Topic: why does one miner not mine all blocks? (Read 286 times)

This is how pool mining works, there is no duplication because the pool is giving a different block template (section) to every miner to look for the magic number, no miner is redoing the guessing of another miner.

I hope things are clear now.

By pool I was referring to the "swimming pool analogy full of balls with numbers" I am thinking rather than all mining rigs turning over the same ball (each mning rig works independently and doesn't know the when other rigs have found the wrong ball) so if each rig "claimed" a section of the swimming pool full of balls it would minimise duplication yes?

I believe he is talking about the pool level not the whole network, if all miners work on the same block template then finding a block will be a lot slower for that pool as a whole.

You cannot force 'a quicker solution'... You are forgetting the Bitcoin algo's built-in requirement that the average time between block is 10 minutes. Roughly every 2 weeks anything that changes that average time triggers a change in difficulty to correct for that change.

Something that 'solves blocks faster' be it more and better hardware or some previously unknown software hack just increases diff so is a null-game.

Assuming by "pool" you really mean a "pool" and not the whole network, then ya mining pools do have something like that, the pool won't send the same block template twice, any small detail you change in the block template opens a whole new door of non-overlapping "guesses".

I undertsand the answers here I think. Basically it is a lottery and the more tickets you purchase (hashrate) the more chance you have of winning but the single ticket holder can still win.
My question is do all miners start at a specific point in the possible solutions or randomly at any point in the pool? I'm thinking that why couldn't a hybrid "proof of work/luck" be incorprataed where a miner is given a certain area in the "pool of balls" ? wouldn't this enable a quicker solution if all miners were searching a smaller section of the pool rather than randomly searching the whole pool?

What's to be unsure about? The block header provided by -ck himself has all the details:

Yes, you noticed everything correctly, but the thing is that the miner changed its hashrate all the time and at the time of finding the block it really was 8.3 TH, but the asic itself had 86TH performance, n any case, this is according to the pool administrator, although he himself is not sure.

You really should correct that erroneous twitter report when reporting about it.
As the block information shows they ran 8.6T, not 86....
Their 7-day peak rate was 86T (rental?) for the user but the current speed was 8.31T

The speeds for the worker found it -- more impressive! Current worker speed was 3.65T, 5-min avg was 3.25T, 1hr was 3.32, 1 day was 3.41 and 7-day avg was 2.49T -- much more impressive of a find than saying they were running 86T at the time...

On January 24, an unknown solo bitcoin miner with an equipment capacity of 86 TH/s found block 720175 in the blockchain, as reported by the administrator of the CKPool pool, Con Kolivas, and this is the third such case in a month, some damn lucky pool for single miners. https://btc.com/btc/block/720175 

Almost

Your % of the network decides your % of the blocks found.

If, however, you have more than 50% of the network, you 'could' keep orphaning all other blocks found by only building off your own blocks.

Though it wouldn't make any sense to lower the fiat value of your mined BTC - since the fiat value of BTC would of course drop if only one entity was mining all the blocks in such a way.

I think the core misconception here was that only one person can have 'the fastest miner' and that one such miner is enough to dwarf the entirety of the rest of the mining community. To mine all blocks, in simple terms you'd need a large majority of the hashpower under your control. Like, not just one of 'the fastest miner', but tons of them. Also, all the 'not fastest' miners, aren't just produced once or twice, but thousands of times, so collectively they have a ton of power and a collectively extremely high likelihood to mine a block.

We have so much hashpower distributed more or less over the whole world, that even if you develop the latest, greatest, cutting-edge, 1nm-whatever ultra-efficient, ground-breaking, out-of-world mining chip... You would still need to probably invest billions into producing a heckton of them as well as tons of infrastructure and buildings to run them all under your control. ^{(to more or less guarantee 'mining all the blocks')}

Except that every block found is by a miner 150TH or less at the moment.

Every single block found.

Nothing unexpected about that.

Simply a case that some people have lots of miners ... but when they find a block, only one of their miners found it.

--

Currently a 126TH/s miner has a 1 in 9,615.5 chance of finding a block, if they mine for 1 full day - so it's 0.01% ... ... ...
If they mine for all 14 days during this difficulty it's 1 in 686.8 chance or 0.146% ... etc.

