Annoying Fallacies

[mootinator]

I keep reading people posting the same ridiculous assumptions over and over in this forum, i.e.:

• People who have mining operations less than X ?H/s are/are going to be irrelevant!

Your hashrate has absolutely nothing to do with whether you are irrelevant or not. If 1GH/s for $100 at 7W is unprofitable, so is 1TH/s for $10,000 at 7000W. Sure there's a nice psychological effect behind getting 1000 times more bitcoin, but you're also losing 1000 times more money on the proposition if you can't get a return on investment. Which leads me into the next "shit bitcoin miners say"...

• Soon only huge mining cartels will exist!

If you're losing money because power, the big mining cartel is losing close to 1000x more. Thus they have more incentive to unplug equipment. There aren't (yet) any chips significantly more power efficient available in high end devices than in low end ones. This does not compute.

Now bring on the hobby market 21nm stuff so I can enjoy greater psychological benefit while overall losing BTC strengthening the network.</rant>

[mgio]

There are tons of fallacies regarding mining.

One of my favorites is that it matters whether you paid for your device with $ or BTC. It doesn't matter at all. If you paid with $, that's money you didn't use buying BTC. It terms of profit, it doesn't matter how you paid. You should look at cost in BTC when you bought and BTC returned by the machine to figure out your profit.

Another great fallacy is that if you have cheap electricity you will eventually make back the money spent to purchase a mining device because it will always be profitable. It's true that it will always make money if you have free electricity but people fail to realize that there is a point where the machine just isn't worth running as it isn't even making a penny a day. And no matter how many portions of a penny you add up, it will never become an amount larger enough to pay off a device.

[Cluster2k]

One of my favorites is that it matters whether you paid for your device with $ or BTC. It doesn't matter at all. If you paid with $, that's money you didn't use buying BTC. It terms of profit, it doesn't matter how you paid. You should look at cost in BTC when you bought and BTC returned by the machine to figure out your profit.

It's not a fallacy.  All of my costs, from mortgage, food, clothing, electricity, taxes, gas, mining hardware, etc are in dollars and not bitcoins.  I have never purchased anything in bitcoins and probably never will.  Buying ASIC hardware using dollars (with protection from PayPal/CC) made much more sense than using bitcoins.  The key question is this: would I have used the dollars I spent on ASIC hardware to buy bitcoins instead?  No, I never considered it.  The bitcoins I generate get turned into dollars to do something useful in my life.  ROI achieved a long time ago.  Some are kept just in case this highly speculative investment ever reaches the point where one bitcoin will pay off my entire mortgage.

Another way of debunking the fallacy is to look at the shares of gold miners.  The shares and dividends are priced in dollars and not ounces of gold.  The annual profits are not announced in gold either.  I have yet to meet an investor who bought shares in a miner and then started comparing spending the same money on gold and silver.  One is buying the commodity, the other is buying the means to generate the commodity.  The two are not the same.  This also appplies to bitcoin.

[bcp19]

One of my favorites is that it matters whether you paid for your device with $ or BTC. It doesn't matter at all. If you paid with $, that's money you didn't use buying BTC. It terms of profit, it doesn't matter how you paid. You should look at cost in BTC when you bought and BTC returned by the machine to figure out your profit.

It's not a fallacy.  All of my costs, from mortgage, food, clothing, electricity, taxes, gas, mining hardware, etc are in dollars and not bitcoins.  I have never purchased anything in bitcoins and probably never will.  Buying ASIC hardware using dollars (with protection from PayPal/CC) made much more sense than using bitcoins.  The key question is this: would I have used the dollars I spent on ASIC hardware to buy bitcoins instead?  No, I never considered it.  The bitcoins I generate get turned into dollars to do something useful in my life.  ROI achieved a long time ago.  Some are kept just in case this highly speculative investment ever reaches the point where one bitcoin will pay off my entire mortgage.

Another way of debunking the fallacy is to look at the shares of gold miners.  The shares and dividends are priced in dollars and not ounces of gold.  The annual profits are not announced in gold either.  I have yet to meet an investor who bought shares in a miner and then started comparing spending the same money on gold and silver.  One is buying the commodity, the other is buying the means to generate the commodity.  The two are not the same.  This also appplies to bitcoin.

I could not agree more.  The only reason I have any bitcoins at all is because I mined them.  I don't care that buying bitcoins would have been a better investment, I invested $ to make $ and I succeeded.  I've mined less than 3 bitcoins (including alts mined and sold for bitcoin) yet my net worth today is over 10 bitcoins.  I invested some wisely and some poorly, but I did better in the long run than most.  Sure, I could have had 25+ if I had been more daring in my investments, but I could just as easily have had 1.

[mootinator]

The key question is this: would I have used the dollars I spent on ASIC hardware to buy bitcoins instead?  No, I never considered it.

I do somewhat agree with this point. I ordered my (ex) BFL Jalapeno without knowing very much about BTC. It took me several small transactions to warm up to the idea of sending some random guy and/or shady looking company on the internet money for BTC before I was comfortable with the idea.

(Though in retrospect I probably should have reserved my skepticism for BFL. At least everyone selling BTC had an actual product in hand.)

There really isn't a good reason not to buy BTC from a trusted exchange, it definitely is tempting for new people to want to purchase a physical thing, even if they're probably just throwing money to the wind.

[mgio]

One of my favorites is that it matters whether you paid for your device with $ or BTC. It doesn't matter at all. If you paid with $, that's money you didn't use buying BTC. It terms of profit, it doesn't matter how you paid. You should look at cost in BTC when you bought and BTC returned by the machine to figure out your profit.

It's not a fallacy.  All of my costs, from mortgage, food, clothing, electricity, taxes, gas, mining hardware, etc are in dollars and not bitcoins.  I have never purchased anything in bitcoins and probably never will.  Buying ASIC hardware using dollars (with protection from PayPal/CC) made much more sense than using bitcoins.  The key question is this: would I have used the dollars I spent on ASIC hardware to buy bitcoins instead?  No, I never considered it.  The bitcoins I generate get turned into dollars to do something useful in my life.  ROI achieved a long time ago.  Some are kept just in case this highly speculative investment ever reaches the point where one bitcoin will pay off my entire mortgage.

Another way of debunking the fallacy is to look at the shares of gold miners.  The shares and dividends are priced in dollars and not ounces of gold.  The annual profits are not announced in gold either.  I have yet to meet an investor who bought shares in a miner and then started comparing spending the same money on gold and silver.  One is buying the commodity, the other is buying the means to generate the commodity.  The two are not the same.  This also appplies to bitcoin.

I could not agree more.  The only reason I have any bitcoins at all is because I mined them.  I don't care that buying bitcoins would have been a better investment, I invested $ to make $ and I succeeded.  I've mined less than 3 bitcoins (including alts mined and sold for bitcoin) yet my net worth today is over 10 bitcoins.  I invested some wisely and some poorly, but I did better in the long run than most.  Sure, I could have had 25+ if I had been more daring in my investments, but I could just as easily have had 1.

It's a fallacy. It doesn't matter whether you paid in bitcoins or dollars. It only matters the value of what you spend and the value of what you get back. And you need
You say: "The key question is this: would I have used the dollars I spent on ASIC hardware to buy bitcoins instead?  No, I never considered it." Why wouldn't have not bought bitcoins instead? Why? It would have been a much better investment. If you are trying to argue that mining is more "fun", fine, but you can't argue that buying BFL hardware in 2012 when bitcoin was only $10 was a better decision than simply buying bitcoins at that point. If you had, you'd end up with more bitcoins right now and thus more dollars. People somehow feel that if you paid in dollars back when you ordered it was a good investment but if you paid in bitcoins it was a poor one. The fact is, it doesn't matter at all.

The difference with gold mining shares is that they are never purchased with gold. They are purchased with dollars. And so you never hear someone say something idiotic like, "good thing you bought those gold mining shares with dollars instead of gold bars!", which is something you hear all the time here (except with BTC replaced for gold).

[Emmie]

OP, this is all assuming small miners and large mining farms use the same equipment.

[Bitweasil]

If you're losing money because power, the big mining cartel is losing close to 1000x more. Thus they have more incentive to unplug equipment. There aren't (yet) any chips significantly more power efficient available in high end devices than in low end ones. This does not compute.

You're wrong, in that you're not considering the surrounding hardware and power costs.

First, location matters.  There are places with much cheaper power than others.  Home miners typically mine where they are - they don't move or rent spaces with cheaper power to mine (and I'd argue if you are renting space specifically to mine, you quality as an industrial miner).

So, industrial miners will be mining in an area with cheaper power.

Second, depending on the scale, they may be able to get industrial rates, in which they guarantee a certain load, and are interruptible loads if needed by the power company.  This reduces rates rather significantly in exchange for a bit of downtime per year.  Bitcoin mining is entirely interruptible - there is no penalty for stopping, other than the time you're not mining.  If your total profits are higher with this system, it's worth doing.

Third, doing power conversion on a large scale (100s of KW or MW scale) can be done more efficiently than your small at home power supply.

Fourth, if you are doing industrial mining, you can use much more efficient cooling systems than typically are used at home.  Look at what some of the big data centers are doing to get ideas here, but you can use cooling towers, air exchange cooling, etc.  And, depending on the hardware, you can run it hot.  It is often more efficient to run hardware up at 100+F and pay less in cooling costs, even calculating in hardware failure rates/reduced hashing rates.  When you're playing with megawatts, this matters.  Supposedly, it's actually cheapest to run data centers up at a temperature where humans cannot enter them to maintain the hardware because it's too damned hot.

Fifth, if you're big enough, you can just go about getting chips designed and fabricated for your specific use case.  I don't know if this is being done or not, but there is certainly no reason that someone who is doing industrial mining is required to use the same core parts that other people are using at home.

Your assertion that big miners are running the same hardware, with the same efficiency, as home miners, is just not taking into account the benefits of scale you can get.

http://www.google.com/about/datacenters/efficiency/internal/ has some interesting resources on data center efficiency as well - industrial mining can be data center scale.

[mootinator]

You're wrong, in that you're not considering the surrounding hardware and power costs.

First, location matters.  There are places with much cheaper power than others.  Home miners typically mine where they are - they don't move or rent spaces with cheaper power to mine (and I'd argue if you are renting space specifically to mine, you quality as an industrial miner).

So, industrial miners will be mining in an area with cheaper power...

Regardless of what the big datacenters are doing, the cheapest power is "free" power. An industrial miner will never get free power. Anyone living in a small apartment or suite where someone else is responsible for the power bill can ignore power costs entirely, they're only limited by the load the existing wiring will handle and the tolerance of whoever actually pays for the power. No reason to think small scale mining will ever become 'irrelevant'.

Fifth, if you're big enough, you can just go about getting chips designed and fabricated for your specific use case.  I don't know if this is being done or not, but there is certainly no reason that someone who is doing industrial mining is required to use the same core parts that other people are using at home.

Your assertion that big miners are running the same hardware, with the same efficiency, as home miners, is just not taking into account the benefits of scale you can get.

http://www.google.com/about/datacenters/efficiency/internal/ has some interesting resources on data center efficiency as well - industrial mining can be data center scale.

No, there certainly is no reason that someone doing industrial mining is required to use the same parts that others are using at home. However, as far as I know every company who has gone to the expense of having an ASIC designed and manufactured has realized the obvious fact there's far less risk selling their hardware than trying to make a buck on it mining. There's certainly no rational reason for any company ever to decide otherwise.

[Emmie]

This is also assuming that the total ROI is 2x or lower for every mining machine.

[mootinator]

This is also assuming that the total ROI is 2x or lower for every mining machine.

You're going to have to explain that one.

[bcp19]

One of my favorites is that it matters whether you paid for your device with $ or BTC. It doesn't matter at all. If you paid with $, that's money you didn't use buying BTC. It terms of profit, it doesn't matter how you paid. You should look at cost in BTC when you bought and BTC returned by the machine to figure out your profit.

It's not a fallacy.  All of my costs, from mortgage, food, clothing, electricity, taxes, gas, mining hardware, etc are in dollars and not bitcoins.  I have never purchased anything in bitcoins and probably never will.  Buying ASIC hardware using dollars (with protection from PayPal/CC) made much more sense than using bitcoins.  The key question is this: would I have used the dollars I spent on ASIC hardware to buy bitcoins instead?  No, I never considered it.  The bitcoins I generate get turned into dollars to do something useful in my life.  ROI achieved a long time ago.  Some are kept just in case this highly speculative investment ever reaches the point where one bitcoin will pay off my entire mortgage.

Another way of debunking the fallacy is to look at the shares of gold miners.  The shares and dividends are priced in dollars and not ounces of gold.  The annual profits are not announced in gold either.  I have yet to meet an investor who bought shares in a miner and then started comparing spending the same money on gold and silver.  One is buying the commodity, the other is buying the means to generate the commodity.  The two are not the same.  This also appplies to bitcoin.

I could not agree more.  The only reason I have any bitcoins at all is because I mined them.  I don't care that buying bitcoins would have been a better investment, I invested $ to make $ and I succeeded.  I've mined less than 3 bitcoins (including alts mined and sold for bitcoin) yet my net worth today is over 10 bitcoins.  I invested some wisely and some poorly, but I did better in the long run than most.  Sure, I could have had 25+ if I had been more daring in my investments, but I could just as easily have had 1.

It's a fallacy. It doesn't matter whether you paid in bitcoins or dollars. It only matters the value of what you spend and the value of what you get back. And you need
You say: "The key question is this: would I have used the dollars I spent on ASIC hardware to buy bitcoins instead?  No, I never considered it." Why wouldn't have not bought bitcoins instead? Why? It would have been a much better investment. If you are trying to argue that mining is more "fun", fine, but you can't argue that buying BFL hardware in 2012 when bitcoin was only $10 was a better decision than simply buying bitcoins at that point. If you had, you'd end up with more bitcoins right now and thus more dollars. People somehow feel that if you paid in dollars back when you ordered it was a good investment but if you paid in bitcoins it was a poor one. The fact is, it doesn't matter at all.

The difference with gold mining shares is that they are never purchased with gold. They are purchased with dollars. And so you never hear someone say something idiotic like, "good thing you bought those gold mining shares with dollars instead of gold bars!", which is something you hear all the time here (except with BTC replaced for gold).

I did not know about Bitcoin in 2012, so your point is moot.

[Emmie]

This is also assuming that the total ROI is 2x or lower for every mining machine.

You're going to have to explain that one.

If ROI is higher than 2x the cost of the mining hardware, bigger farms will have an exponentially linearly higher win than smaller ones.

[mootinator]

This is also assuming that the total ROI is 2x or lower for every mining machine.

You're going to have to explain that one.

If ROI is higher than 2x the cost of the mining hardware, bigger farms will have an exponentially higher win than smaller ones.

No. That isn't how math works.

If each dollar spent returns 3x the cost of the hardware for big farms they're earning 3c (Where c is the investment). That's still a linear relationship.

Even if small miners are getting x=1.5c, the big miners are only getting 2x if their return is 3c. If bigger farms were having an exponentially higher win than small miners they'd have to be making something like x²

[Rannasha]

This is also assuming that the total ROI is 2x or lower for every mining machine.

You're going to have to explain that one.

If ROI is higher than 2x the cost of the mining hardware, bigger farms will have an exponentially higher win than smaller ones.

No. That isn't how math works.

If each dollar spent returns 3x the cost of the hardware for big farms they're earning 3c (Where c is the investment). That's still a linear relationship.

Even if small miners are getting x=1.5c, the big miners are only getting 2x if their return is 3c. If bigger farms were having an exponentially higher win than small miners they'd have to be making something like x²

That would a quadratic relationship. Exponential would be 2^x

[Emmie]

This is also assuming that the total ROI is 2x or lower for every mining machine.

You're going to have to explain that one.

If ROI is higher than 2x the cost of the mining hardware, bigger farms will have an exponentially higher win than smaller ones.

No. That isn't how math works.

If each dollar spent returns 3x the cost of the hardware for big farms they're earning 3c (Where c is the investment). That's still a linear relationship.

Even if small miners are getting x=1.5c, the big miners are only getting 2x if their return is 3c. If bigger farms were having an exponentially higher win than small miners they'd have to be making something like x²

I'm thinking like, invest a thousand, gain two thousand. Invest a million, gain two million. The more money you have to play with the more you can earn. Just like buying and selling stock.

[mootinator]

I'm thinking like, invest a thousand, gain two thousand. Invest a million, gain two million. The more money you have to play with the more you can earn. Just like buying and selling stock.

Of course. If that weren't the case nobody would have any theoretical incentive at all to build a big mining datacenter.

[xuemike]

• People who have mining operations less than X ?H/s are/are going to be irrelevant!

Your hashrate has absolutely nothing to do with whether you are irrelevant or not. If 1GH/s for $100 at 7W is unprofitable, so is 1TH/s for $10,000 at 7000W. Sure there's a nice psychological effect behind getting 1000 times more bitcoin, but you're also losing 1000 times more money on the proposition if you can't get a return on investment. Which leads me into the next "shit bitcoin miners say"...

• Soon only huge mining cartels will exist!

There are generally advantages in having bigger farms vs smaller farms.  The overhead cost becomes less for a bigger farm / GH/S.  Reason being. Cooling becomes more streamlined, the usage of space is now more efficient (ie having all your farms in 1 place), cost for computers as host etc all contribute to over head cost of mining.