Hey Inaba,
From Block Stats page
68 145590 2011-09-16 12:43:35 04:54:45 312085 -82.22% 1755425 6/120 1677 0.29812350
(+9.88%) 0.00270750
67 145567 2011-09-16 07:48:49 07:46:51 444131 -74.7% 1755425 29/120 2743 0.33792404
(+8.62%) 0.00306896
66 145534 2011-09-16 00:01:57 2 d 01:56:52 2815761 +60.4% 1755425 62/120 17235 0.32480914
(+5.78%) 0.00294986
my calculator says 0.96085668 for me
but from Account page
Unconfirmed Reward: 1.29878072
Am I misunderstanding something?
Update: I see it now. I'm getting the middle one twice!
Topic: [1200 TH] EMC: 0 Fee DGM. Anonymous PPS. US & EU servers. No Registration! - page 183. (Read 499629 times)
September 16, 2011, 10:51:28 AM
September 16, 2011, 09:33:04 AM
Ok, so I done broke it! Of course we solved two blocks super fast before I could fix it.
Short form explanation: I forgot that I was storing the pool score in a static variable in the getwork server... thus when we switched over to the new scoring system, I reset all the scores in the database but neglected to reset the pool score in the getwork static memory, so it was calculating based off the test scoring pool numbers which were substantially higher than they should have been.
Because of this, I had to revert the last two blocks to the geometric method, which works out for most everyone I suspect anyway, since it was such a short two rounds.
Everyone should have been paid properly now and your scores should be updated.
I have properly reset all the variables and now THIS round should be the first DGS score round (block 69) that is live.
When the block solves, it will still show 0 0 0 for everyone like these last two blocks did for a short time while I sanity check the numbers before making them live and allocating rewards. Depending on the round length of block 69, your pay could be substantially less than you expect - it all depends on how long the block takes. But rest assured that BTC that is not paid in this round will be paid as time goes on and you would eventually make it up as blocks get solved. I keep harping on this because I don't want anyone to be surprised when their payout on the first couple of blocks is less than they expect.
Just think of it as your payment being spread out a bit over several blocks instead of allocated on just one block at at time. As we solve more blocks and your payment gets spread out over several blocks, it will increase to the proportional amount and continue at that level as long as you continue mining. When you stop mining, it will slowly taper off, just like it slowly built up.
Short form explanation: I forgot that I was storing the pool score in a static variable in the getwork server... thus when we switched over to the new scoring system, I reset all the scores in the database but neglected to reset the pool score in the getwork static memory, so it was calculating based off the test scoring pool numbers which were substantially higher than they should have been.
Because of this, I had to revert the last two blocks to the geometric method, which works out for most everyone I suspect anyway, since it was such a short two rounds.
Everyone should have been paid properly now and your scores should be updated.
I have properly reset all the variables and now THIS round should be the first DGS score round (block 69) that is live.
When the block solves, it will still show 0 0 0 for everyone like these last two blocks did for a short time while I sanity check the numbers before making them live and allocating rewards. Depending on the round length of block 69, your pay could be substantially less than you expect - it all depends on how long the block takes. But rest assured that BTC that is not paid in this round will be paid as time goes on and you would eventually make it up as blocks get solved. I keep harping on this because I don't want anyone to be surprised when their payout on the first couple of blocks is less than they expect.
Just think of it as your payment being spread out a bit over several blocks instead of allocated on just one block at at time. As we solve more blocks and your payment gets spread out over several blocks, it will increase to the proportional amount and continue at that level as long as you continue mining. When you stop mining, it will slowly taper off, just like it slowly built up.
September 16, 2011, 09:20:38 AM
Yay!
September 16, 2011, 08:49:57 AM
0 shares are the coolest
=]
perhaps we could switch to triple trigonomic or quadruple quantum scoring?
inaba will straighten it all out. he always has.
it's a good thing he has no life other than bitcoin pools.
i know i don't.
/sarcasm mode OFF
'monkey
=]
perhaps we could switch to triple trigonomic or quadruple quantum scoring?
inaba will straighten it all out. he always has.
it's a good thing he has no life other than bitcoin pools.
i know i don't.
/sarcasm mode OFF
'monkey
September 16, 2011, 08:49:06 AM
Bitcoin first: Underpaid slave labour!
We need the union.
We need the union.
September 16, 2011, 06:56:23 AM
Anyway, pool is now running on DGS scoring. Your first payout will be somewhat less than what you expect.
Well, he didn't lie
Update: We got another block!
Doesn't look like we're gettin' anything for it but we're knockin' um out.
September 16, 2011, 06:51:02 AM
I agree with kripz
This is my block stats too:
67 145567 2011-09-16 07:48:49 07:46:51 444131 -74.7% 1755425 16/120 0 0 0
I am sure Inaba will fix it when he wakes up and has the time.
This is my block stats too:
67 145567 2011-09-16 07:48:49 07:46:51 444131 -74.7% 1755425 16/120 0 0 0
I am sure Inaba will fix it when he wakes up and has the time.
September 16, 2011, 04:05:07 AM
Quote
67 145567 2011-09-16 07:48:49 07:46:51 444131 -74.7% 1755425 3/120 0 0 0
Broken? 0 shares
September 15, 2011, 10:33:18 PM
So yeah, I had taken two of my rigs down within a couple hours of the block being solved. Bleh!
Anyway, pool is now running on DGS scoring. Your first payout will be somewhat less than what you expect. I will try to get the estimated payout to be more accurate either tonight or tomorrow. Right now it's running against the old calculations which will be much higher than you'll get for your first payout.
Anyway, pool is now running on DGS scoring. Your first payout will be somewhat less than what you expect. I will try to get the estimated payout to be more accurate either tonight or tomorrow. Right now it's running against the old calculations which will be much higher than you'll get for your first payout.
September 15, 2011, 08:23:07 PM
oops
September 15, 2011, 06:30:33 PM
One of you guys POKE IT ALREADY!!
No, don't poke it just yet. My rigs are down.. September 15, 2011, 06:06:19 PM
Okay, this block is two days old.
One of you guys POKE IT ALREADY!!
One of you guys POKE IT ALREADY!!
September 14, 2011, 01:36:22 PM
Did you guys see that somebody with 500 GH/s hashing power hopped in on EMC about 5 days ago? It still shows up in the "7 day historical hashrate graph". It looks like it was ~250 GH/s, but that day in the 24-hour view it was about 450 GH/s.
That was when Inaba stopped playing games and mined a little.
Edit: just for clarification
September 14, 2011, 01:26:14 PM
Really wish they'd get rid of this craptacular forum software 
The whole forum went down because of an attack The first and second block after the change will see a lower than average payout for everyone as the scores slowly build up, as time goes on and blocks get solved, your payout will increase to what you expect (+ whatever we get from hoppers).
Looking forward to the DGS. I'm less skeptical than before although I haven't studied it yet.Did you guys see that somebody with 500 GH/s hashing power hopped in on EMC about 5 days ago? It still shows up in the "7 day historical hashrate graph". It looks like it was ~250 GH/s, but that day in the 24-hour view it was about 450 GH/s.
September 13, 2011, 08:09:10 PM
So, numbers look good and there's even a small payout on invalid blocks (not that we've had many).
So, barring any unforseen circumstances, I will switch the payout system to the Double Geometric System after this block solves.
If anyone has any questions, comments or reasons not to, please let me know now!
So, barring any unforseen circumstances, I will switch the payout system to the Double Geometric System after this block solves.
If anyone has any questions, comments or reasons not to, please let me know now!
September 12, 2011, 06:43:01 PM
No, not yet. There are some discrepancies with the calculations I am still working out.
September 12, 2011, 08:02:23 AM
We found a couple of Blocks last night...so is the Double G scoing live now?
September 11, 2011, 11:30:53 AM
I want to be up front and clear about it, though:
The first and second block after the change will see a lower than average payout for everyone as the scores slowly build up, as time goes on and blocks get solved, your payout will increase to what you expect (+ whatever we get from hoppers). If you ever stop mining, your payout will slowly decrease (even if you aren't mining in the block) which will effectively even out what you did not make at the start of the new system.
