Difference between revisions of "Talk:Why pooled mining"

From Bitcoin Wiki
Jump to: navigation, search
(This equation is plain wrong)
 
(6 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 +
== This equation is plain wrong ==
 +
 +
p + .9p + .9^2 p + .9^3 p ....
 +
 +
You cannot sum up sequential probabilities. You should count the opposite probability. See https://en.wikipedia.org/wiki/Probability#Mathematical_treatment
 +
 +
Note: 10% difficulty rise is quite unexpected, see: http://bitcoin.sipa.be/
 +
 
== Request for deletion ==
 
== Request for deletion ==
 
This whole page is about spreading FUD about solo mining and comes to conclusions like:
 
This whole page is about spreading FUD about solo mining and comes to conclusions like:
Line 5: Line 13:
 
--[[User:Giszmo|Giszmo]] 21:37, 21 June 2011 (GMT)
 
--[[User:Giszmo|Giszmo]] 21:37, 21 June 2011 (GMT)
 
* I am reverting the delete request because your arguments are complete nonsense. The math is probably correct, and the page was contributed by a well-known and trusted member of the community. If there are some errors in the math, perhaps correction is needed, but certainly not deletion. Consider especially that the conclusions it supports are also correct: solo mining ''is'' always inferior to pooled mining. --[[User:Luke-jr|Luke-jr]] 14:28, 22 June 2011 (GMT)
 
* I am reverting the delete request because your arguments are complete nonsense. The math is probably correct, and the page was contributed by a well-known and trusted member of the community. If there are some errors in the math, perhaps correction is needed, but certainly not deletion. Consider especially that the conclusions it supports are also correct: solo mining ''is'' always inferior to pooled mining. --[[User:Luke-jr|Luke-jr]] 14:28, 22 June 2011 (GMT)
 +
** ... says the owner of a pool. The article claims that with a 500Mh/s machine and a difficulty of 567358 I will have an average time to generate of 56 days. So far so good. But then it concludes that for the same setup there will not be a generation ever with 96% probability. Sorry but the 56 days figure is the 50% threshold already. If I have a 50% chance to hit in 56 days how can that probability possibly go down for "forever"? --[[User:Giszmo|Giszmo]] 22:46, 22 June 2011 (GMT)
 +
*** It seems you don't understand how mining works. At least every 14 days, the difficulty goes up. The fact that I happen to run a pool does not change the fact that solo mining is less profitable than mining in a pool. --[[User:Luke-jr|Luke-jr]] 00:56, 23 June 2011 (GMT)
 +
**** It seems you don't understand how mining works. The difficulty '''changes''' every 2016 blocks. If it happens after more than 14 days, the difficulty goes down. I see your POV biased as you run a pool if that was not clear in my prior post and I strongly oppose your statement that pool mining is more profitable than solo mining. In fact pushing work (so called long polling) is the only mechanism that may result in a 2% increased profit for the mining process but not necessarily for the miner. The pool may be subject to attacks (miners that take the profit but don't report solved blocks, DOS attacks, ...) and most pools take fees or at least withhold the transaction fees.
 +
**** Regardless of the details we throw back and forth here, the article is plain wrong.
 +
***** How can on the one hand for the given starting conditions the estimated time to find a block be every 56 days but on the other hand the probability to find it in the first days with constant conditions '''not''' be above 4%?
 +
***** How can the author get through with the claim that the network total will rise ad infinitum? (Yes, at the moment it rises much faster than 10%/14 days. It's rather 5%/day.)
 +
**** I have better things to do than this stupid edit war but I also hate to see people being tricked into pool mining as I see a [[Talk:Weaknesses#Pools|big threat]] for bitcoin coming from pools. I would like to see a well done estimate for "not getting a block in a year". The article as it stands is wrong. How should we go with it? Is there any template for "the content is strongly disputed. please help improving it"? --[[User:Giszmo|Giszmo]] 10:47, 23 June 2011 (GMT)
 +
 +
== Alternative content ==
 +
 +
I requested to delete the article as it was plain wrong and off by magnitudes. Now I took the time to do the maths myself. I would say the result is impressive and I don't mind debating certain assumptions but hope to find support to replace the old content. The claim that pool mining pays out more still is also wrong. I would only support the claim that via a pool it is highly unlikely to not get a cash out in a certain time.
 +
--[[User:Giszmo|Giszmo]] 14:10, 24 June 2011 (GMT)

