Difficulty in Mining BTC Wires

Updated FAQs for newcomers

TL:DR: Don't bother mining if you want to get rich yo. You're way too late to the party.
Welcome to the exciting and often stressful world of bitcoin! You are wondering what looks like a once in a lifetime opportunity to get rich quick. Of course you guys probably heard about this "mining" process but what is this?
Simply put, a bitcoin mining machine that performs complicated calculations and when deemed correct by the network, receives a block which contains 25 bitcoins (XBT). This is how bitcoins are generated. So your brain instantly thinks, "Holy shit, how can I get on this gold rush?"
Before you proceed further, I would like to explain the concept of mining further. Bitcoin is limited 21m in circulation. It is coded to release a certain number of blocks at a certain time frame, ie: this year the network will release close to 500,000 bitcoins. What this means is that the more people (or specifically the amount of mining power) mine, the less each person gets. The network tries to keep to this time frame through the process of difficulty adjustments which makes the calculations harder and this happens every 2 weeks. So every 2 weeks, you get less bitcoins with the same hash rate (mining power) based on what the difficulty changes are. Recently, the changes have been pretty staggering, jumping 226% in 2 months. You can see the difficulty changes here.
Now, why are these changes so large?
A bit of a simple history. Bitcoin's algorithm runs on SHA-256. This algorithm can be solved using many hardware, from CPU to GPU and dedicated hardware (Application Specific Integrated Circuits). When bitcoin first started, mining on CPU was a trivial process, you can pretty much earn 50 XBT (the block size then) every few hours between Q1 and Q2 of 2010.
In late 2010, due to the difficulty increase that is reducing the effectiveness of CPU mining, people started to harness GPU mining. Only AMD GPU's architecture design are better optimized for bitcoin mining so this is what the community used. Immediate improvements of more than 10x was not uncommon.
In time of course, GPUs reached their limit and people started to build dedicated. In the same vein as the CPU to GPU transition, similar performance increase was common. These ASICs can only perform SHA-256 calculation so they can be highly optimized. Their performance mainly depends on the die size of the chips exactly like CPU chips.
In general, think of bitcoin mining's technological advancement no different to mining gold. Gold panning (CPUs) vs pickaxes (GPUs) vs machinery (ASICs) and we are still in the ASIC mining race.
ASIC mining started with ASICMiner and Avalon being first to the market, both producing 130nm and 110nm chips. The technology are antiquated in comparison to CPUs and GPUs which are now 22nm with 14nm slated for Q1 next year by Intel but they are cheap to manufacture and with performance gains similar to the CPU to GPU transition, they were highly successful and popular for early adopters. At that point in time since there were less competing manufacturers and the low batch runs of their products, miners became really rich due to the slow increase in difficulty.
The good days came to an end mid August with an unprecedented 35% increase in difficulty. This is due to existing manufacturers selling more hardware and many other players coming onto the market with better hardware (smaller die). Since die shrinking knowledge and manufacturing process are well known along with a large technological gap (110nm vs 22nm), you get an arms race. Current ASIC makers are closing in on our technological limit and until everyone catches up, the difficulty jumps will be high because it is just too easy to get a performance increase. Most newer products run at 28nm and most chips are not well optimized, so it will be around another 6 to 9 months before we see hit a hard plateau with 22nm or 14nm chips. The estimated time frame is because manufacturing chips at 22nm or 14nm is a more difficult and expensive task. In the meantime most manufacturers will probably settle at 28nm and we will reach a soft plateau in about 3 months.
Now, you might ask these questions and should have them answered and if you have not thought about them at all, then you probably should not touch bitcoin until you understand cause you are highly unprepared and probably lose lots of money.
No. If you have to ask, please do not touch bitcoin yet. You will spend more on electricity cost than mining any substantial bitcoin. Seriously. At all. A 7990 would produce a pitiful 0.02879 XBT (USD $14 @ $500/XBT exchange rate) for the next 30 days starting 23 Nov 2013 at 35% difficulty increase.
And if you think you can mine on your laptop either on a CPU or GPU, you are probably going to melt it before you even get 0.01 XBT.
Probably not because you probably forgot that GPUs and CPUs produce a ton of heat and noise. You can try but I see no point earning < $20 bucks per month.
No, because your machine will probably not mine as much as buying bitcoins. This situation is called the opportunity cost. While you can still make money if XBT rise in value, it is a fallacy.
IE: if you start mining on 1 Dec 2013, a KnC Jupiter running at 450Gh/sec (KnC lies as not all chips run at 550Gh/sec) will yield you a total revenue of 9.5189 XBT with a profit of 0.7859 XBT in profit by 30th Jan 2014 at a constant difficulty increase of 35%. The opportunity cost is: 8.5910 XBT @ USD $580/XBT with USD $5,000 which is the cost of a KnC Jupiter. This is the best you can earn and it's a bloody optimistic assumption because:
The only circumstances where you will earn money is when XBT exchange rates is so high that it makes the opportunity cost pales in comparison. Unfortunately this is not the case. If XBT stabilized at 900/XBT today (20 Nov 2013) then we might have a good case.
The risk is just generally not worth it. Unless you have at least a hundred thousand and can make a contract with a manufacturer for a lower cost, do not bother. Just wait until the arms race is over then you can start mining.
Okay, go buy an AsicMiner USB Block Erupter. They are cheap and pretty fun to have.
Sure, just read the answer below on who NOT to go for. You are doing bitcoin a service by securing the network and you have our (the users') gratitude.
You can check out the manufacturers and their products below along with a calculator here.
If you still insist on buying, do not to go for BFL. Their track record is horrid and borderline scammish. KnC fucked up a lot with defective boards and chips. Personally, I think CoinTerra is the best choice.
Alternatively, you can go on the secondary market to buy a delivered product. You can get a better deal there if you know how to do your "return on investment (ROI)" calculation. Personally, I will go for a 45%-50% difficulty increase for the next 3 months for my calculations and a 2% pool fee.
However, most products on ebay are sold at a cost much higher than it should. bitcointalk.org is a cheaper place because everyone knows what are the true value is so you will find less options. If you are unclear or need assistance, please post a question.
I actually do not use any of the pools recommended to the left because I think they lack features.
My favourite is Bitminter (Variable fees based on features used; max 2%). It has all advanced features for a pool, very responsive and helpful owner on IRC. Variable fees is good for those who do not need a large feature set, even with all features turned on, it is still cheap.
Eligius (0% fees) has high value for money but lacks features. It has anonymous mining which might be attractive to certain subset of people but not for others. Many other community member and I disagree highly with the opinions of the owner on the direction of bitcoin. I do use his pool for now but I do so only because I share my miners with a few partners and anonymous mining allows us to monitor the machines without using an account. Bitminter uses only OpenID which is problematic for me.
BTC Guild (3% fees) is another big pool and is fully featured and does charge a premium for their fees. That said, they are the most stable of the lot. I do use them but do so only because my hoster uses them for monitoring. I try not to use them because a pool with a very large hash rate (they are the largest) presents a large vulnerability to bitcoin's network if compromised.
All of them pay out transaction fees.
submitted by Coz131 to BitcoinMining [link] [comments]

Common misconceptions of forking blockchains

TL;DR There are some misconceptions about forking out there. Here are some information about soft and hard forks and what they mean in the context of the dao attack.
The past two days I have read a lot of reddit posts, comments, blog posts and media article regarding whether one should (soft and/or hard) fork or not as a reaction to the dao attack.
I have noticed that there are many common misconceptions on what a soft/hard fork is, what these forks mean and how the process of forking looks like.
In my opinion it is essential to clear up these misunderstandings in order to give every community member the best possibility to form a profound opinion and have a fruitful and action-oriented discourse.
Definitions of soft and hard forks
I assume that this post is most important for community members without technical background therefore I keep the following defintions simple:
Soft fork: A soft fork is a change in the protocol (set of rules) in a way that some previous valid transactions/blocks are no longer valid. Since everyone following the old rule set will recognise the new blocks as valid, a softfork is backward-compatible.
Hard fork: A hard fork is a change in the protocol (set of rules) in a way that some previous invalid transactions/blocks are valid after the change.
So how could the forks regarding the dao attack look like?
Soft fork: The doa attacker drained part of theDAO's fund in a child dao which is a new theDAO contract. Each dao contract has the same hash. A possible soft fork could disallow value transaction from contracts with these specific hash. The ether would be locked and can't be transfered anylonger.
Hard fork: In a hard fork all rule changes are possible. A hard fork would be required for getting the ether drained in the child dao back in the mother dao contract. This could be done be reverting all illegimate splits. Then one could deploy this contract and set the rule to forward every interaction of the mother dao to this new DAO which only functionality is to withdraw.
Other common misconceptions:
A hard fork means there will be a rollback and the Ether I just bought will vanish.
As I said before, in a hard fork all rule changes are possible and in principle even rollbacks. Nevertheless a rollback -a reversion of all transaction to a specific block- is absolutely unlikely to get community consensus and surely not indispensable in hard forks. Hard forks which might be proposed will only revert specific transaction if any (e.g.illegimate splits).
In a fork situation the miners decide which chain will "win".
One could say that in order to have a successful and reliable soft fork (eg. the dao attack soft fork) a coordination of miner would be helpful but in general soft and hard forking is much more community consensus.
Ethereum is much more flexible in adjusting the difficulty than Bitcoin. In Bitcoin it would result in a disaster if a large portion of miners decide to go along with the other chain than the rest of the community since the block time would increase enormously on the commonly used chain. Therefore forks in Bitcoin are prevoted by the miners - and yes in Bitcoin one can argue it is miner's choice. In Ethereum that's not the case.
In a fork scenario a client version with the fork and one without it is released. Then each community member decides which branch to go along simply by using one chain eg sending transaction, trading coins and running contracts/dapps with or without the forked client version. Exchanges most likely follow the demand of the community members. Under the assumption that a miner is a buisness man, miners therefore choose the chain to mine on which they earn most.
Miners choose the chain n where X(n)=(Hashrate/TotalHashrate(n))*CoinValue(n) is maximum. As you can see miners decision depends on the CoinValue of each chain which depends highly on the community decision.
Forking is an evil thing and should be avoided under every circumstances.
In my opinion forking the protocol of a blockchain has 3 purposes:
1) Keep the consensus system alive.
Forks are needed for bug fixing or are used to disincentivize attacks game theoretically.
2) Improve the consensus system.
Forks provide the possibility to add functionality or improve certain aspects of the consensus system.
3) Set and redefine the moral framework the consensus system runs in.
In my opinion forking is the purest form of social consensus. People come together in order to discuss, negatiate and finally form an agreement or build a new rule set. This can be done for fixing an previous agreement or for improving the capability of the previous agreement. Furthermore it can redefine or extends the morality of the previous agreement. Ethereum is a consensus system where 'code is law'. It is a system which is build for human interaction (at least it is a tool for humans) but, nevertheless, it lacks of empathy. IMHO it is the duty of its community to take care of the system and check once in a while if the moral framework has to be adjusted.
So another important purpose of forks is to adjust the moral framework the consenus system runs in.
I dont think forking is evil. It is an indispensable tool.
To illustrate the idea of this moral framework, I posted a thought experiment here. Check it out and discuss about how the moral framework of Ethereum should look like.
submitted by PhiStr90 to ethereum [link] [comments]

