Yes profitability made huge jumps in the last couple of months (which may be a signal to wait not buy GPUs, see below), but you can't extrapolate today's anomalous returns without taking on huge risks. Risks most people contemplating the bandwagon do not understand. When you tell newbs their cards will be paid off in a few months you are essentially telling them difficulty will not increase and bitcoin prices in the mid ten thousands (and the accompanying alt coin spikes and enthusiastic bidding on nicehash, etc) is the new normal, or that prices will keep rising.
Also did you put a watt meter on the power outlet to your rig to measure the actual power being consumed? Don't rely on online claims of card X uses Y watts.
Are you using non-baseline pricing for your power costs? Locally baseline is $0.16 and newbs use that in online calculators to get to their paid in a few months type of numbers and then are surprised when they are actually billed at $0.25 when the GPUs blow past baseline usage. Fortunately for them Time of Use rate plans are coming and if they select the right one, not the default one, they can get around $0.14 if they shut down for the 6 hours of high demand pricing or $0.18 if they run 24/7. Again, newbs need to do a lot of research before joining the scene.
OK, so you are arguing that recent price spikes have let you pay off all your cards. Congrats. Now go back and on the day you bought each card figure out how many bitcoins you could have purchased with that money on that day. Do that for each card. Add up all those hypothetical bitcoins. What is the current value of those coins. Its likely more than what you made mining. You could buy all those cards today with those bitcoin proceeds and probably have lots of cash left over. This is the point I was trying to make in scenario (2) in the original post. Rising prices is not a signal to buy a miner, rising prices is a signal to buy coins and delay purchasing a miner. Unless of course you need the GPU for non-mining purposes.
If you don't do this alternative use of the money calculation you can't really say you came out ahead on mining. Newbs need to understand this sort of thing before joining the scene.
Bitcoin has fewer flaws than any crypto currency. There is one fundamental "flaw" most people complain about...
Here's the one fundamental flaw that gets little attention: mining is not decentralized.
Bitcoin has deviated from its design and its security is compromised. The security of the bitcoin blockchain is based on the assumption of a globally distributed network of ordinary users using their computers to maintain the blockchain (mining). There are two deviations. First, mining is dominated by a small group of individuals/organizations with expensive specialized hardware (ASICs). This makes a 51% attack plausible should a cartel form, note that a mining pool reached 50% a few years ago. Secondly, 70% of miners are located in a single country and dependent upon government supplied inexpensive electricity. This makes government interference or manipulation plausible. With distributed mining cartel and government interference was not plausible, but with the current mining situation both are plausible.
This is fixable, it is software after all. However the necessary software updates (ASIC resistant mining algorithms, updated as necessary - already in use by other coins; maybe a move from PoW to PoS - we'll see how that goes with etherium; etc) have to overcome inertia, politics and greed - the voting on such changes will be done by the miners who like the current system and have expensive ASiC hardware that need to be paid for. Users can only abandon bitcoin and move to another coin.
Bitcoin may be replaced in the short/medium term, I don't know about Buffets 5 year timeframe but in general he is likely correct. The future is blockchain technology, any individual coin is just a user of that technology and entire replaceable.
What we need is a car analogy.;-) Blockchain technology is like internal combustion engine technology. Bitcoin is like the Ford Model T, an early user of the technology, the first to gain wide acceptance and mindshare, but entirely replaceable by the inevitable competitor with newer and better technology.
Now some of you might be tempted to mention network effect. Network effect requires a high switching cost. Users have little switching cost moving from one coin to another. Miners have a high switching cost and that is why they will resist algorithm changes, but users are quite free to move on to something new.
It seems that these companies (Microsoft and Ubuntu and others) are forgetting everything about sound software development practices here. They're in such a hurry to deploy patches that they aren't taking the time to fully test them. The cure is worse than the ailment.
Both Microsoft and Ubuntu are plagued by the vast permutations of hardware out there, all the combinations of motherboard, cpu, video, etc. Aren't there identified problems with various anti-virus software? Did some driver developer out there try something tricky too that is incompatible with the fix(es)? Historically various problems with Windows came from 3rd party drivers not necessarily Microsoft itself, perhaps Ubuntu is having similar problems?
That is today's profitability. You can't project that into the future. More realistic profitability calculators factor in difficulty increases.
Also power costs seem off. $0.15 looks more like baseline residential pricing. GPU mining will likely blow way past baseline and into the next higher tier of pricing. As I mentioned earlier you need to make sure comparisons are using beyond baseline pricing.
If coin prices decline, difficulty decreases and if coin prices rise, difficulty increases. I think they balance themselves out.
No. More miners (more hashing power technically) more difficulty. Less miners less difficulty. More people joining the mining scene increases difficulty regardless of what price is doing.
Plus, there are automated pool that will mine the most profitable coin and auto-exchange it to your preferred coin.
Not a "plus", that is what your profitability link above assumes, constant switching to the more profitable algorithms as alt-coin enthusiasts bid for your hashing power.
Who know what will happen. From the past it looks like 3 month ROI is plausible. It has been better before and it has been worse but just an average of 3 months.
The past is what I described in scenario (2) earlier. That was actual data since the summer, mildly rounded for convenience. Those nicehash estimates do not project out as you assume. Plus they understate costs since they only factor the GPU power requirements. As I mentioned earlier you have to use a watt meter and measure what the entire computer is consuming when it is hashing.
I totally agree with you that investing has been far more profitable than mining. If you just want to get a sweet GPU or system paid for my mining in about 3 months without a lot of work, it looks to be possible. Mining rigs or cypto investing is almost a part time job since you have to keep on top of so many things and maybe not the best use of your time.
Even if pays for itself in 6 months, it's still a good deal. Most stores will give you a 0% APR for 6 month loan for buying over $500.
If you want the GPU for non-mining purposes, sure, mine to subsidize the purchase. Why leave the system off/sleeping if mining is profitable? Maybe go up a model and get a slightly more expensive card, scenario (1) earlier. I'm just arguing that if the only reason for the GPU is mining things need to be viewed differently, a delay to purchase possibly warranted depending on what coin price expectations are.
They are different kinds of investments. Of course you can make more overall by day trading coins. That does not remove the profitability of the mining that allows those coins to function in the first place.
As I said, if you are buying a GPU for non-mining reasons you might as well mine *iff* profitable. But if you are buying a GPU only to make money mining then just buying coins needs to be an alternative you consider.
We can incorporate the dense mining rig into the day trading you're talking about, pointing it at newer lower difficulty coins for the long term.
As I mentioned elsewhere, delay that until coin prices are no longer increases and you may be better off. In scenario 2 above by delaying the two purchases by 3 and 6 months respectively you still get the GPUs for the long term but you also have an additional $2K.
The cryptocurrency technology is elegantly constructed to sustain itself. It's counter-intuitive, or just plain wrong, to take the profitability of both sides separately so hard.
Now we have a topic change, not mining for profit but mining to altruistically support the blockchain network. I was only arguing over the profit motive.
They can still buy and use cheap ASICs, just have to ship them outside China.
ASICs have a short lifespan and no salvage value beyond scrap metal pricing for their heat sinks. Every day in transit and setting up at a new location means an ASIC is likely losing its most profitable current mining time, difficulty increases reduce profitability.
The Chinese miners will just set up shop elsewhere. Maybe they won't move at all since local authorities in China benefit greatly when underutilized power capacity is used by bitcoin.
In theory, but its going to hurt to see their ASiCs lose their more profitable days, perhaps even no longer be profitable in the timeframe necessary to relocate.
Case (2) above is based on real numbers, modestly rounded for convenience. Reality is 3x the earning just buying coins directly 6 months ago compared to buying gpus last summer and last fall. Delaying those purchases 3 to 6 months is the difference between paid for $1,000 GPUs (mining) and paid for $1,000 GPUs plus an additional $2,000 in cash (buying coins directly).
Sorry, I described the scenario incorrectly. 3x is the result of buying coins rather than a GPU 6 months ago and again 3 months ago. I erroneously implied by all the coins six months ago, that would result in 4.5x but that is an unfair comparison since money is being spent at different times. Spending money at the same time is the fair comparison and that is a 3x result.
What are you blabbering on about?
It's cheaper to have heatsinks made in China and shipped to most places in the world.
Why would you think it's cheaper than shipping alone to locally manufacture them?
I was actually thinking source them locally, not necessarily manufacture them locally, thoughts and typing diverged.
Fill up a container, stick it on a boat. You'll pay a flat rate for the container.
And while in ocean transit the ASIC experiences a drop in profitability. Also I am are talking about individuals getting their hosted in China ASICs delivered to them. As for farms, containers may be more viable but again transit time and setting up a new farming site eats in the more profitable days those ASIC would have had.
Case (2) above is based on real numbers, modestly rounded for convenience. Reality is 3x the earning just buying coins directly 6 months ago compared to buying gpus last summer and last fall. Delaying those purchases 3 to 6 months is the difference between paid for $1,000 GPUs (mining) and paid for $1,000 GPUs plus an additional $2,000 in cash (buying coins directly).
Paying some of the highest local market rates, my normal workstation with an outdated video card pulls in real actual profit by the real actual honest numbers.
As I said, if you are buying a GPU for non-mining purposes then mining to subsize the GPU is fine. Might even justify getting a slightly better GPU. But if the GPUs are being bought solely for mining then things radically change.
Can you guys start shutting up if you don't want to put the work in?
Work includes investigating the alternatives, as I did. Its not as simple as a rig being profitable, you need to calculate the alternative uses of that money, model different scenarios (coin prices, difficulty increase rates, etc).
Buy a dual 1080 Ti. Stick it your computer and it will pay for itself in 3 months. You can also game on it once in a while.
*Iff* coin prices do not decline.
**Iff** difficulty levels do not increase.
Normal difficulty increases will likely stretch that 3 months to 6 months. 3 months is an extrapolation of today's revenue, poor planning to bank on that.
Now if you want to gamble on increasing coin prices, you can take the money for those two 1080 Ti's and invest that up front by buying the coins directly at an expected relative low. You may be tempted to make an "averaging in" argument, but we aren't talking about consistent payroll deductions into a 401K here. Every two weeks the earning power of those 1080 Ti's will likely decline, your "contributions" will decline. If you are expecting the sort of volatility that "averaging in" helps protect against you are also expecting to stretch out the timeframe those 1080 Ti's need to reach profitability, if ever.
Again, *iff* you are gambling on increasing coin prices you may make more by just buying coins up front and then cashing out after a few months. Then buy a 1080 Ti for gaming if you really want one and if you think coins prices have stabilized or will decline. If you expect further increases its probably best to delay the 1080 Ti a little longer. Read case (2) above, its based on actual mining revenue and coin price of the last six months, numbers slightly rounded for convenience. 3x the revenue for delaying the GPU purchases.
Just to check, that 3 month projection of yours is subtracting out power at above baseline residential pricing?
Seriously, your 3 month claim to break-even makes me skeptical, if you had said 6 months that would have been more plausible. Sit down with a spreadsheet and model different scenarios. People who do so often find things are quite different than more casual analysis predicts.
The free market guarantees meritocracy, right? right??? RIGHT!?!?
Adam Smith, "An Inquiry into the Nature and Causes of the Wealth of Nations" - 1776, did not predict joy on every single day, he predicted better outcomes in the long run than otherwise available.
So yes, capitalism allows us to switch to macOS, Linux, FreeBSD, etc to reward those and/or punish Microsoft. That is the mechanism by which a meritocracy is determined.
If you want to get in on the cryptocurrency mining scene...
Stop yourself, don't do it. Take either of these two paths:
(1) You were going to buy a GPU anyway for some non-mining reason. Go ahead, buy the GPU. Maybe, **maybe**, buy a model up one level of performance/price from what you would have otherwise bought, if its a low to midrange model. Say if you were otherwise planning on a GTX 1050 Ti 4GB ***maybe*** get a GTX 1060 6GB, ****maybe****. If you were otherwise getting a high performer, say a 1070 Ti for that 4K monitor, do *****not***** go up a level to a 1080 Ti. Then let the GPU mine when you are not using the machine. Use a watt meter to determine the total power consumption of your machine to determine power usage, do *not* trust online references that say your GPU uses so many watts. When your GPU is mining other parts of the computer are also drawing power, especially the CPU which may also be mining. You want to know the total system power and make sure your mining proceeds exceed that amount. Be sure to use above baseline residential power rates in your calculations, do *not* just look at your current bill and expect the current rate. If it is not profitable to mine do *not* fall into the trap that "the coin price will eventually rise and make it profitable", that is a losing game. Instead, take whatever money you would spend on power and just buy the coins directly, you will have more coins that way if the price rises. But above all else, do not get into the mindset of joining the mining scene, that is a path to losing money. Stay in the scene "the GPU I have anyway can make some coin when I'm not using it".
(2) Take whatever money you were willing to spend on a GPU for mining and just buy coins with that money. You will likely do better that way if the price rises. Many miners fall into the trap that they are profitable and pat themselves on the back. They do not consider the opportunity cost of the alternative of just buying coins directly. The following are very rough estimates but the point will nonetheless be clear. Lets say you spent $500 on a GPU last summer and another $500 on a GPU last fall. At above baseline residential power rates maybe you have about an extra $1,000 after factoring in power. Congrats your GPUs are now paid. However your friend bought $500 worth of bitcoin in the summer and another $500 in the fall and now has $3,000 worth of bitcoin. You are at net $0, he is at net $2,0000. If you are willing to gamble on increasing coin prices you may be better off just buying coins directly. Things are not as simple as a mining rig being profitable, the opportunity cost of the just buy directly must be considered. Many other things must also be considered before joining the mining scene.
For a while you couldn't turn a profit with a GPU. With insane bitcoin prices you can earn (I'm told) about $250/month after electric cost with a 1080ti. Which is why the price of 1080s has gone up.
But you are not mining bitcoin. If you are receiving bitcoins from GPU mining you are likely selling your hashing power to someone doing alt-coin mining and who is paying for this hashing in bitcoin. Because who wants whatever goofy alt-coin they want to mine.:-)
Find me some mining hardware that will make enough money to make back my investment before its obsolete. It was only profitable in China because the government paid for everything.
Especially after you pay to have that heavy ASIC device shipped from China to wherever your are. Maybe you can pay to have heat sinks removed and have domestically manufacture heatsinks attached after the ASIC blades arrives. Either way, hurts you profitability and makes ASiC obsolescence that much closer.:-)
The biggest, and maybe only problem that will arise will be the difficulty being too high when they shut down the farms, leading to a period of slow block solves until the difficulty adjusts. The network will work just fine without the chinese mining farms
And the blockchain network will be more secure. Bitcoin has deviated from its original design in two ways and both compromise blockchain security. First, miners are no longer a diverse group of ordinary users and their personal computers, mining is dominated by a relatively small number of ASIC farms. This makes 51% attacks more plausible, we had one pool get to 50% a few years ago. Secondly, ASIC mining is concentrated in a single country, 70% of the hashate give or take. This obviously destroys the notion that bitcoin is beyond government meddling. These ASIC miners are dependent upon cheap government controlled power.
It sucks to have invested money in ASIC hardware and colocated you gear there but this move would help to get bitcoin back on track. Hopefully closer to the globally and widely distributed mining that is necessary for blockchain security, something we do not have today.
If you want to get in on the cryptocurrency mining scene...
Stop yourself, don't do it. Take either of these two paths:
(1) You were going to buy a GPU anyway for some non-mining reason. Go ahead, buy the GPU. Maybe, **maybe**, buy a model up one level of performance/price from what you would have otherwise bought, if its a low to midrange model. Say if you were otherwise planning on a GTX 1050 Ti 4GB ***maybe*** get a GTX 1060 6GB, ****maybe****. If you were otherwise getting a high performer, say a 1070 Ti for that 4K monitor, do *****not***** go up a level to a 1080 Ti. Then let the GPU mine when you are not using the machine. Use a watt meter to determine the total power consumption of your machine to determine power usage, do *not* trust online references that say your GPU uses so many watts. When your GPU is mining other parts of the computer are also drawing power, especially the CPU which may also be mining. You want to know the total system power and make sure your mining proceeds exceed that amount. Be sure to use above baseline residential power rates in your calculations, do *not* just look at your current bill and expect the current rate. If it is not profitable to mine do *not* fall into the trap that "the coin price will eventually rise and make it profitable", that is a losing game. Instead, take whatever money you would spend on power and just buy the coins directly, you will have more coins that way if the price rises. But above all else, do not get into the mindset of joining the mining scene, that is a path to losing money. Stay in the scene "the GPU I have anyway can make some coin when I'm not using it".
(2) Take whatever money you were willing to spend on a GPU for mining and just buy coins with that money. You will likely do better that way if the price rises. Many miners fall into the trap that they are profitable and pat themselves on the back. They do not consider the opportunity cost of the alternative of just buying coins directly. The following are very rough estimates but the point will nonetheless be clear. Lets say you spent $500 on a GPU last summer and another $500 on a GPU last fall. At above baseline residential power rates maybe you have about an extra $1,000 after factoring in power. Congrats your GPUs are now paid. However your friend bought $500 worth of bitcoin in the summer and another $500 in the fall and now has $3,000 worth of bitcoin. You are at net $0, he is at net $2,0000. If you are willing to gamble on increasing coin prices you may be better off just buying coins directly. Things are not as simple as a mining rig being profitable, the opportunity cost of the just buy directly must be considered. Many other things must also be considered before joining the mining scene.
When a currency can quickly gain or lose half it's value with no apparent reason and no apparent cause, many people won't want to take them for fear they will lose money on the deal.
There is a cause, wall street speculation. Bitcoin's price rose 65x as wall street speculators were joining the pre-existing home based speculators.
Regarding BSD... At the computer swap meet in 1993 I picked up two dirt cheap CDs, Yggdrasil and FreeBSD. Given a BSD background from university days I tried FreeBSD first, it crashed during install. Then I tried Yggdrasil, to borrow from Apple, it "just worked". And so the Linux vs BSD decision was made for me with respect to PC desktop *nix. Today its back to BSD via macOS mostly, a little Linux via VMs or old headless PCs in the closet.
Linux Now Has its First Open Source RISC-V Processor, Slashdot.
To answer AC's question a few moths later: "What's the big advantage with RISC over ARM or x86?"
Meltdown, Spetre.
ARM is a RISC architecture, and plenty of RISC architectures suffer from Spectre
I think the GP was using RISC as an abbreviation for the RISC-V brandname, not RISC as in the architecture design approach. Indicating an advantage of RISC-V CPUs over ARM or x86 CPUs.
Slashdot Mirror
Yes profitability made huge jumps in the last couple of months (which may be a signal to wait not buy GPUs, see below), but you can't extrapolate today's anomalous returns without taking on huge risks. Risks most people contemplating the bandwagon do not understand. When you tell newbs their cards will be paid off in a few months you are essentially telling them difficulty will not increase and bitcoin prices in the mid ten thousands (and the accompanying alt coin spikes and enthusiastic bidding on nicehash, etc) is the new normal, or that prices will keep rising.
Also did you put a watt meter on the power outlet to your rig to measure the actual power being consumed? Don't rely on online claims of card X uses Y watts.
Are you using non-baseline pricing for your power costs? Locally baseline is $0.16 and newbs use that in online calculators to get to their paid in a few months type of numbers and then are surprised when they are actually billed at $0.25 when the GPUs blow past baseline usage. Fortunately for them Time of Use rate plans are coming and if they select the right one, not the default one, they can get around $0.14 if they shut down for the 6 hours of high demand pricing or $0.18 if they run 24/7. Again, newbs need to do a lot of research before joining the scene.
OK, so you are arguing that recent price spikes have let you pay off all your cards. Congrats. Now go back and on the day you bought each card figure out how many bitcoins you could have purchased with that money on that day. Do that for each card. Add up all those hypothetical bitcoins. What is the current value of those coins. Its likely more than what you made mining. You could buy all those cards today with those bitcoin proceeds and probably have lots of cash left over. This is the point I was trying to make in scenario (2) in the original post. Rising prices is not a signal to buy a miner, rising prices is a signal to buy coins and delay purchasing a miner. Unless of course you need the GPU for non-mining purposes.
If you don't do this alternative use of the money calculation you can't really say you came out ahead on mining. Newbs need to understand this sort of thing before joining the scene.
Bitcoin has fewer flaws than any crypto currency. There is one fundamental "flaw" most people complain about ...
Here's the one fundamental flaw that gets little attention: mining is not decentralized.
;-) Blockchain technology is like internal combustion engine technology. Bitcoin is like the Ford Model T, an early user of the technology, the first to gain wide acceptance and mindshare, but entirely replaceable by the inevitable competitor with newer and better technology.
Bitcoin has deviated from its design and its security is compromised. The security of the bitcoin blockchain is based on the assumption of a globally distributed network of ordinary users using their computers to maintain the blockchain (mining). There are two deviations. First, mining is dominated by a small group of individuals/organizations with expensive specialized hardware (ASICs). This makes a 51% attack plausible should a cartel form, note that a mining pool reached 50% a few years ago. Secondly, 70% of miners are located in a single country and dependent upon government supplied inexpensive electricity. This makes government interference or manipulation plausible. With distributed mining cartel and government interference was not plausible, but with the current mining situation both are plausible.
This is fixable, it is software after all. However the necessary software updates (ASIC resistant mining algorithms, updated as necessary - already in use by other coins; maybe a move from PoW to PoS - we'll see how that goes with etherium; etc) have to overcome inertia, politics and greed - the voting on such changes will be done by the miners who like the current system and have expensive ASiC hardware that need to be paid for. Users can only abandon bitcoin and move to another coin.
Bitcoin may be replaced in the short/medium term, I don't know about Buffets 5 year timeframe but in general he is likely correct. The future is blockchain technology, any individual coin is just a user of that technology and entire replaceable.
What we need is a car analogy.
Now some of you might be tempted to mention network effect. Network effect requires a high switching cost. Users have little switching cost moving from one coin to another. Miners have a high switching cost and that is why they will resist algorithm changes, but users are quite free to move on to something new.
It's 2AM And do you know where your BTC is?
Yes, the blockchain tells me. ;-)
It seems that these companies (Microsoft and Ubuntu and others) are forgetting everything about sound software development practices here. They're in such a hurry to deploy patches that they aren't taking the time to fully test them. The cure is worse than the ailment.
Both Microsoft and Ubuntu are plagued by the vast permutations of hardware out there, all the combinations of motherboard, cpu, video, etc. Aren't there identified problems with various anti-virus software? Did some driver developer out there try something tricky too that is incompatible with the fix(es)? Historically various problems with Windows came from 3rd party drivers not necessarily Microsoft itself, perhaps Ubuntu is having similar problems?
That's a very dismal perspective ;-)
Mining will inherently end up centralised wherever power is available the cheapest...
Not with ASIC resistant algorithms and periodic updates to the algorithms. Its already done elsewhere.
12345? That's the same combination as my luggage!
Per TSA regulations :-)
... using one of the many predictable default hostnames ...
Good thing I renamed mine to "FutureCorruptedBackup" ;-)
https://www.nicehash.com/profitability-calculator/nvidia-gtx-1080-ti?e=0.15¤cy=USD
That is today's profitability. You can't project that into the future. More realistic profitability calculators factor in difficulty increases.
Also power costs seem off. $0.15 looks more like baseline residential pricing. GPU mining will likely blow way past baseline and into the next higher tier of pricing. As I mentioned earlier you need to make sure comparisons are using beyond baseline pricing.
If coin prices decline, difficulty decreases and if coin prices rise, difficulty increases. I think they balance themselves out.
No. More miners (more hashing power technically) more difficulty. Less miners less difficulty. More people joining the mining scene increases difficulty regardless of what price is doing.
Plus, there are automated pool that will mine the most profitable coin and auto-exchange it to your preferred coin.
Not a "plus", that is what your profitability link above assumes, constant switching to the more profitable algorithms as alt-coin enthusiasts bid for your hashing power.
Who know what will happen. From the past it looks like 3 month ROI is plausible. It has been better before and it has been worse but just an average of 3 months.
The past is what I described in scenario (2) earlier. That was actual data since the summer, mildly rounded for convenience. Those nicehash estimates do not project out as you assume. Plus they understate costs since they only factor the GPU power requirements. As I mentioned earlier you have to use a watt meter and measure what the entire computer is consuming when it is hashing.
I totally agree with you that investing has been far more profitable than mining. If you just want to get a sweet GPU or system paid for my mining in about 3 months without a lot of work, it looks to be possible. Mining rigs or cypto investing is almost a part time job since you have to keep on top of so many things and maybe not the best use of your time. Even if pays for itself in 6 months, it's still a good deal. Most stores will give you a 0% APR for 6 month loan for buying over $500.
If you want the GPU for non-mining purposes, sure, mine to subsidize the purchase. Why leave the system off/sleeping if mining is profitable? Maybe go up a model and get a slightly more expensive card, scenario (1) earlier. I'm just arguing that if the only reason for the GPU is mining things need to be viewed differently, a delay to purchase possibly warranted depending on what coin price expectations are.
They are different kinds of investments. Of course you can make more overall by day trading coins. That does not remove the profitability of the mining that allows those coins to function in the first place.
As I said, if you are buying a GPU for non-mining reasons you might as well mine *iff* profitable. But if you are buying a GPU only to make money mining then just buying coins needs to be an alternative you consider.
We can incorporate the dense mining rig into the day trading you're talking about, pointing it at newer lower difficulty coins for the long term.
As I mentioned elsewhere, delay that until coin prices are no longer increases and you may be better off. In scenario 2 above by delaying the two purchases by 3 and 6 months respectively you still get the GPUs for the long term but you also have an additional $2K.
The cryptocurrency technology is elegantly constructed to sustain itself. It's counter-intuitive, or just plain wrong, to take the profitability of both sides separately so hard.
Now we have a topic change, not mining for profit but mining to altruistically support the blockchain network. I was only arguing over the profit motive.
They can still buy and use cheap ASICs, just have to ship them outside China.
ASICs have a short lifespan and no salvage value beyond scrap metal pricing for their heat sinks. Every day in transit and setting up at a new location means an ASIC is likely losing its most profitable current mining time, difficulty increases reduce profitability.
The Chinese miners will just set up shop elsewhere. Maybe they won't move at all since local authorities in China benefit greatly when underutilized power capacity is used by bitcoin.
In theory, but its going to hurt to see their ASiCs lose their more profitable days, perhaps even no longer be profitable in the timeframe necessary to relocate.
If you run the real numbers, it's profitable.
Case (2) above is based on real numbers, modestly rounded for convenience. Reality is 3x the earning just buying coins directly 6 months ago compared to buying gpus last summer and last fall. Delaying those purchases 3 to 6 months is the difference between paid for $1,000 GPUs (mining) and paid for $1,000 GPUs plus an additional $2,000 in cash (buying coins directly).
Sorry, I described the scenario incorrectly. 3x is the result of buying coins rather than a GPU 6 months ago and again 3 months ago. I erroneously implied by all the coins six months ago, that would result in 4.5x but that is an unfair comparison since money is being spent at different times. Spending money at the same time is the fair comparison and that is a 3x result.
What are you blabbering on about? It's cheaper to have heatsinks made in China and shipped to most places in the world. Why would you think it's cheaper than shipping alone to locally manufacture them?
I was actually thinking source them locally, not necessarily manufacture them locally, thoughts and typing diverged.
Fill up a container, stick it on a boat. You'll pay a flat rate for the container.
And while in ocean transit the ASIC experiences a drop in profitability. Also I am are talking about individuals getting their hosted in China ASICs delivered to them. As for farms, containers may be more viable but again transit time and setting up a new farming site eats in the more profitable days those ASIC would have had.
If you run the real numbers, it's profitable.
Case (2) above is based on real numbers, modestly rounded for convenience. Reality is 3x the earning just buying coins directly 6 months ago compared to buying gpus last summer and last fall. Delaying those purchases 3 to 6 months is the difference between paid for $1,000 GPUs (mining) and paid for $1,000 GPUs plus an additional $2,000 in cash (buying coins directly).
Paying some of the highest local market rates, my normal workstation with an outdated video card pulls in real actual profit by the real actual honest numbers.
As I said, if you are buying a GPU for non-mining purposes then mining to subsize the GPU is fine. Might even justify getting a slightly better GPU. But if the GPUs are being bought solely for mining then things radically change.
Can you guys start shutting up if you don't want to put the work in?
Work includes investigating the alternatives, as I did. Its not as simple as a rig being profitable, you need to calculate the alternative uses of that money, model different scenarios (coin prices, difficulty increase rates, etc).
Buy a dual 1080 Ti. Stick it your computer and it will pay for itself in 3 months. You can also game on it once in a while.
*Iff* coin prices do not decline.
**Iff** difficulty levels do not increase.
Normal difficulty increases will likely stretch that 3 months to 6 months. 3 months is an extrapolation of today's revenue, poor planning to bank on that.
Now if you want to gamble on increasing coin prices, you can take the money for those two 1080 Ti's and invest that up front by buying the coins directly at an expected relative low. You may be tempted to make an "averaging in" argument, but we aren't talking about consistent payroll deductions into a 401K here. Every two weeks the earning power of those 1080 Ti's will likely decline, your "contributions" will decline. If you are expecting the sort of volatility that "averaging in" helps protect against you are also expecting to stretch out the timeframe those 1080 Ti's need to reach profitability, if ever.
Again, *iff* you are gambling on increasing coin prices you may make more by just buying coins up front and then cashing out after a few months. Then buy a 1080 Ti for gaming if you really want one and if you think coins prices have stabilized or will decline. If you expect further increases its probably best to delay the 1080 Ti a little longer. Read case (2) above, its based on actual mining revenue and coin price of the last six months, numbers slightly rounded for convenience. 3x the revenue for delaying the GPU purchases.
Just to check, that 3 month projection of yours is subtracting out power at above baseline residential pricing?
Seriously, your 3 month claim to break-even makes me skeptical, if you had said 6 months that would have been more plausible. Sit down with a spreadsheet and model different scenarios. People who do so often find things are quite different than more casual analysis predicts.
The free market guarantees meritocracy, right? right??? RIGHT!?!?
Adam Smith, "An Inquiry into the Nature and Causes of the Wealth of Nations" - 1776, did not predict joy on every single day, he predicted better outcomes in the long run than otherwise available.
So yes, capitalism allows us to switch to macOS, Linux, FreeBSD, etc to reward those and/or punish Microsoft. That is the mechanism by which a meritocracy is determined.
Your Bug Report Status: Works as expected.
If you want to get in on the cryptocurrency mining scene ...
Stop yourself, don't do it. Take either of these two paths:
(1) You were going to buy a GPU anyway for some non-mining reason. Go ahead, buy the GPU. Maybe, **maybe**, buy a model up one level of performance/price from what you would have otherwise bought, if its a low to midrange model. Say if you were otherwise planning on a GTX 1050 Ti 4GB ***maybe*** get a GTX 1060 6GB, ****maybe****. If you were otherwise getting a high performer, say a 1070 Ti for that 4K monitor, do *****not***** go up a level to a 1080 Ti. Then let the GPU mine when you are not using the machine. Use a watt meter to determine the total power consumption of your machine to determine power usage, do *not* trust online references that say your GPU uses so many watts. When your GPU is mining other parts of the computer are also drawing power, especially the CPU which may also be mining. You want to know the total system power and make sure your mining proceeds exceed that amount. Be sure to use above baseline residential power rates in your calculations, do *not* just look at your current bill and expect the current rate. If it is not profitable to mine do *not* fall into the trap that "the coin price will eventually rise and make it profitable", that is a losing game. Instead, take whatever money you would spend on power and just buy the coins directly, you will have more coins that way if the price rises. But above all else, do not get into the mindset of joining the mining scene, that is a path to losing money. Stay in the scene "the GPU I have anyway can make some coin when I'm not using it".
(2) Take whatever money you were willing to spend on a GPU for mining and just buy coins with that money. You will likely do better that way if the price rises. Many miners fall into the trap that they are profitable and pat themselves on the back. They do not consider the opportunity cost of the alternative of just buying coins directly. The following are very rough estimates but the point will nonetheless be clear. Lets say you spent $500 on a GPU last summer and another $500 on a GPU last fall. At above baseline residential power rates maybe you have about an extra $1,000 after factoring in power. Congrats your GPUs are now paid. However your friend bought $500 worth of bitcoin in the summer and another $500 in the fall and now has $3,000 worth of bitcoin. You are at net $0, he is at net $2,0000. If you are willing to gamble on increasing coin prices you may be better off just buying coins directly. Things are not as simple as a mining rig being profitable, the opportunity cost of the just buy directly must be considered. Many other things must also be considered before joining the mining scene.
For a while you couldn't turn a profit with a GPU. With insane bitcoin prices you can earn (I'm told) about $250/month after electric cost with a 1080ti. Which is why the price of 1080s has gone up.
But you are not mining bitcoin. If you are receiving bitcoins from GPU mining you are likely selling your hashing power to someone doing alt-coin mining and who is paying for this hashing in bitcoin. Because who wants whatever goofy alt-coin they want to mine. :-)
Find me some mining hardware that will make enough money to make back my investment before its obsolete. It was only profitable in China because the government paid for everything.
Especially after you pay to have that heavy ASIC device shipped from China to wherever your are. Maybe you can pay to have heat sinks removed and have domestically manufacture heatsinks attached after the ASIC blades arrives. Either way, hurts you profitability and makes ASiC obsolescence that much closer. :-)
The biggest, and maybe only problem that will arise will be the difficulty being too high when they shut down the farms, leading to a period of slow block solves until the difficulty adjusts. The network will work just fine without the chinese mining farms
And the blockchain network will be more secure. Bitcoin has deviated from its original design in two ways and both compromise blockchain security. First, miners are no longer a diverse group of ordinary users and their personal computers, mining is dominated by a relatively small number of ASIC farms. This makes 51% attacks more plausible, we had one pool get to 50% a few years ago. Secondly, ASIC mining is concentrated in a single country, 70% of the hashate give or take. This obviously destroys the notion that bitcoin is beyond government meddling. These ASIC miners are dependent upon cheap government controlled power.
It sucks to have invested money in ASIC hardware and colocated you gear there but this move would help to get bitcoin back on track. Hopefully closer to the globally and widely distributed mining that is necessary for blockchain security, something we do not have today.
If you want to get in on the cryptocurrency mining scene ...
Stop yourself, don't do it. Take either of these two paths:
(1) You were going to buy a GPU anyway for some non-mining reason. Go ahead, buy the GPU. Maybe, **maybe**, buy a model up one level of performance/price from what you would have otherwise bought, if its a low to midrange model. Say if you were otherwise planning on a GTX 1050 Ti 4GB ***maybe*** get a GTX 1060 6GB, ****maybe****. If you were otherwise getting a high performer, say a 1070 Ti for that 4K monitor, do *****not***** go up a level to a 1080 Ti. Then let the GPU mine when you are not using the machine. Use a watt meter to determine the total power consumption of your machine to determine power usage, do *not* trust online references that say your GPU uses so many watts. When your GPU is mining other parts of the computer are also drawing power, especially the CPU which may also be mining. You want to know the total system power and make sure your mining proceeds exceed that amount. Be sure to use above baseline residential power rates in your calculations, do *not* just look at your current bill and expect the current rate. If it is not profitable to mine do *not* fall into the trap that "the coin price will eventually rise and make it profitable", that is a losing game. Instead, take whatever money you would spend on power and just buy the coins directly, you will have more coins that way if the price rises. But above all else, do not get into the mindset of joining the mining scene, that is a path to losing money. Stay in the scene "the GPU I have anyway can make some coin when I'm not using it".
(2) Take whatever money you were willing to spend on a GPU for mining and just buy coins with that money. You will likely do better that way if the price rises. Many miners fall into the trap that they are profitable and pat themselves on the back. They do not consider the opportunity cost of the alternative of just buying coins directly. The following are very rough estimates but the point will nonetheless be clear. Lets say you spent $500 on a GPU last summer and another $500 on a GPU last fall. At above baseline residential power rates maybe you have about an extra $1,000 after factoring in power. Congrats your GPUs are now paid. However your friend bought $500 worth of bitcoin in the summer and another $500 in the fall and now has $3,000 worth of bitcoin. You are at net $0, he is at net $2,0000. If you are willing to gamble on increasing coin prices you may be better off just buying coins directly. Things are not as simple as a mining rig being profitable, the opportunity cost of the just buy directly must be considered. Many other things must also be considered before joining the mining scene.
When a currency can quickly gain or lose half it's value with no apparent reason and no apparent cause, many people won't want to take them for fear they will lose money on the deal.
There is a cause, wall street speculation. Bitcoin's price rose 65x as wall street speculators were joining the pre-existing home based speculators.
Regarding BSD ... At the computer swap meet in 1993 I picked up two dirt cheap CDs, Yggdrasil and FreeBSD. Given a BSD background from university days I tried FreeBSD first, it crashed during install. Then I tried Yggdrasil, to borrow from Apple, it "just worked". And so the Linux vs BSD decision was made for me with respect to PC desktop *nix. Today its back to BSD via macOS mostly, a little Linux via VMs or old headless PCs in the closet.
Linux Now Has its First Open Source RISC-V Processor, Slashdot. To answer AC's question a few moths later: "What's the big advantage with RISC over ARM or x86?" Meltdown, Spetre.
ARM is a RISC architecture, and plenty of RISC architectures suffer from Spectre
I think the GP was using RISC as an abbreviation for the RISC-V brandname, not RISC as in the architecture design approach. Indicating an advantage of RISC-V CPUs over ARM or x86 CPUs.