Topic: [1050 TH] BitMinter.com [1% PPLNS,Pays TxFees +MergedMining,Stratum,GBT,vardiff] - page 341. (Read 837101 times)
December 10, 2011, 09:12:27 AM
Is it possible to convert Bitcoins into Namecoins?
December 10, 2011, 05:48:47 AM
Already feared it was something like that. Nice of you to pay it anyway. This mine pooling business is becoming a real money printing machine for you, isnt it 
?
?Haha, yes. Bottom line so far is a loss of just over 200 bitcoins. Maybe it's time to implement that donation feature soon.
December 10, 2011, 05:40:01 AM
This mine pooling business is becoming a real money printing machine for you, isnt it ?
?
what do you mean ? 
December 10, 2011, 04:56:13 AM
Already feared it was something like that. Nice of you to pay it anyway. This mine pooling business is becoming a real money printing machine for you, isnt it ?
? December 09, 2011, 05:24:50 PM
Block 213 is orphaned (aka "invalid"). Another pool created the block a few seconds before us, and our block was orphaned upon creation. As if the bad luck lately wasn't enough.
When I was reading bitcoind and namecoind source code to decipher how to implement merged mining it seems I made a mistake which lead to a bug with detection of invalid blocks. These things can happen when you hurry to implement merged mining based on undocumented features. Still, doing so was my own decision, so I can't complain.
Please note that the server always tells bitcoind to create a block when it gets a proof of work that is good enough. The long blocks lately are just plain bad luck. This bug only affects the detection of invalid blocks. It cannot lead to loss of money for miners, only for the pool op.
There never was any 50 BTC from block 213, and I didn't actually intend to pay for invalid blocks. But the 50 BTC is already paid out, and it's my fault for not properly detecting failed blocks. I'll pay it out of my own pocket.
The bad luck has to end some time, right?
When I was reading bitcoind and namecoind source code to decipher how to implement merged mining it seems I made a mistake which lead to a bug with detection of invalid blocks. These things can happen when you hurry to implement merged mining based on undocumented features. Still, doing so was my own decision, so I can't complain.
Please note that the server always tells bitcoind to create a block when it gets a proof of work that is good enough. The long blocks lately are just plain bad luck. This bug only affects the detection of invalid blocks. It cannot lead to loss of money for miners, only for the pool op.
There never was any 50 BTC from block 213, and I didn't actually intend to pay for invalid blocks. But the 50 BTC is already paid out, and it's my fault for not properly detecting failed blocks. I'll pay it out of my own pocket.
The bad luck has to end some time, right?
December 09, 2011, 03:17:44 PM
Just noticed block 213 got confirmed and paid out instantly?
Looking into it..
December 09, 2011, 03:06:29 PM
Just noticed block 213 got confirmed and paid out instantly?
December 09, 2011, 12:13:04 PM
And what about the client? It could be coincidence of course, but it seems our luck turned sour around the time of the last update. Might be worth double checking what client the miners are using that have been finding blocks lately.
The regular (non-beta) version hasn't changed in a long time. And I haven't changed anything in the beta that should affect this. 4 days ago I made an NMC block and I'm always running the beta.
We're just having a long streak of bad luck, just like we had a long streak of good luck earlier. It would be preferable with more stable payouts of course, and I'm looking into what I can do to reduce the variance.
December 09, 2011, 07:16:35 AM
I'm pretty certain it's not a bug causing this. When the server sees a proof of work that beats the current difficulty it logs that fact before doing anything else. I'm not seeing any such messages without the next message stating that a block was created.
And what about the client? It could be coincidence of course, but it seems our luck turned sour around the time of the last update. Might be worth double checking what client the miners are using that have been finding blocks lately.
December 09, 2011, 06:55:16 AM
Congrats 
Now turn off this million share block limit
Now turn off this million share block limit December 09, 2011, 04:34:48 AM
There's an interview with me today (friday 2011.12.09) in the Norwegian business paper Dagens Næringsliv ("Business Today"), on page 24, about BitMinter.com and bitcoins in general. There's also a side article about Meze Grill. If you understand Norwegian you can buy a copy of today's paper at http://www.dn.no/avis/bestilling/dagens/eavis?execution=e1s1
December 07, 2011, 05:05:10 AM
A couple weeks back we had 70+ Ghps with 90 users. Right now we have 44 Ghps with 101 users. It's not so much people leaving, but rather old miners spreading our their hashrate among multiple pools to reduce variance.
I'm pretty certain it's not a bug causing this. When the server sees a proof of work that beats the current difficulty it logs that fact before doing anything else. I'm not seeing any such messages without the next message stating that a block was created.
Variance is really annoying in small pools. What you are talking about, conspirosphere, is definitely the way to go. I've already been making some plans along those lines. Let's see what I can come up with..
I'm pretty certain it's not a bug causing this. When the server sees a proof of work that beats the current difficulty it logs that fact before doing anything else. I'm not seeing any such messages without the next message stating that a block was created.
Variance is really annoying in small pools. What you are talking about, conspirosphere, is definitely the way to go. I've already been making some plans along those lines. Let's see what I can come up with..
December 07, 2011, 04:25:03 AM
Its rather clear whats going on. We are averaging over 3 million shares per block for quite some time now. Not sure what the statistical chances are over the last 11 blocks, but its definitely extremely unlucky. I suspect many worry there is something wrong with the pool or miner, or at least do like I started doing and spread their luck a bit by mining part time for bitminter and part time in a bigger pool.
December 07, 2011, 02:56:54 AM
What's going on ? Sad to see so many leave. F...ing shame 
December 05, 2011, 03:59:37 PM
Maybe there is a way for smaller pools: just contribute the whole pool's hashrate to a bigger pool, or -even better- find a way to cooperate among small pools in a bigger pool of pools. A dozen of pools with the speed of Bitminter would make a top-tier pool. Opinions?
December 04, 2011, 12:58:25 PM
At the moment we have good luck at eclipse. Pay out is fair over there. But no other pool i tried out so far comes close to BitMinters stale rates and homepage .
. December 04, 2011, 09:26:20 AM
If the variance in two pools is the same, then switching between them or staying in the same one makes no difference luck-wise. They both offer the same chances for future payouts. The exception is when you can know something about future payouts, as is the case with proportional pools.
Look, I'm sorry, since I use to stand with the underdogs and I like the pool, BUT I made about 9 BTC in november mining on Bitminter, while I am making 0.5 BTC/day on BTCGuild. I have an analogy about chances which may explain why smaller pools seem to underperform more often than not: If you bet on black or red playing roulette, you have a 50%/50% chance to win, but if you bet on a number, you have 1/72 chances, if I remember well, so you might gain more if you are lucky (only at roulette, since poolmining you get the same fraction of winning in bigger and smaller pools), but rarely *if ever*. Most likely, you would end losing most of the time at any given time.
Of course, in the loooooooooong term, smaller pools should perform like the biggest.
And in the short term chances of a small pool doing far better than the big pools are the same as doing worse. How much did you make in september and october? Past month was terrible, but youve already had the bad luck, switching to another pool now cant undo that, it can only consolidate your loss with almost no chance of making up for it.
Anyway, its your call of course.
December 04, 2011, 09:13:37 AM
If the variance in two pools is the same, then switching between them or staying in the same one makes no difference luck-wise. They both offer the same chances for future payouts. The exception is when you can know something about future payouts, as is the case with proportional pools.
Look, I'm sorry, since I use to stand with the underdogs and I like the pool, BUT I made about 9 BTC in november mining on Bitminter, while I am making 0.5 BTC/day on BTCGuild. I have an analogy about chances which may explain why smaller pools seem to underperform more often than not: If you bet on black or red playing roulette, you have a 50%/50% chance to win, but if you bet on a number, you have 1/72 chances, if I remember well, so you might gain more if you are lucky, but rarely *if ever*. Most likely, betting on numbers at roulette or mining in smaller pools you would end losing most of the times at any given time.
Of course, in the loooooooooong term, smaller pools should perform like the biggest.
December 04, 2011, 08:28:25 AM
If the variance in two pools is the same, then switching between them or staying in the same one makes no difference luck-wise. They both offer the same chances for future payouts. The exception is when you can know something about future payouts, as is the case with proportional pools.
A (theoretical) infinitely large pool has no variance and is almost completely predictable. If you switch to such a pool after a stroke of bad luck at a small pool like ours, you will never make up for the bad luck youve already had. If you stick with the unlucky smaller pool with higher variance, its only a matter of time before the bad luck averages out.
December 04, 2011, 08:05:48 AM
If the variance in two pools is the same, then switching between them or staying in the same one makes no difference luck-wise. They both offer the same chances for future payouts. The exception is when you can know something about future payouts, as is the case with proportional pools.
I can understand some users switching to pools with higher hashrate (and therefore lower variance), or PPS pools. Variance is a pain in a small pool when it's having bad luck. It is also very difficult to grow a pool up to a decent size. People leave because there aren't enough people. It's a catch-22.
Copying Wikipedia
A personal appeal from the operator of the BitMinter pool: If you like this pool then please keep mining here so it can grow.
In other news, a new beta release over in the miner thread: https://bitcointalksearch.org/topic/m.640162
I can understand some users switching to pools with higher hashrate (and therefore lower variance), or PPS pools. Variance is a pain in a small pool when it's having bad luck. It is also very difficult to grow a pool up to a decent size. People leave because there aren't enough people. It's a catch-22.
Copying Wikipedia
A personal appeal from the operator of the BitMinter pool: If you like this pool then please keep mining here so it can grow.In other news, a new beta release over in the miner thread: https://bitcointalksearch.org/topic/m.640162
Jump to: