Antminer T19 May Not Affect Bitcoin Hash Rate but Keeps Bitmain Ahead
The Antminer T19 by Bitmain may not have a big impact on the Bitcoin network, and it comes out amid the firm’s internal and post-halving uncertainty. Earlier this week, Chinese mining-hardware juggernaut Bitmain unveiled its new product, an application-specific integrated circuit called Antminer T19. The Bitcoin (BTC) mining unit is the latest to join the new generation of ASICs — state-of-the-art devices designed to mitigate increased mining difficulty by maximizing the terahashes-per-second output. The Antminer T19 announcement comes amid the post-halving uncertainty and follows the company’s recent problems with its S17 units. So, can this new machine help Bitmain to reinforce its somewhat hobbled position in the mining sector? T19: The cheaper S19 According to the official announcement, the Antminer T19 features a mining speed of 84 TH/s and a power efficiency of 37.5 joules per TH. The chips used in the new device are the same as those equipped in the Antminer S19 and S19 Pro, though it uses the new APW12 version of the power supply system that allows the device to start up faster. Bitmain usually markets its Antminer T devices as the most cost-effective ones, while the S-series models are presented as the top of the line in terms of productivity for their respective generation, Johnson Xu — the head of research and analytics at Tokensight — explained to Cointelegraph. According to data from F2Pool, one of the largest Bitcoin mining pools, Antminer T19s can generate $3.97 of profit each day, while Antminer S19s and Antminer S19 Pros can earn $4.86 and $6.24, respectively, based on an average electricity cost of $0.05 per kilowatt-hour. Antminer T19s, which consume 3,150 watts, are being sold for $1,749 per unit. Antminer S19 machines, on the other hand, cost $1,785 and consume 3,250 watts. Antminer S19 Pro devices, the most efficient of three, are considerably more expensive and go for $2,407. The reason Bitmain is producing another model for the 19 series is due to what is known as "binning" chips, Marc Fresa — the founder of mining firmware company Asic.to — explained to Cointelegraph: “When chips are designed they are meant to achieve specific performance levels. Chips that fail to hit their target numbers, such as not achieving the power standards or their thermal output, are often ‘Binned.’ Instead of throwing these chips in the garbage bin, these chips are resold into another unit with a lower performance level. In the case of Bitmain S19 chips that don’t make the cutoff are then sold in the T19 for cheaper since they do not perform as well as the counterpart.” The rollout of a new model “has nothing to do with the fact that machines are not selling well,” Fresa went on to argue, citing the post-halving uncertainty: “The biggest reason machines probably are not selling as well as manufacturers would like is because we are on a bit of a tipping point; The halving just happened, the price can go anyway and the difficulty is continuing to drop.” Product diversification is a common strategy for mining hardware producers, given that customers tend to aim for different specifications, Kristy-Leigh Minehan, a consultant and the former chief technology officer of Genesis Mining, told Cointelegraph: “ASICs don’t really allow for one model as consumers expect a certain performance level from a machine, and unfortunately silicon is not a perfect process — many times you’ll get a batch that performs better or worse than projected due to the nature of the materials. Thus, you end up with 5–10 different model numbers.” It is not yet clear how efficient the 19-series devices are because they have not shipped at scale, as Leo Zhang, the founder of Anicca Research, summed up in a conversation with Cointelegraph. The first batch of S19 units reportedly shipped out around May 12, while the T19 shipments will start between June 21 and June 30. It is also worth noting that, at this time, Bitmain only sells up to two T19 miners per user “to prevent hoarding.” Hardware problems and competitors The latest generation of Bitmain ASICs follows the release of the S17 units, which have received mostly mixed-to-negative reviews in the community. In early May, Arseniy Grusha, the co-founder of crypto consulting and mining firm Wattum, created a Telegram group for consumers unsatisfied with the S17 units they purchased from Bitmain. As Grusha explained to Cointelegraph at the time, out of the 420 Antminer S17+ devices his company bought, roughly 30%, or around 130 machines, turned out to be bad units. Similarly, Samson Mow, the chief strategy officer of blockchain infrastructure firm Blockstream, tweeted earlier in April that Bitmain customers have a 20%–30% failure rate with Antminer S17 and T17 units. “The Antminer 17 series is generally considered not great,” added Zhang. He additionally noted that Chinese hardware company and competitor Micro BT has been stepping on Bitmain’s toes lately with the release of its highly productive M30 series, which prompted Bitmain to step up its efforts: “Whatsminer gained significant market share in the past two years. According to their COO, in 2019 MicroBT sold ~35% of the network hashrate. Needless to say Bitmain is under a lot of pressure both from competitors and internal politics. They have been working on the 19 series for a while. The specs and price look very attractive.” Minehan confirmed that MicroBT has been gaining traction on the market, but refrained from saying that Bitmain is losing market share as a result: “I think MicroBT is offering option and bringing in new participants, and giving farms a choice. Most farms will have both Bitmain and MicroBT side by side, rather than exclusively host one manufacturer.” “I would say that MicroBT has taken up the existing market share that Canaan has left,” she added, referring to another China-based mining player that recently reported a net loss of $5.6 million in the first quarter of 2020 and cut the price of its mining hardware by up to 50%. Indeed, some large-scale operations seem to be diversifying their equipment with MicroBT units. Earlier this week, United States mining firm Marathon Patent Group announced that it had installed 700 Whatsminer M30S+ ASICs produced by MicroBT. However, it is also reportedly waiting for a delivery of 1,160 Antminer S19 Pro units produced by Bitmain, meaning that it also remains loyal to the current market leader. Will the hash rate be affected? Bitcoin’s hash rate plummeted 30% soon after the halving occurred as much of the older generation equipment became unprofitable due to the increased mining difficulty. That spurred miners to reshuffle, upgrading their current rigs and selling older machines to places where electricity is cheaper — meaning that some of them had to temporarily unplug. The situation has stabilized since, with the hash rate fluctuating around 100 TH/s for the past few days. Some experts attribute that to the start of the wet season in Sichuan, a southwest Chinese province where miners take advantage of low hydroelectricity prices between May and October. The arrival of the new generation of ASICs is expected to drive the hash rate even higher, at least once upgraded units become widely available. So, will the newly revealed T19 model make any impact on the state of the network? Experts agree that it won’t affect the hash rate to a major degree, as it’s a lower output model compared with the S19 series and MicroBT’s M30 series. Minehan said she doesn’t expect the T19 model “to have a huge impact that’s an immediate cause of concern,” as “most likely this is a run of <3500 units of a particular bin quality.” Similarly, Mark D’Aria, the CEO of crypto consulting firm Bitpro, told Cointelegraph: “There isn’t a strong reason to expect the new model to significantly affect the hashrate. It might be a slightly more compelling option to a miner with extraordinarily inexpensive electricity, but otherwise they likely would have just purchased an S19 instead.” Bitmain continues to hold leadership despite internal struggle At the end of the day, manufacturers are always in an arms race, and mining machines are simply commodity products, Zhang argued in a conversation with Cointelegraph: “Besides price, performance, and failure rate, there are not many factors that can help a manufacturer differentiate from the others. The relentless competition led to where we are today.” According to Zhang, as the iteration rate naturally slows down in the future, there will be more facilities using “creative thermal design such as immersion cooling,” hoping to maximize the mining efficiency beyond just using most powerful machines. As for now, Bitmain remains the leader of the mining race, despite having to deal with the largely defunct 17 series and an intensifying power struggle between its two co-founders, Jihan Wu and Micree Zhan, which recently resulted in reports of a street brawl. “Due to its recent internal issues, Bitmain is facing challenges to keep its strong position in the future thus they started to look at other things to expand its industry influences,” Xu told Cointelegraph. He added that Bitmain “will still dominate the industry position in the near future due to its network effect,” although its current problems might allow competitors such as MicroBT to catch up. Earlier this week, the power struggle inside Bitmain intensified even further as Micree Zhan, an ousted executive of the mining titan, reportedly led a group of private guards to overtake the company’s office in Beijing. Meanwhile, Bitmain continues to expand its operations. Last week, the mining company revealed it was extending its “Ant Training Academy” certification program to North America, with the first courses set to launch in the fall. As such, Bitmain seems to be doubling down on the U.S.-based mining sector, which has been growing recently. The Beijing-based company already operates what it classifies as “the world’s largest” mining facility in Rockdale, Texas, which has a planned capacity of 50 megawatts that can later be expanded to 300 megawatts.
To acquire more than 50 percent of the Bitcoin network, an attacker would need at minimum the same number of ASICs that are already in circulation. Network hash rate / terahashes per ASIC = number of ASICs necessary to acquire 50 percent * 41,483,931 network terahashes / 14 terahashes = 2,963,138 S9is * 41,483,931 network terahashes / 23 terahashes = 1,803,649 T15s Calculations for total current ASIC usage: 2,963,138 * 1.75 (crude estimate accounting for older models) = 5,185,492
2,963,138 S9is * $307 per S9i = $909,683,366
1,803,649 T15s * $913 per T15 = $1,646,731,537
Including bulk discount of 10 percent for the size of the purchase and economies of scale.
Total for S9is after discount = $818,715,029
Total for T15s after discount = $1,482,058,383
Infrastructure costs such as housing are a multiple of hardware costs from anywhere between an additional 22 percent to 40 percent, according to the CEO of a large-scale mining operation based in Canada. We will use the low figure to again account for economies of scale.
Variable Costs: Costs of daily electricity consumption for S9i: * 2,963,138 ASICs * 1320W = 3,911,342,160 Watts * 3,911,342,160 Watts / 1000 = 3,911,342 kWh * 3,911,342 kWh * 24 hours = 93,872,212 kWh per day * 93,872,212 * $0.03 per kWh = $2,816,166 per day in electricity Electricity costs based the low-end of the electricity rates in Washington State, the state with the average lowest electricity costs in the United States: https://www.electricitylocal.com/states/washington/ Costs of daily electricity consumption for T15: * 1,803,649 T15s * 1541 Watts = 2,779,423,109 Watts * 2,779,423,109 Watts / 1000 = 2,779,423 kW * 2,779,424 kW * 24 hours = 66,706,155 kWh * 66,706,155 kWh * $0.03 per kWh = $2,001,185 per day in electricity Cost of labor and maintenance above electric costs: 10%. This is an estimate based on an interview with two different large-scale miners in Canada and Georgia who chose to remain anonymous. Average kWh: ((((3,911,342,160 Watts + 2,779,423,109 Watts) / 2) * 24 hours) * 365 days) / 1E12 watt hrs per terawatt hrs = 29.3 TWh This estimate is short of Denmark’s oft-cited annual energy consumption of 32 TWh, closer to Morocco's consumption of 29 TWh.
Check out my new Crypto mining and solution business. For checking it out I'll be giving back a little bit.
Someone is happy. Miner life. Old McDonald had a farm. Money machines. Out of my way plebs. 5 Bitcoin Antminer giveaway. Wolfe Crypto Assets https://www.facebook.com/WolfCryptoIdea/ We will be giving away 2 Bitcoin Antminers at the end of June. 2 different people will win 1 each. 1 entry for a like of our Facebook page. 1 like per share of our Facebook page. 1 entry for any service used. 1 entry per Twitter retweet. 1 entry for Instagram like. On reddit give Wolfe Crypto Assets a shout out and screen shot it. PM me the info and I'll enter you in. All comments here will get you entered in. All inquiries will get you entered. ;) (10 Terahash T9)
bitcoin mining profitable in the US? Where are my calculations off?
Someone tell me where my calculations are wrong. Amazon has this miner advertised: Antminer S9 ~14.0TH/s @ .098W/GH 16nm ASIC Bitcoin Miner So that would consume 14000*0.98=1372 watts. Given my electricity costs (0.12 $/kwhr), I would make $8.19 for every $1 of electricity. In a month, I would make $884. That can't be right. Where did I screw up? Here is a python script to calculate that:
edit: thanks to Personthingman2. 25 vs 12.5 block reward. when I change that, my script outputs: rev=0.000194444444444 cost=4.74798641087e-05 ratio=4.09530330582 profit per month=380.93219223 This is a $4000 unit, so it pays for itself in 10 months. OK. So whether I will ever make money on this depends heavily on the growth of the network hashing rate over time, and the increase in BTC price. edit2: I am guessing that the answer to my question is that I would be lucky for the unit to keep working long enough to pay for itself. It would likely break down before reaching that point.
Why Bitcoin will eventually need a constant, positive Inflation Rate - only practical solution for a long term secure network
I'm going to present numbers that show inflation is the ONLY way to secure the bitcoin network in the long term. this problem wont show up for quite a while, until the block reward is negligible...say in the late 2030's early 2040s, but i still think its important to think about early on. Some back ground numbers i will use in my calculations -Visa network handles about 2000 transactions per second = 63B tx/year -All monetary transactions (cash, cc, etc) = 20000 tx/second = 620B tx/year -$400 for a 1 terahash bitcoin mining rig, using about 500W of power -World M2 Money supply is about $60T (US m2 is about $15T) Lets assume bitcoin is the sole world currency. all bitcoins in the world are worth $60T, or about $3M/bitcoin. What is an acceptable cost to gain control of 51% of the network? $10B? no...apple has $80-$100B in just cash on its balance sheet. $100B? no, again see apple's cash. $500B? Defense dept spending is about that per year. Getting there, but i still think it should cost more. somethin in the range of $1T seems right. So we want the ANNUAL miners revenue to be say $1T on a monetary base of $60T (about 1.7%) Can this $1T come from transaction fees alone? no. Here's the numbers on a cost per tx basis. $1T/620B tx per year = $2.90 per transaction. Not feasible at all. Thats more expensive than credit cards, and its a fantasy if people believe the anonymity gains etc will be enough for people to agree to those insanely high fees. Does off blockchain transactions help? no. Again, the miners need to get that $1T somehow. off blockchain just means few tx numbers but higher per tx fees. The total product is still the same. which ultimately means cost per tx is the same whether its on or off blockchain. So transaction fees and/or off blockchain transactions wont help. What is the other solution? i submit that its a constant inflation rate in the form of an increasing block reward. the block reward should not trend to 0 btc per block by 2140, but instead be set at say 1-4%/year of the current number of bitcoins in existence. with 21M coins in 2040, 630K coins per year or 11 coins per block (instead of the less than 1 coin per block that is currently schedule in the bitcoin protocol). At 3% per year, there would be 30M coins by 2053, 50M coins by 2070, 100M by 2093. inflation is still a set 3% per year. what this does is allow us to get to that $1T/yr in mining revenue while still keeping cost per transaction at say less than 10 cents per tx. 3% of 60T = $1.8T...which would be a very secure network. someone realistically going to spend that much to gain 51% of the network? furthermore logistically it would be tough to get that much. math is tech wrong, but about right. using $400/terahash, that would mean 1.8T/400= 4.5B 1 terahash mining units, consuming 2.3 terawatts of energy. 15TW is average world wide use, so that would mean 2.3/15 = 15% of total energy usage goes to securing bitcoin. Again, these are extreme numbers. maybe i'm being overly paranoid, but i don't see anyone else crunching these numbers, and i'm not comfortable with "we'll deal with the problem when we get to it in 20 years)
Stability of the difficulty. A weakness you may not be aware of.
EDIT: Here's a TL;DR Sorry if I rambled on a bit there. I'll try to make my point a bit more concise here. TL;DR: If bitcoin starts to gain mainstream success, eventually a large percentage of the miners will be in it for the profit and not for the good of bitcoin. My fear is a crash in bitcoin's value after achieving general acceptance as a form of payment could cause a crash in hash rate from miners shutting down when the boss sees a drop in profitability. A crash in hash rate could possibly destroy bitcoin's usefulness as a form of exchange, driving the price down further, killing the currency. In my post I describe a possible solution to make scaling down difficulty to match a sudden drop in hash rate smother. If this interests you, then please read my post and share your thoughts. I feel like this is an obvious flaw in the protocol that was overlooked. Am I wrong? If so, please tell me because I'd like to have my worries eased. ASIC miners and the rapid increase in difficulty have created a new point of failure in the bitcoin network that was never that large of a problem before. No, I’m not talking about the increased centralisation of mining that you see so many people complaining about. I’m talking about the rising difficulty itself and the way the network scales the difficulty. It allows for smooth transitions in the upward direction, but a sharp decline in hash rate could kill bitcoin completely. The readjusting of the difficulty doesn’t really happen every two weeks, it’s really every 2,016 blocks which should be about two weeks in theory. If we were to lose a large chunk of the hash rate, lets say 66%, block discovery time will increase to 30 minutes per block. Maybe that’s not a huge deal, but it definitely impacts the convenience and usability of bitcoin as a means of exchange and would certainly impact the price negatively. Also, it’s possible that we would be stuck at this slow confirmation rate for 6 weeks in this example. Perhaps it is an unlikely scenario that such a large amount of hash rate is lost within a short span, but think about it this way. Imagine a point in the future where ASIC miners account for over 98% of the hash rate. Maybe bitcoin is well on it’s way into mainstream acceptance in this future. There may be large corporations that own large mining farms. These corporations may be publicly traded companies with a responsibility to maximize profits for their shareholders. If the value of bitcoin were to crash, there’s a good chance that some of these corporations may shut down their mining operations because it no longer has a good ROI in the opinion of their CEO. Maybe the next reward halving is coming up soon, that could also cause miners to shut down. This would slow down transaction confirmations, impacting it’s usefulness as a means of exchange and driving the price down even lower, driving even more miners to shut down. This would continue exponentially until the only remaining hashing power is us, the early adopters, true believers, and ASIC manufacturers. We could very easily end up in a situation with hour long confirmation times or longer, and the next difficulty adjustment being months away. That would essentially completely kill off bitcoin. It’s not logical to assume there will never be a large drop in hash rate between now and 2140. It’s not possible to predict political events that far into the future. Maybe world war 3 happens and China decides to unplug their whole country from the outside internet. Maybe Butterfly Labs successfully ships out 1,000,000 terahash miners that run on sunshine dust and unicorn farts and they quickly become the defacto standard, and then it’s discovered that they are a fire hazard. Many people are killed, homes are lost, and people just start turning them off out of fear. You just don't know what will happen that far into the future. I have an idea about how the protocol could be modified to protect against this sort of scenario and allow for a sharp decrease in hash rate without losing its usefulness as a means of exchange. There should be an emergency mode that drastically cuts the difficulty rate. This emergency mode can be requested by any node in the network but will only occur if the network has consensus from the nodes. The request could be triggered by two possible events. The first trigger event should be if no blocks are discovered for a certain span of time, lets say an hour and a half. The second event should be if a certain number of consecutive blocks take over 25 minutes to discover. The reason for using 25 minutes as the trigger is because that would require a loss of 55% so in theory a single person couldn't trigger it without having over 50% and in that scenario we have worse problems than the falling hash rate. Once the event is triggered, the difficulty should be slashed by a factor of 10x and reduce the block reward to 0. If it’s impossible to have no reward blocks, then use a dust amount of bitcoin like a few Satoshies. If the event was triggered by 25 minute confirmation times then the new emergency block discovery time will be 2.5 minutes. This will have the immediate effect of speeding up transactions that have possibly been waiting awhile for confirmations. The rapid block time will have the secondary effect of quickly and more accurately calculating the new hash rate of the network. (More data points over a shorter spread of time.) The removal of the block reward serves to disincentivize miners with lots of hashing power from trying to trigger the event on purpose so they could make more bitcoins at a lower difficulty rate and also prevents the creation of large amount of bitcoins since it would be moving into an unknown block discovery rate. This emergency event should not last long because miners would just shut down even faster with no block reward. Lets say it only lasts 96 blocks. Assuming 2.5 minute confirmation times, 96 blocks would take 4 hours to mine but that number could vary depending on how much hash rate was lost. Lets put this in perspective. A sudden loss of 55% of the hash rate would result in a loss of 4 hours of block rewards. A loss of 90% would be 16 hours with no block rewards. After 96 blocks have been mined, the network can make an estimation on the new hash rate based on the speed those blocks took, and generate a new difficulty to resume normal 10 minute confirmation times and normal block rewards. However, such a short span of time may not be enough to generate an accurate difficulty for the network. The network should recalculate the difficulty again after 432 blocks (3 days), and then resume the normal 2 week schedule. Each time an emergency event is triggered, the next scheduled halving of the block reward should be moved back by 96 blocks. That way these events have no effect on the mathematical total of bitcoins that can possibly be mined by 2140. Even if these events are very unlikely and it’s possible that this scenario will never play out, it would be added security to the value of bitcoin to have fail safes in place in the event of huge losses in hash rate. It could only add to the strength of bitcoin. It gives it the robustness it needs to survive a crash in price in a world where the miners are mostly interested in profits and ROI, not the good of bitcoin. EDIT: Here's another TL;DR of my proposed solution Basicly the nodes can request an emergency drop in hash rate if it notices block confirmation times rise above 25 minutes. This would require a loss of 55% of the hash rate. If the nodes have consensus on this request then the difficulty gets slashed by a factor of 10 and the network will mine 96 blocks with no block reward. That will take 4 hours with a 55% hash rate loss and 16 hours at a 90% hash rate loss. After the 96 blocks are mined, the network will calculate the appropriate difficulty based on the discovery rate that those blocks were mined. Normal block confirmations and block rewards will then resume at that point. What makes it nice is it will only be triggered by extreme cases. Also, mining for 4 to 16 hours for the good of bitcoin is an easier pill to swallow than just keep mining indefinitely into a possibly dying system for the good of bitcoin.
Where is the network difficulty headed, come November?
Reposted for accuracy. (Read: My math skills are the result of public education.) KNCMiner announced today that they're doing encapsulation on their new Scrypt ASIC chips, and then when they're completed, will be shipping to Stockholm for integration and testing, buildout and finally...shipping! I have read on forums that they have sold 3,000 Titans via pre-order, for batch 1, at 250MH/s nominal performance, each. I figured it was time to look at my "hashrate/difficulty prediction" again and see where it may actually be, by the time the snow's falling. All of the below is calculated with a Litecoin price of around $5. Let's assume for a moment that both Alpha Technology and Mining ASICs Technologies have also sold around 3,000 systems on pre-order (probably a safe bet) and all three expect to ship in September-October. 9,000 systems @ 250MH/s = 2,250,000MH/s. That's somewhere in the neighborhood of 2.25 TH/s being added to the network in roughly two months' time...that we can account for. The current network hashrate, as I post this? Not quite 1 TH/s...it's 896 GH/s. But at the current rate of network expansion, we're going to be 1 TH/s by the time these systems ship, easily. So...let's say we're looking at a 3.5 TH/s Litecoin network by November. What does that mean? When the Bitcoin network hit 3.5 TH/s back in May of 2011, the difficulty was around 244,000. Litecoin's difficulty is currently around 28,000. You can probably see where this is going, already. Fun with mining calculators time. Say you have one 250MH/s miner and deployed it TODAY (impossible, but for the sake of argument). You're looking at pulling in 9 LTC/day with it. If you pay $.10/kwh you're very lucky not to live in California, but we'll say that's the case. You pay around $4.50/day in power. So you walk away with $42.50 worth of Litecoin, at $5/each. If you somehow managed to freeze the network at that difficulty and the coin at that price, you'd pay off your $9,200 purchase of hardware in roughly seven months or so...or if you bought a Titan at $10,000 you're looking closer to eight. But since difficulty marches on, forget that entire concept. Now...say you get your system after all three companies have shipped and their customers have deployed them, and we've seen the network rocked to the tune of two-and-a-quarter terahashes per second. Oh, it's a rosy picture... Now, with the network difficulty having blown up to 244,000 the miner with a 250MH/s system is mining 1.03 Litecoin per day. And if my estimates are correct...this is NOVEMBER, we're talking about. At the current price of $5/LTC and $.10/kwh you are pulling down a healthy $0.80/day in profits, after power. If you again had the power to freeze the hashrate and price, you'd be able to pay off that hardware purchase in, oh...roughly 35 years. To have a REASONABLE shot at getting a return on your investment (around 5-6 months), Litecoin will need to be $70 by November and climbing steadily, in concert with network hashrate. Bear in mind, again that there is nowhere else for that hashrate to go but Litecoin. Nothing else will profit the Scrypt miner. So what will happen? There is built-in hardware cost here that has to be recouped and the only real way of doing that is by mining...and there's only one game in town for Scrypt mining: Litecoin. It's going to be a really, really wild fourth quarter for this year. Either the miners mine and hoard, decreasing supply and demand increases radically, or miners take heavy losses on hardware, can't afford to run them and the Litecoin network contracts until they CAN make money with them. In the interests of self-preservation, I have a feeling miners will start hoarding. Soon.
I have been helping a friend develop business strategies at a Bitcoin start-up over the last few months. In the course of this work, the topic of Bitcoin mining appears often to be fraught with misinformation and uncertainty, especially for individual miners who unfortunately may find it difficult to return an adequate profit in many cases. This informal guide covers some important issues prospective miners should consider to avoid headaches and financial loss. The information is derived from experience deploying a 400 TH/s system scheduled to come online in around December. Opinions are my own; I’m happy to entertain constructive feedback. This year, the Bitcoin network will award miners nearly USD 500 million, at the current price of USD 375 per bitcoin, to participate in a process known as mining. Unsurprisingly, this has attracted significant interest not only from Bitcoin advocates, but from speculators and investors as well. Regardless of one’s motivations, the business of Bitcoin mining must ultimately be profitable, or at least operationally viable, if there is to be any chance of success. HOME MINING Acquiring and personally managing ASIC miners is probably the most fulfilling way to mine bitcoins. It provides the greatest level of transparency, but requires a certain level of technical proficiency to set up and run. Advantages: 1) No hosting fees payable 2) Full control of operating parameters 3) Direct payment from mining pool Disadvantages: 1) Purchasing the latest mining hardware is inherently risky because the ongoing development of energy-efficient ASIC chips requires expertise, time and millions of dollars. R&D is usually funded by customer prepayments with no guarantee of timeliness or success. It is not uncommon for miners to incur financial loss and opportunity costs when a supplier fails to deliver 2) The retail price of hardware is typically marked up anywhere from 25% to 500%, or more, depending on market conditions. This creates a barrier to profitability, making it harder for miners to recoup hardware costs if they are unable to negotiate for volume discounts 3) Shipping fees and import tariffs can cost hundreds of dollars per unit, especially if importing equipment from overseas. This adds to the cost of hardware and must be taken into account when calculating the return on investment 4) Shipping time varies greatly. Each day spent in transit incurs an opportunity cost 5) Miners need to set aside space, usually in the home, to locate mining equipment 6) Many mining units may generate excessive noise, and heat that requires around the clock ventilation to maintain an optimal operating temperature range 7) The average mining unit draws up to three amps of current. A system containing twenty units could easily exceed the power limit in a typical home 8) Electricity is by far the largest expense in any mining operation, making up around 90 percent of operating costs. If the price of residential power is materially higher than the rate paid by commercial operators, it makes home mining uncompetitive CLOUD MINING Buying into a cloud mining service is often marketed as a convenient and hassle-free way to get in on Bitcoin mining. As the mining assets are managed by an intermediary, getting a breakdown of operating costs prior to purchase often proves difficult. This makes it challenging for potential customers to make a fully informed buying decision. The unspoken truth is that some cloud miners incorporate obsolete equipment—cheap miners from previous generations or liquidated, unprofitable hardware—into their cloud to sell to unsuspecting customers. Older mining units can consume 80% more power than the current generation miners, leaving very little profit for the customer. In addition to the acquisition price, those in the market for cloud mining should consider the power consumption of the cloud on offer, including changes over time as new mining units are added to increase total capacity. Advantages: 1) Start earning immediately. No waiting weeks or months for equipment delivery, installation and set up 2) Convenient and fully managed mining service means customer needs not be technically inclined or involved in day-to-day operations 3) Professional hosting service ensures optimal performance and low operating costs. Commercial hosts may be able to purchase electricity for a materially lower cost than residential customers 4) Acquisition price is often reasonable. Sometimes, possibly, too good to be true 5) Some platforms allow miners to sell their assets to other traders Disadvantages: 1) Not all hashing power is comparable. For the same acquisition cost, more energy-efficient miners are better because they use less power and return higher profits. When buying hashing power from a cloud, the buyer should ensure he is not getting obsolete hardware. Often this is not possible to verify without a basic understanding of the costs involved, however subpar earnings is a good indication that further investigation is required 2) Hosting and cloud management fees are typically payable. Sometimes there is little transparency in pricing, resulting in unexpected cost to the customer 3) Miner has little input into how the cloud is managed COSTS BREAKDOWN The amount of money earned from Bitcoin mining over a short period of time, say one week, is fairly easy to calculate. Given mining is a zero-sum game where new entrants dilute existing participants and the mining reward is roughly shared on the basis of each miner’s contribution to the overall hash rate, we can derive profit by estimating the income and costs. Mining Income: Weekly mining bitcoins created = 25,200 = 25 bitcoins x 6 times per hour x 24 hours x 7 days Assuming hash rate is at 300,000 TH/s, bitcoins earned weekly per one terahash of processing power = 0.084 bitcoins = (1 terahash/ 300,000 terahash) x 25,200 bitcoins Table 1: Weekly earnings per one terahash of computing power