Additionally, as an incentive to remain in the pool, on long blocks (those that exceed the expected shares), your payout will be slightly more than you would normally expect and on short blocks (those that are less than the expected shares) your payout will be slightly less than you would expect.
Looking at the numbers, it will take at least 5 or 6 blocks for the pool to normalize into a more or less steady state, so payouts will be "not quite right" in comparison to what you expect from proportional, or even the current geometric scoring.
I'd like to clarify that this does not affect your expected eventual income per share. In the old system, if you submit a share now you will get for it at round's end, on average, (50 BTC) / (Difficulty), or pB as I like to call it. In the new system, you'll still get on average pB per share (even if this share is submitted now in the startup phase) but it will take more time to clear, so you could for example get 0.5pB in the next block, 0.25pB in the one after, then 0.125pB and so on (these are made up numbers). This, again, is done to reduce your variance - the eventual total will be closer to the expected value than it was in the old system. The first and second block after the change will see a lower than average payout for everyone as the scores slowly build up, as time goes on and blocks get solved, your payout will increase to what you expect (+ whatever we get from hoppers). If you ever stop mining, your payout will slowly decrease (even if you aren't mining in the block) which will effectively even out what you did not make at the start of the new system.
Additionally, as an incentive to remain in the pool, on long blocks (those that exceed the expected shares), your payout will be slightly more than you would normally expect and on short blocks (those that are less than the expected shares) your payout will be slightly less than you would expect.
Looking at the numbers, it will take at least 5 or 6 blocks for the pool to normalize into a more or less steady state, so payouts will be "not quite right" in comparison to what you expect from proportional, or even the current geometric scoring.
September 11, 2011, 10:57:47 AM
With the difficulty about to go lower in the next couple days, hopefully those half dozen blocks won't take long.
September 11, 2011, 10:39:51 AM
Really wish they'd get rid of this craptacular forum software
Solved a few blocks while we were away!
I have been vetting the numbers for DGS, and I think they are looking proper. I just ran the last 4 blocks by Meni to make sure everything is calculating properly, so we shall see. If so, I'll probably change it over after this current block.
I want to be up front and clear about it, though:
The first and second block after the change will see a lower than average payout for everyone as the scores slowly build up, as time goes on and blocks get solved, your payout will increase to what you expect (+ whatever we get from hoppers). If you ever stop mining, your payout will slowly decrease (even if you aren't mining in the block) which will effectively even out what you did not make at the start of the new system.
Additionally, as an incentive to remain in the pool, on long blocks (those that exceed the expected shares), your payout will be slightly more than you would normally expect and on short blocks (those that are less than the expected shares) your payout will be slightly less than you would expect.
Looking at the numbers, it will take at least 5 or 6 blocks for the pool to normalize into a more or less steady state, so payouts will be "not quite right" in comparison to what you expect from proportional, or even the current geometric scoring.
Solved a few blocks while we were away!
I have been vetting the numbers for DGS, and I think they are looking proper. I just ran the last 4 blocks by Meni to make sure everything is calculating properly, so we shall see. If so, I'll probably change it over after this current block.
I want to be up front and clear about it, though:
The first and second block after the change will see a lower than average payout for everyone as the scores slowly build up, as time goes on and blocks get solved, your payout will increase to what you expect (+ whatever we get from hoppers). If you ever stop mining, your payout will slowly decrease (even if you aren't mining in the block) which will effectively even out what you did not make at the start of the new system.
Additionally, as an incentive to remain in the pool, on long blocks (those that exceed the expected shares), your payout will be slightly more than you would normally expect and on short blocks (those that are less than the expected shares) your payout will be slightly less than you would expect.
Looking at the numbers, it will take at least 5 or 6 blocks for the pool to normalize into a more or less steady state, so payouts will be "not quite right" in comparison to what you expect from proportional, or even the current geometric scoring.
Jump to: