To stop centralization, you need to cap the Difficulty, not the Blocksize !!!

[GermanGiant]

I think a big problem with bitcoin is the average joe cannot mine succesfully like before though.

That is a myth. The "average joe" has never been able to make a decent (if any) profit from mining. Even when CPU mining dominated, people generally mined at a loss.

How come CPU mining be unprofitable for those who mined then and sold later ?

Before bitcoins had value, miners spent money mining worthless bitcoins. Once bitcoins had value, the value of the mined bitcoins were still less than the cost to mine them.

Holding bitcoins has been very profitable for some people, including the miners that mined at a loss.

So, in simple terms, you can not just mine and sell. You need to mine, hold and sell.

[bitgolden]

It completely new things how to cap difficulty and how it could prevent some miners take 51% attack. If some one explain that would be more helpful. Thanks in advance.

[GermanGiant]

It completely new things how to cap difficulty and how it could prevent some miners take 51% attack. If some one explain that would be more helpful. Thanks in advance.

Please read this...

I see, there are people saying that increasing block size will increase centralization. This is true. But, the heart of bitcoin, i.e. mining, is already centralized. And this is due to pool mining. I hear, miners will decide with their hash power that which BIP will win. But, this is practically pool owners are chosing, not an average miner. If you want to stop this centralization, break the pools. This requires capping the difficulty, not the blocksize.

Capping the difficulty means that blocks would be found faster and faster as hashrate increases, accelerating bitcoin towards the final 21 million coin money supply point much faster than under the current plan. Miners would find blocks faster, but the halvings would come much faster so they'd be getting less and less when they did score a block.

I'm not sure how capping difficulty would change anything. If anything the only miners that could stay mining in such an environment would be the most efficient (of course this is tending to be true today, but would be even more so under these conditions). That would increase centralization of mining, not decrease it.

No. If difficulty is capped, block reward needs to be adjusted accordingly, so that all bitcoins do not get mined faster. Lowering block reward in capping difficulty will de-incentivize big hash power, wich is resource hungry. So, mining will be back to small machines again, if we can take down the difficulty and lower the block reward.

[adamstgBit]

if you cap Difficulty there is no cap on how fast blocks will be solved

you could get in a situation that blocks take ~1 second to solve.

orphen rate would go sky high.

no way.

[RussianRaibow]

if you cap Difficulty there is no cap on how fast blocks will be solved

you could get in a situation that blocks take ~1 second to solve.

orphen rate would go sky high.

no way.

But, there are many Alt coins where the average time to find a block is close to 1 second. Do they have high orphan rate ?

[RGBKey]

If we cap the difficulty, the network will no longer be scalable, we will have blocks every few minutes instead of 10.

[mallard]

But, there are many Alt coins where the average time to find a block is close to 1 second. Do they have high orphan rate ?

Probably.
Ethereum has a 10 second block time which causes many orphans, but I think they are called Uncles and it's still possible to get a block reward from them.

[GermanGiant]

In the light of Luke Jr's proposal to adjust difficulty on block halving, this thread has become relevant again.

[BlockyChain]

I see, there are people saying that increasing block size will increase centralization. This is true. But, the heart of bitcoin, i.e. mining, is already centralized. And this is due to pool mining. I hear, miners will decide with their hash power that which BIP will win. But, this is practically pool owners are chosing, not an average miner. If you want to stop this centralization, break the pools. This requires capping the difficulty, not the blocksize.

March 2, 2016 Update : Core devs are now planning a Hardfork to fix difficulty drop algorithm.

That isn't 100% true. If lets say you mine in AntPool, and they vote for a decision you don't want. What do you do? Simple. Switch! Lots of pools out there and surely one will suit your needs?

[franky1]

this is where luke is wrong.

difficulty is not based on total network hashrate of all miners hash combined. but on the time a miner can solve a block.

and it has been proven many times that MANY miners can make a block around 10 minutes. with a few varients of seconds between each
EG 6 of the top 20 pools REGULARLY make blocks around 10 minutes

so even if a few miners dropped off wont cause any issues at all. because miners can still make blocks in the same time period as always.

the debate of lukes should not be about the reward halving causing miners to drop out to cause prolonged time of blocksolving. which someone like him (a pool owner) should know all about. but instead its should be about (i think its his hidden reasoning) making it slightly easier to mine to give the little guys a chance to get some profits instead of the remaining big guys leaping forward with less competition and getting more rewards more often.

to me doing a hardfork purely for difficulty is stupid. because within a few weeks the difficulty would go up again. and thus its a waste of code and hard forking just for such a temporary measure of letting his little pool have a slight betterchance of profit for a couple weeks.

his proposal has nothing to do with blocktimes.
because if all the top 6 pools can make blocks in 10 minutes. give or take a few seconds.. then if 3 dropped out there are still 3 big pools making blocks in 10 minutes. thus less competition and thus the remaining 3 will solve more blocks between them rather then sharing between 6.