Latest revision as of 08:43, 20 July 2012

This equation is plain wrong

p + .9p + .9^2 p + .9^3 p ....

You cannot sum up sequential probabilities. You should count the opposite probability. See https://en.wikipedia.org/wiki/Probability#Mathematical_treatment

Note: 10% difficulty rise is quite unexpected, see: http://bitcoin.sipa.be/

Request for deletion

This whole page is about spreading FUD about solo mining and comes to conclusions like:

And conversely, your probability of /never generating a block even after millions of years/ is about 96+ percent.

... given a 500Mh/s machine. I studied maths but sorry in this case I don't even bother to look into the formulas presented as the "experiment bitcoin" proofs it very far wrong. --Giszmo 21:37, 21 June 2011 (GMT)

  • I am reverting the delete request because your arguments are complete nonsense. The math is probably correct, and the page was contributed by a well-known and trusted member of the community. If there are some errors in the math, perhaps correction is needed, but certainly not deletion. Consider especially that the conclusions it supports are also correct: solo mining is always inferior to pooled mining. --Luke-jr 14:28, 22 June 2011 (GMT)
    • ... says the owner of a pool. The article claims that with a 500Mh/s machine and a difficulty of 567358 I will have an average time to generate of 56 days. So far so good. But then it concludes that for the same setup there will not be a generation ever with 96% probability. Sorry but the 56 days figure is the 50% threshold already. If I have a 50% chance to hit in 56 days how can that probability possibly go down for "forever"? --Giszmo 22:46, 22 June 2011 (GMT)
      • It seems you don't understand how mining works. At least every 14 days, the difficulty goes up. The fact that I happen to run a pool does not change the fact that solo mining is less profitable than mining in a pool. --Luke-jr 00:56, 23 June 2011 (GMT)
        • It seems you don't understand how mining works. The difficulty changes every 2016 blocks. If it happens after more than 14 days, the difficulty goes down. I see your POV biased as you run a pool if that was not clear in my prior post and I strongly oppose your statement that pool mining is more profitable than solo mining. In fact pushing work (so called long polling) is the only mechanism that may result in a 2% increased profit for the mining process but not necessarily for the miner. The pool may be subject to attacks (miners that take the profit but don't report solved blocks, DOS attacks, ...) and most pools take fees or at least withhold the transaction fees.
        • Regardless of the details we throw back and forth here, the article is plain wrong.
          • How can on the one hand for the given starting conditions the estimated time to find a block be every 56 days but on the other hand the probability to find it in the first days with constant conditions not be above 4%?
          • How can the author get through with the claim that the network total will rise ad infinitum? (Yes, at the moment it rises much faster than 10%/14 days. It's rather 5%/day.)
        • I have better things to do than this stupid edit war but I also hate to see people being tricked into pool mining as I see a big threat for bitcoin coming from pools. I would like to see a well done estimate for "not getting a block in a year". The article as it stands is wrong. How should we go with it? Is there any template for "the content is strongly disputed. please help improving it"? --Giszmo 10:47, 23 June 2011 (GMT)

Alternative content

I requested to delete the article as it was plain wrong and off by magnitudes. Now I took the time to do the maths myself. I would say the result is impressive and I don't mind debating certain assumptions but hope to find support to replace the old content. The claim that pool mining pays out more still is also wrong. I would only support the claim that via a pool it is highly unlikely to not get a cash out in a certain time. --Giszmo 14:10, 24 June 2011 (GMT)