Bitcoin Mining Difficulty Drops A Historical 16% - Bitcoin Hashrate Down 45% - Why It's A Good Thing How to start Bitcoin mining for beginners (SUPER EASY ... How to do Bitcoin Mining tutorial: Bitcoin mining with CGMiner - YouTube BITCOIN MINING DIFFICULTY EXPLAINED IN 10 MINUTES!

What Happens When Mining Difficulty Rises? On August 27, before the price took off, the difficulty of the network was: 7.5 TH, and it’s hashrate: 500 GH/s (approximately 16000 video cards). This price increase caused the hashrate to increase to 1.12 TH/s (1120 GH/s) (about 35000 video cards), and the difficulty flew up to 16,728 TH. As a result, the rig’s profit, which was mining the coin ... Mining pools also have a pool-specific share difficulty setting a lower limit for shares. #2 Relationship with hash rate. One of the critical metrics in judging the health of a proof-of-work network is hash rate. Simply put, hashrate shows you how powerful the miners are within the network. Higher the bitcoin network hashrate, higher it’s overall security and speed. However, these networks ... The mining software constructs a block using the template (described below) and creates a block header. It then sends the 80-byte block header to its mining hardware (an ASIC) along with a target threshold (difficulty setting). The mining hardware iterates through every possible value for the block header nonce and generates the corresponding hash. The Bitcoin difficulty chart provides the current Bitcoin difficulty (BTC diff) target as well as a historical data graph visualizing Bitcoin mining difficulty chart values with BTC difficulty adjustments (both increases and decreases) defaulted to today with timeline options of 1 day, 1 week, 1 month, 3 months, 6 months, 1 year, 3 years, and all time Last but not the least; it’s possible that the mining pool itself uses a ‘share difficulty setting to mine a particular cryptocurrency; Mining Difficulty, in the world of bitcoin, automatically adjusts after every 2,016 blocks on the network. Based on how many people were mining, their combined hashpower, and the time it took to find 2,016 blocks, the difficulty will either increase or ...

[index] [14269] [29988] [1674] [44597] [30081] [43469] [43896] [49991] [1984] [5014]

Bitcoin Mining Difficulty Drops A Historical 16% - Bitcoin Hashrate Down 45% - Why It's A Good Thing

Bitcoin mining difficulty drops a historical 16%, bitcoin hashrate down 45%. Bitcoin mining difficulty explained in this video. Start mining bitcoins in the warrior mining btc mining farm: https ... In this video I show you how to start mining Bitcoins with CGMiner and an account at your favorite miningpool. Get CGMiner at: https://bitcointalk.org/index.... 🍓 Best Bitcoin Mining Software That Work in 2020 🍓 - Duration: 5 ... Nexalt Solo Miner Setup - Duration: 5:20. Muhammad Nouman 1,918 views. 5:20. How to make a Concrete Counter Top in 1 hour ... BITCOIN MINING DIFFICULTY EXPLAINED IN 10 MINUTES! GenicsTheCrypto. Loading... Unsubscribe from GenicsTheCrypto? Cancel Unsubscribe. Working... Subscribe Subscribed Unsubscribe 5.02K. Loading ... In this video, I attempt to describe how crypto mining difficulty works and how it affects profitability. I also crunch some numbers to show alternative methods for determining profits based on ...

#