our luck is down.  There is always a minimum that stays with Bitminter though.  And I want to leave it all here really.  I have bought an S3 for a reason.
newbie
Activity: 35
Merit: 0
There is a way to avoid "stale" block?
Pages:
Jump to: