[4+ EH] Slush Pool (slushpool.com); Overt AsicBoost; World First Mining Pool

[Epoch]

Almost all short blocks my payout is 20% less.  I am not quite sure why, am I doing anything wrong or is this something to do with pool hoppers?

I've noticed the same thing. On short rounds (say, less than 5 or 10 minutes) my payout is consistently 10%-20% less than my average. I understand that the variance is expected to be high on short rounds, but then I would expect half of the short rounds to payout MORE than my average and the other half to pay out LESS.

But what I have been consistently seeing is that on short rounds, my payout is invariably less than average. Never do I get more than my average (or even my average). I'm not sure what the issue is. It might be pool hoppers (Slush isn't completely immune to hopping) ... I'm not sure what else it could be.

[organofcorti]

Almost all short blocks my payout is 20% less.  I am not quite sure why, am I doing anything wrong or is this something to do with pool hoppers?

I've noticed the same thing. On short rounds (say, less than 5 or 10 minutes) my payout is consistently 10%-20% less than my average. I understand that the variance is expected to be high on short rounds, but then I would expect half of the short rounds to payout MORE than my average and the other half to pay out LESS.

But what I have been consistently seeing is that on short rounds, my payout is invariably less than average. Never do I get more than my average (or even my average). I'm not sure what the issue is. It might be pool hoppers (Slush isn't completely immune to hopping) ... I'm not sure what else it could be.

Unlikely to be hoppers. If this was a standard prop pool, hoppers would have to double the pool hashrate in order for full time miners to lose 20%.

Since hopping is much less lucrative here at Slush's pool, the hashrate increase for the first say 10-15% of the round would have to be higher. I don't see the pool hashrate even doubling for the first 10-15% of the round, so I'd say the decline in pay you're seeing can't be attributed (at least not most of it) to pool hopping.

[simc]

Could the dip at the beginning of the round be due to Slush's long polling optimisation, are slower miners loosing out to quicker ones?

[slush]

The last 3 blocks (10416, 10417, 10418) have been marked as invalid ... is something wrong? 

Cross-check for block validity on blockexplorer failed for some reason (I think BE was down for a moment). I fixed it manually right now.

[slush]

Littleshop, Epoch - your payout in short rounds is lower, because there's more hashpower on the beginning of the round (yes, thanks to pool hopers). There's around 10-20% of increase in pool rate, which explains your math (because your own hashrate is smaller portion of total hashrate).

[organofcorti]

Littleshop, Epoch - your payout in short rounds is lower, because there's more hashpower on the beginning of the round (yes, thanks to pool hopers). There's around 10-20% of increase in pool rate, which explains your math (because your own hashrate is smaller portion of total hashrate).

I think a 20% increase in hashrate causing a 20% decrease in fulltime miner earnings is unlikely. I'm willing to accept it if you can show me a proof though.

[Clipse]

slow miners allways get less rewards on short rounds with high hashrate pools, nothing strange about it.

Higher hashrate users would submit most of the shares in short rounds claiming most of the rewards.

[organofcorti]

slow miners allways get less rewards on short rounds with high hashrate pools, nothing strange about it.

Higher hashrate users would submit most of the shares in short rounds claiming most of the rewards.

Slow miners shouldn't get less reward on short rounds. If you're constantly hashing, then the likelihood of submitting a share in the first ten seconds of a round should be the same as at any other ten second interval. So I'd expect an increase in variance, but no change in expected value of a share due to the length of a round and miner hashrate.

[slush]

So I'd expect an increase in variance, but no change in expected value of a share due to the length of a round and miner hashrate.

...and did you asked those users how large dataset they're comparing?

Those +20% in hashrate was my tip, pool knows only 30min average hashrate so I don't know exactly what's the boost on the beginning of the round.

[organofcorti]

So I'd expect an increase in variance, but no change in expected value of a share due to the length of a round and miner hashrate.

...and did you asked those users how large dataset they're comparing?

No, and it did occur to me that it might just be variance they're experiencing. However,  Littleshop, Epoch were asking if pool hopping could cause a consistent loss of income and I pointed out that a 20% loss of income would require a massive hashrate boost until whatever the current correct hop point is.

Those +20% in hashrate was my tip, pool knows only 30min average hashrate so I don't know exactly what's the boost on the beginning of the round.

But you could work it out - anyone can. Your site provides shares submitted since start of round and time since start of round, so you can easily calculate the equivalent hashrate. Time how long the first say 130 000 shares (ish? Not at my abacus atm) of the round take to submit.

[simc]

So I'd expect an increase in variance, but no change in expected value of a share due to the length of a round and miner hashrate.

...and did you asked those users how large dataset they're comparing?

No, and it did occur to me that it might just be variance they're experiencing. However,  Littleshop, Epoch were asking if pool hopping could cause a consistent loss of income and I pointed out that a 20% loss of income would require a massive hashrate boost until whatever the current correct hop point is.

Those +20% in hashrate was my tip, pool knows only 30min average hashrate so I don't know exactly what's the boost on the beginning of the round.

But you could work it out - anyone can. Your site provides shares submitted since start of round and time since start of round, so you can easily calculate the equivalent hashrate. Time how long the first say 130 000 shares (ish? Not at my abacus atm) of the round take to submit.

OK I've had my abacus out and run it over my stats for yesterday.  I only have 24 hours worth of data so the sample size is a bit small really but there does seem to be a trend towards short rounds having a higher average hash rate.  Some short rounds I have done better than average others worse, but even factoring in the higher hash rate I'm a little down but at this point I'm only considering 6 samples of rounds less than 10 minutes so not exactly conclusive.  The other way to look at it is that the miners that are pool hopping are loosing out to me on the longer rounds as their shares decay, I don't think it's easily possible to calculate my submitted shares and thus payout per share for a round.

Results here
https://docs.google.com/spreadsheet/ccc?key=0Ahmqn5gYZuJNdDRVb2doZGtocHZEYVVlY21EYXFrV3c

[organofcorti]

OK I've had my abacus out and run it over my stats for yesterday.  I only have 24 hours worth of data so the sample size is a bit small really but there does seem to be a trend towards short rounds having a higher average hash rate.  Some short rounds I have done better than average others worse, but even factoring in the higher hash rate I'm a little down but at this point I'm only considering 6 samples of rounds less than 10 minutes so not exactly conclusive.  The other way to look at it is that the miners that are pool hopping are loosing out to me on the longer rounds as their shares decay, I don't think it's easily possible to calculate my submitted shares and thus payout per share for a round.

Results here
https://docs.google.com/spreadsheet/ccc?key=0Ahmqn5gYZuJNdDRVb2doZGtocHZEYVVlY21EYXFrV3c

I've analysed your data here. Comments are welcome, but keep them to the 'How to hop' thread, since this is getting a bit off topic.

[disclaimer201]

Sorry, to interrupt. I've generated 6 blocks now in 7 months and received 213.5 BTC @ slush's pool and one block somewhere else with say 10 btc payout. Should I have gone solo? I really get the feeling I should have. The pool is fine, but still it's a bit frustrating.

[digital]

Sorry, to interrupt. I've generated 6 blocks now in 7 months and received 213.5 BTC @ slush's pool and one block somewhere else with say 10 btc payout. Should I have gone solo? I really get the feeling I should have. The pool is fine, but still it's a bit frustrating.

Your better off going with the pool.  It will even out in the long run trust me.

I started out the same way, only on a smaller scale.  I found 3 blocks in my first three months while only receiving about 50 btc.  But since then I haven't found any blocks (im still at three) and i've been paid over 200 btc.

In reality you pretty much get paid the same either way, just with a pool the payouts come on a regular basis.  With solo mining it could be a year between blocks...

[dishwara]

Not only you, many i see saying same thing, including me.

When they first join pool, they find some blocks, but after that they never find any blocks.
It's really strange to me.
How can a new joined miner can able to find blocks when he is new, but not when he is old?
or some one tricking us that new ones found blocks by saying you found blocks in statistics?

I found 3 blocks joining pool & with in a month, that's around last year march, after that i so far found only 2 blocks & i stopped mining in oct.

[ensign_lee]

Is something wrong with the share counting?

I just noticed that slush says that my two miners haven't been doing any work at all for the past 14 hours, and yet my comp has been making all the noises and what not that it normally does when mining, and guiminer says that it has been producing hashes.

ensignlee.Daniel1 and ensignlee.Daniel2 if that would help troubleshoot for you, slush.

[Barlog]

guiminer says that it has been producing hashes.

ensignlee.Daniel1 and ensignlee.Daniel2 if that would help troubleshoot for you, slush.

Guiminer use opencl (poclbm) core? Try to change it phoenix or other.

[ensign_lee]

guiminer says that it has been producing hashes.

ensignlee.Daniel1 and ensignlee.Daniel2 if that would help troubleshoot for you, slush.

Guiminer use opencl (poclbm) core? Try to change it phoenix or other.

Why it suddenly change and no longer count my shares?

[BinoX]

There is something wrong with the default GUIminer engine (poclbm) for some reason.

Seems to have started in the last day or so, although I'm not sure what's so special about today that would cause it to change...

https://bitcointalk.org/index.php?topic=62590.0 has other people having problems with guiminer on other pools as well

[ensign_lee]

There is something wrong with the default GUIminer engine (poclbm) for some reason.

Seems to have started in the last day or so, although I'm not sure what's so special about today that would cause it to change...

https://bitcointalk.org/index.php?topic=62590.0 has other people having problems with guiminer on other pools as well

Oh, so it's guiminer's fault, not slush. What the hell?

And what the heck is my computer doing if it's not submitting shares anyhow? It's obviously still working...