As new miners enter the market, an increase in hash rate dilutes the mining reward. This is the source of much uncertainty in mining because it is difficult to accurately forecast the rate of increase. Dilution reduces a miner’s income while the amount of work is the same. Mining Costs: Electricity typically comprises around 90 percent of total operating costs. The two determinants of electricity cost are price and the amount of electricity consumed. If we take a hypothetical 700 GH/s system that is rated at 490 watts, we can normalise it: 0.7 kW per one terahash = (1 terahash / 0.7 terahash) x 0.49 kW Electricity used per week is: 117.6 kWh = 0.7 kW x 24 hours x 7 days If we know the cost of electricity, the dollar value of electricity consumed in one week can be estimated. For reference, power prices in Australia are between USD 14 cents (commercial rate) and 19 cents (residential rate). China averages around 8 cents, while other places can be cheaper. For example, in Georgia, USA the cost of commercial electricity is around 6.5 cents per kWh. Table 2: Weekly electricity cost of running a one terahash system
Other costs to consider include mining pool fee (typically 1 percent of earnings), hosting fee (depends on host) and other expenses such as air conditioning if hosting at home, maintenance, etc. Profit: Using the assumptions that hash rate is at 300,000 TH/s and bitcoin price is USD 375, we can work out the profit. Moreover, knowing the basic cost of Bitcoin mining can help prospective miners avoid offers that are too good to be true. To simplify, we ignore other running costs: Profit = (bitcoin price x bitcoins earned) - electricity expense Table 3: Estimated profit from running a one terahash system for one week
These figures serve as a good benchmark for comparing your personal performance. Where the electricity price is known, the difference between the calculated and actual profits can be attributed to two things: 1) Energy efficiency of mining units can cause significant deviation, especially when the cost of electricity is high. This is usually the case if obsolete equipment is being used 2) Hosting fee, mining pool fee and other costs also contribute to the difference Return on Investment: The rate of return is a measure of how much miners make for a given investment size. Implied annualised return = (52 weeks x profit per week) / (hardware cost + shipping fees + tariffs + installation and setup costs) The current price of ASIC miners runs at around USD 500 per terahash, excluding international delivery and insurance that can cost between five to 20 dollars per kg ($50 to $200 per unit). As a general rule, higher operating profit and lower capital costs are preferred. Investors endeavour to break even quickly on the initial hardware investment and make a profit on top of that. The problem with this model is that it implies the hash rate remains unchanged for the entire year. In reality, the hash rate is likely to increase depending on a variety of factors. Therefore, the annual profit forecast is sensitive to changes in the hash rate as well as bitcoin price. This is a complex and interesting topic that deserves its own post. Please, keep in mind that actual mining results will very likely be less than what is indicated by this simple calculation. Under some scenarios, even informed miners can experience financial loss. OPERATING RISKS 1) Liquidity risk: Bitcoin trading is rather shallow. As such, miners may experience high trading frictions when selling bitcoins to obtain cash. A bid-ask spread of up to 10% is not uncommon in some cases. Furthermore, most mining businesses rely on the liquidation of mined bitcoins to cover operating expenses such as electricity and hosting. The combination of these two factors may result in unexpected trading costs to the miner if there is insufficient demand from bitcoin buyers. 2) Price risk: Bitcoin is highly speculative and this is reflected in its price volatility. There is no guarantee that it won’t be worthless by next year. Therefore, the miner should keep in mind that the market price is just as important as the amount of bitcoins he holds. Bitcoin price is influenced by multiple factors outside of the scope of this discussion. 3) Competition risk: Bitcoin mining is a zero-sum game. While the size of the reward is fixed, new entrants are permitted to enter at anytime reducing all miners’ share of the reward. When bitcoin price is high, more new competitors are attracted to mining, further eroding all participants’ income. 4) As a function of the Bitcoin protocol, the mining reward will be halved between May and June of 2016. When this happens, all miners will experience an immediate decline of 50 percent in income with many operators becoming unviable. This effectively gives new entrants less than 1.5 years to break even and turn a profit. The short window of opportunity is troublesome because it makes mining significantly less profitable as the deadline draws near.
