Does a high pool difficulty lower anyone's profits? - page 4.

FlungSpun

sr. member

Activity: 414

Merit: 251

Quick crib from Wikipedia cos that's always right.

imagine that with the dominoes H2 of middlecoin fame decides he wants to minimise his incoming traffic to keep his pool slick and reactive
but he need to do this fairly.

He tells all miners that he is not interested in dominoes that add up to 7 (they are the most common to be found)

UNLESS A 7 HASH SOLVES THE BLOCK they wont get sent.

Everyone has to abide by the same rules and have exactly the same chance of finding the outlier dominoes and be paid for the proof they are doing work.
by removing the central column of probability there is less network traffic generated but the odds remain the same for all.

http://en.wikipedia.org/wiki/Probability_distribution

Agrippa

newbie

Activity: 14

Merit: 0

Well it's nice to see some consensus here after all the insanity of the middlecoin thread.

Like I was trying to say before, complaining that your miner would have gotten to get a share had the block lasted just a few more seconds is not only wrong because it is a complete misunderstanding of probability, but because by that logic you might as well complain about all the wasted hashes that would have yielded a share on the very next block.

mueslo

member

Activity: 94

Merit: 10

Quote from: Liquidfire on August 22, 2013, 03:11:37 PM

I now think the correct distribution for my simulation would be the Poisson distribution. Can any math geeks confirm or deny this for me?

If you had actually read the rebuttals, you'd know we already wrote that.

https://bitcointalksearch.org/topic/m.2979579

FlungSpun

sr. member

Activity: 414

Merit: 251

I'm pretty sure the last three statement have nailed this discussion.

turtle83

sr. member

Activity: 322

Merit: 250

Supersonic

Quote from: Eli0t on August 22, 2013, 03:43:37 PM

its possible to prove you worked on a solution for every block if the share diff is reasonable but when its high there can be long periods where you cant prove your working on a solution. when a pool doesnt credit you for the work you put in your wasting your time

In the long run it averages out completely. the hashers would have payout proportional to their hashrate. in the short run, with higher diffs low hashers have an increased variance...

The only reason for higher diff shares is efficiency, because lesser load on the networks, CPU, etc.

Quote

Does a high pool difficulty lower anyone's profits?

No. It only increases variance in the short term... which could go both ways.

Damnsammit

sr. member

Activity: 406

Merit: 250

Regardless of your hashrate, a higher difficulty reduces latency between the miners and the pool server.

The drawback is that it also increases the day-to-day variance. In the long-run, it doesn't matter, though.

FlungSpun

sr. member

Activity: 414

Merit: 251

I really don't want to get drawn into this argument because I understand the mechanics but realise I won't be able to explain it clearly enough.

From a confidence perspective it would be nice for smaller hashers to be able to prove work for every block (lower diff)
But
The ratio between the high and low rates of share submission will not change (ergo payouts) because the big hashers will just inundate the pool with PoW shares.
There is one provision with that statement, it takes a relatively long time to average out the larger variance the small miners suffer.

Eli0t

sr. member

Activity: 252

Merit: 250

its possible to prove you worked on a solution for every block if the share diff is reasonable but when its high there can be long periods where you cant prove your working on a solution. when a pool doesnt credit you for the work you put in your wasting your time

FlungSpun

sr. member

Activity: 414

Merit: 251

Quote from: Eli0t on August 22, 2013, 03:12:50 PM

it creates very low profits for low hashers who cant submit a share before the round ends on fast coins

A - Low hashers cannot expect to submit a share for every block on a fast coin, what makes anyone think that they should be submitting for every block?

B It makes no difference because they will still submit on blocks in relation to their hashrate and earn accordingly.

and what happens when a block is solved in 1 share and it happens to come from a low hasher?
should all the big hasher cry foul ?

GSnak

full member

Activity: 182

Merit: 100

Quote from: jdebunt on August 22, 2013, 03:12:04 PM

only thing that lowers profitability is idiots dumping their coins to the lower buy order....

I thought that too, until I noticed all my orders getting left behind. Now I pretty much dump on the top bid.

Hydroponica

full member

Activity: 182

Merit: 100

fml

Holy fuck, why don't you stop discussing it, and lower share difficulty for a day, and see what happens.

Eli0t

sr. member

Activity: 252

Merit: 250

it creates very low profits for low hashers who cant submit a share before the round ends on fast coins

your pool sells coins at the lowest price possible, this also reduces profits

jdebunt

legendary

Activity: 1596

Merit: 1010

only thing that lowers profitability is idiots dumping their coins to the lower buy order....

Liquidfire

newbie

Activity: 28

Merit: 0

I now think the correct distribution for my simulation would be the Poisson distribution. Can any math geeks confirm or deny this for me?

h2odysee

full member

Activity: 238

Merit: 119

There has been a lot of discussion about this in the middlecoin.com thread: https://bitcointalk.org/index.php?topic=259649.new#new

Let's move it here, to keep the topics separate.

Topic: Does a high pool difficulty lower anyone's profits? - page 4. (Read 4425 times)