Really funny how people never understand the simple statistics of mining ... but go rabid when someone posts what people have been posting here on the forum for years ... that's twitter for you - full of twits  

Pity he can't do math ... though he has no computing degree so I guess that explains it.

Yes, indeed, a large percentage depends on luck here, for example, a fresh example according to the CKPool pool administrator on January 11, an unknown bitcoin miner with an equipment capacity of 126 TH/s found solo block 718 124 https://www.btc.com/btc/block/718124 , the probability of success with the current bitcoin hashrate was less than 0.0001%. 
https://twitter.com/ckpooldev/status/1480843321043865602

Simplest explanation:
It's random.

2 people: A rolls 1000 dice and B rolls 1 dice.
You win when you roll a 6
A can't stop B rolling a 6
Thus A gets (on average) 1000 times the reward of B ... but B still gets his reward.

Or ... https://kano.is/index.php?k=mining

I guess you'll have to read the whitepaper....
But no, the fastest miner won't solve all the blocks.

I'll try to give you a very, very, very simplified example:
You have a big swimming pool filled with numbered balls... There are 1.000.000.000.000 balls in the pool, each ball is numbered incrementally (from 0 to 999.999.999.999). Your task is to draw a ball with number 100 or less (that's the difficulty). If you find this ball, you get $1000 (the block reward) and any coins hidden inside the ball (the fees). If it takes to long for a "good" ball to be found, the consensus is that the number 100 might shift upwards or downwards (the difficulty retarget), this is checked every x "correct" balls that have been found. If it took to long for the network to "solve" the problem, they might change the difficulty to "all balls equal or less than 120 are good", if (on average) the time between 2 "good" balls being found was to short, the difficulty might change to "all balls equal or less than 80 are considered good".

By hand, you can check 1 ball per second...
Let's assume you have a machine that draws and checks 100 balls per second (the fastest asic), will this machine be guaranteed to "win" every time? If you are alone around this pool, the answer is: yes... But in reality, the network is open, everybody is allowed to mine, so you won't be the only one standing around the pool, trying their luck in claiming the >$1000 reward.

Now, let's assume everybody is able to buy a ball checking machine (ASIC's are sold commercially you know, everybody is allowed to buy one, unless their governement is quite backwards)... Around your pool filled with balls you'll have a couple dozen people picking and checking balls by hand (cpu mining), you'll have a couple people that invented a simpler machine themselfs and they're able to check 10 balls per second (gpu mining), and you have all kind of people with commercial ball picking machines (asic's) ranging from 50 balls per second to 100 balls per second.

Offcourse, after a while, it becomes clear that the people picking the balls by hand (cpu mining) and the ones picking balls using a homemade device (gpu mining) have a big disadvantage... Since picking and checking balls requires effort from them, they're pushed out of the ball picking (mining) environment in favor of the ASIC's... But it's not because you have the fastest ASIC, you'll win all the time... there are hundreds like you, and hundreds more with slightly slower machines... Your odds are about equivalent to your hashrate devided by the total hashrate (on average).

Now, this is a gross oversimplification... But to a point it is correct...
When you're mining, you're building a new block header using information from the previous block's header, a merkle tree and some other info... But most importantly, it also contains a nonce (which is basically a field that can be filled with random data).
Your ASIC fills the nonce with incremental data at a very, very, very fast pace (in the Th/s range), then it executes a sha256d function on the resulting block header and checks if the result is UNDER the current target (like with the balls, you're randomly picking them and checking if their number is equal to or smaller than 100).
The more hashpower you have, the bigger the odds of finding a header whose sha256d hash is under the current target... But the target is soooo small, that, on average, all ASIC's in the world will take ~10 minutes between finding such a header. Unless you own ALL asic's in the world, you won't be guaranteed to solve ALL blocks. It's like the swimming pool filled with balls: unless you're the only one picking balls from the pool and checking them, there is no guarantee that you'll win every time... Sure, the more ball checking machines you own, the bigger your odds become, but all in all your odds are about your hashrate divided by the total hashrate of all people standing around the swimming pool filled with numbered balls.

I hope this makes it clearer... If not: really, read the whitepaper... There are even simplified versions and youtube videos around that teach this stuff...

Guys, I have a one very noob question. .If the "fastest" miner is the successful one and gets the mining rewards, why does that same one not get all of the rewards, if it's the fastest? Is there a regulation that one should maybe skip a block or something?