How Bitcoin block chain and mining work (probably > ELI5)
I actually posted this in a different thread, but thought it might be useful to others. Let me know of any errors! Bitcoin maintains a list of every transaction in a ledger called the blockchain. The block chain is divided up into sequential blocks that contain the movement of bitcoins from one address to another. Each new block has a set of new transactions along with a special transaction that generates new bitcoins and a link to the most recent block in the block chain. During the creation of this new block, new transactions are being passed around to all of the computers connected to the Bitcoin peer-to-peer network. Some of the computers connected to the network are creating or watching for transactions, others are miners that are trying to add new blocks to the block chain. A miner's goal is to be the first one to add the next block to the block chain. He receives a blockreward of (currently) 25 bitcoin plus all of the transaction fees (so its best to add as many paid transactions as you can). For all of the computers on the network to accept the miner's block as the next block, the miner's computer has to do some math that involves the calculation of hashes. A hash is the result of a function that takes arbitrary data and make a single number out of it. There are many different hashing functions, but the one used by Bitcoin, SHA256, has properties that are particularly good for Bitcoin's purpose. Here are a couple of examples of SHA256 I ran on my computer:
The long list of numbers and letters is the SHA256 hash of "hello you" and "hello yoo" in hexadecimal (base 16 numbers). The numbers above may not look that big, but they are actually greater than the number of atoms in the Milky Way galaxy. For SHA256, no matter how large or small the input is, the result will always be a 256 bit number somewhere from 0 to 2256 -1. The same input always gives the same result. Notice that while there was only one letter difference between the two inputs in the example above, the resulting hashes look nothing alike. In fact, they are so different that we can't even guess what any of the numbers will be until we calculate the full hash. This property, along with the extremely large set of possible hash results and the repeatability of the hashing function is what makes it useful to Bitcoin. Instead of using those simple little sentences, Bitcoin uses the block chain as the input to the SHA256 hash. With this, any computer on the network can ensure that no data has changed in the transaction ledger since even a small change of 0.00000001BTC anywhere in the block chain will cause a completely different hash result. In addition to ensuring the block chain hasn't been tampered with, the miner uses SHA256 to calculate the hash of the next block added to the block chain in what is called the proofofwork. For this calculation, he needs to come up with a hash that meets a certain criteria: the hash must be less than a certain value. This number is what is called Bitcoin's difficulty. The difficulty is adjusted so that only one miner every 10 minutes (approximately) can solve the proof of work problem. If the miners all just calculated the hash of the current transactions from the block chain ledger and the new transactions, they would all get the exact same value (which would almost never be less than the difficulty value). So there is another value input into this hash calculation called the nonce. This is another big number, but can be any number the miner chooses. By trying new nonce values and recalculating the new block hash over and over, eventually one of the miners will come up with a hash value that is less than the difficulty value. This is simply trial and error - sometimes a new block is calculated in minutes and other times it can take over an hour. The more hashes you can do in a second, the better your chance of solving the problem. This is why miners are keenly interested in gigahashes and terahashes per second. When a miner has found a nonce that works with the list of new transactions and meets the difficulty criteria, he notifies all of the other computers on the network of the new block (with one of the transactions being his own address with the transaction fees and block reward!). All of the computers in the Bitcoin network now verify that this new block is valid and meets the proof of work criteria. Those on the network sending and receiving transactions see the new block as 1 confirmation for any transactions in that block. Miners now start over again with a new block and the hope that they will be the one to discover the next nonce. (When miners don't all move on to the same next block, you run into block chain splits and potentially nefarious things like 51% attacks, but its in everyone's best interest to all be working on the same block chain - the longest block chain.)
We have compiled a short table showing how much Bitcoin you could mine with 1 THS hashrate starting November 1st up until the estimate for February 1st this year. This period covers 3 months and as you can see the coins mined for just these 3 months pretty much halved due to the rapid increase in the network hashrate and thus the network difficulty. The table shows the numbers of coins you ... 10 terahash miner. ¿Cuánto cuesta un ordenador para minar Bitcoin?. Cuando la red alcanza un hash rate de 10 TH/s significa que puede hacer 10 billones de cálculos por segundo. Minería. La minería en Bitcoin es el proceso de. Independientemente de la perspectiva anti ASIC de parte de la comunidad minera, el uso de estos dispositivos sigue vigente. On Sep. 10, 2019, at 12 p.m. EDT, it used to be four% extra successful to mine the BCH chain. Do you wish to have to maximise your Bitcoin Mining doable? Plug your personal into the arena’s maximum successful Bitcoin mining pool or get began with no need to possess thru one among our aggressive Bitcoin cloud mining contracts.. Each chains’ SHA256 hashrate mixed on Sep. 10 crossed the 100 ... Find out what your expected return is depending on your hash rate and electricity cost. Find out if it's profitable to mine Bitcoin, Ethereum, Litecoin, DASH or Monero. Do you think you've got what it takes to join the tough world of cryptocurrency mining? If you want a Single miner to do it, then you need the miner to be the size of big room (Possibly more). For the moment, the miners available only have capabilities in range of 15–30 TH/s. 1 PH would be equivalent to 1000 TH. So you could do the m...
Hashflare Bitcoin Mining I Hashflare TeraHash Update
Get yours here : terahash-mining.com. Thatch Roof House: Full Bushcraft Shelter Build with Hand Tools Saxon House - Duration: 1:21:06. TA Outdoors Recommended for you Close. This video is unavailable. How to BitCoin mine using fast ASIC mining hardware - Duration: 27:15. Barnacules Nerdgasm 1,689,855 views. 27:15. Building a 3.5kWh DIY Solar Generator for $650 - Start to Finish - Duration: 33 ... How to CPU Mine Bitcoins 2018 Updated Windows 10 - Duration: 6:29. howtwos101 50,101 views. 6:29. NiceHash how to buy hashing power tutorial - Duration: 7:54. NiceHash 181,743 views. 7:54 . PROOF ... Hashflare Bitcoin Mining I Hashflare TeraHash Update // In this video wanted to update you awesome folks on bitcoin mining and how profitable bitcoin mining can be. I love hashflare and use it in ...