Topic: Law of unintended consequence. (Read 2184 times)

Yeah, I've been spending my time getting a miner together and working. Even though there is a learning curve to this, it is a worthwhile pursuit indeed. I still have much to learn, but at least I will earn.

cbeast still has not responded to my answer.  I'm rather disappointed, I was looking forward to a rousing debate.  I guess I scared him off with my superior debate skills. 

The irony of this statement is only exceeded by the wild ass guess that followed.

You're making a wild-ass guess there.

Without making a wild-ass guess, assuming that 5/6 of the average hashrate over the previous difficulty computation period drops out in the next period:

The number of blocks a miner can expect to solve in a given time period is independent of the current network hashrate (being a function strictly of the difficulty and the miner's hashrate: if difficulty means that 2.8*10^15 hashes are required to solve a block, and you're hashing at 100 megahashes/sec, then you'll average 1/7777.78 blocks solved per hour).  Therefore the expected reward from coin generation is unchanged regardless of the network's hashrate.  The number of transactions that can reasonably be expected to be included per block would increase by a factor of roughly six (probably less as some transactions may have inputs from within the previous hour).  So if the transaction fees before were averaging n% of the generation reward, they would be 6xn% of the generation reward now, where x is the increase in average transaction fees added by users to get their transactions processed faster.

****

Even if transaction fees double, the increase in topline for a miner is roughly 12%: if it was 100 mBTC/hr before, then it's 112 mBTC/hr now.  The only miners who can be reasonably expected to start mining are thus those whose reckoning of the cost of mining (however they may reckon that cost) is between 100 and 112 mBTC/hr (if their reckoning is a greater cost than 112 mBTC/hr, they're not going to start; if their reckoning is below 100 mBTC/hr, they're already mining).  How much mining capacity falls into that camp, I don't know, but it's probably not prudent to assume that it's enough to make things much better.

There's bound to be a delay in the transaction fees increasing: many users don't pay a huge amount of attention to the forums or to the blockchain.  They won't find out that transaction processing is slow until they've tried to send coins and found that it's not getting confirmed.  At that point, it's too late to adjust the fee and probably not worth sending the coins again with a bigger fee.

Whether any benefit accrues to any individual user in terms of processing speed in this situation from upping the transaction fee they're willing to pay is also questionable.  An individual user deciding he's willing to pay 1 BTC to move 1 bitcent is unlikely to increase the network hashrate in the short term (catching a 1 BTC tx fee is great for the miner who solves that particular block, but what miner is going to decide to start mining because of that fee, given that the odds of solving the block aren't that great (and in the case of deciding to join a pool, one's share of that isn't going to be significant)).  If even such a dramatic increase in tx fee by a given user is unlikely to noticeably increase the speed at which the transaction gets included in the chain, is there really any incentive for any user to increase the fee?  If you have something that needs to be done by somebody for the common good but that something is to the detriment of the one who does it, don't be surprised if that needed thing doesn't get done (unless some means is developed for the community to reward those who "took one for the team", but there's no real guarantee of such a means developing in this context).

I never said that mining was free, only that energy costs were effectively free if you had a use for that heat.  If I lived in Iceland, where the heating season lasts roughly 50 out of 52 weeks per year, a miner tied to my thermostat is a great thing.  However, if the odds of my success at mining profits were to suddenly jump up due to temporal conditions, I might as well move the thermostat to 72 from 69 and enjoy the warmth.  (F not C, but you get the idea)  The rest of the time I just leave it set on 69 and run on auto.  If it finds a block, I get a bonus.  If it never finds a block, I've not lost anything for doing it, as the heat bill is still the same.
I contest your numbers are accurate, as I think that users that want to get their transactions processed will pay more than otherwise, but I'm not willing to do the math to see what that would be.


I don't know why, neither do you.  This is the point.  You're just guessing as much as I am.

Don't assume that I'm mining

If mining is free, then why on earth is a miner who's given it any thought at all on the sidelines?  Perhaps, it might be ventured, that some non-monetary cost (the noise made by the mining rigs, one's significant other threatening to leave if mining continues, etc.) pushed the benefit-cost difference negative.  Regardless of the reason the hash capacity is on the sidelines, it seems unlikely that a 6% increase in benefit to turning on the rig would lead to large number of those differences turning positive.

What primary costs are there to mining in the general case anyway?  Electricity, Equipment, and Bandwidth are all I can think of off the top of my head.  Equipment cost shouldn't affect the decision of whether to take capacity offline.  Electricity and bandwidth cost should be broadly the same over a given period of real time mining, regardless of the block-solving rate(this is true whether your electricity/bandwidth is free or unbelievably expensive (and if electricity and bandwidth are free, again, why are you on the sidelines?)).

Don't assume that all miners have the same primary costs as yourself, or even most.  Mining is literally free if you heat with electric anyway, and it's winter where you are.

Hence, "not much".

Electricity cost would be the most logical reason for a miner being on the sidelines (depreciation affects the decision of whether to buy/expand a rig and start mining, but not the decision of whether to keep mining because the cost of the rig is a sunk cost: once you've spent the money on a mining rig, maximal return is achieved by running that rig into the ground).  If 5/6 of the hashing power dropped out, it would then take an hour to solve a block but the remaining miners would have six times the chance of solving a block.  The EV for the remaining miners of coin generation over a given period of time doesn't change.  Adding marginal hash capacity from the sidelines would only very slightly increase the EV over what it was pre-fall-in-hash-rate.  As for transaction fees, we could start with assuming those would be increased by a factor of six per solved block (though the factor is probably lower: some number of the transactions that would have occurred later in the hour that got backed up are spending coins that were spent earlier that hour).  If transaction fees are typically 1% of the generation reward (the vast majority of the most recent blocks have transaction fees that won't get to that level for the better part of a decade), then we've got an increase in EV for a miner of about 6%, so only those miners for whom electricity cost was 106% of their return would come back in: any miner whose electricity cost was more than that is still losing money if they come in and can't be reasonably expected to come off the sidelines.

If the hashing power drops significantly enough to drag out the average block time above an hour, than not only does the odds of a particular miner catching the next block increase (even if it takes longer) but the odds of catching transactions due to backups of processing increases.  So there is certainly an economic incentive for a miner on the sidelines to jump in if some major miners were to suddenly drop out, even before the difficulty is adjusted.

The biggest worry has to be that there's a steep enough fall in the hash rate that it starts taking, say, an hour to solve a block (and potentially 4-6 weeks to adjust the difficulty).  There's not much incentive for previously sidelined hash capacity to come online because the difficulty hasn't adjusted to increase the probability of you finding a block.  Meanwhile, six hours to confirm transfers negates a major selling point and basis for demand for bitcoin.

Less extreme than this is a bit of market manipulation by the miner.  Consider a miner who assembles a rig (or collection of rigs) with, say, half the hashing power of the pre-join network (so this miner has 1/3 of the network after he joins).  He joins immediately after a difficulty adjustment and thus collects approximately 33,600 BTC in 224 hours before the difficulty adjusts.  When the difficulty adjusts, the miner cuts the power to the rigs.  With the difficulty adjustment and loss of 1/3 of the hash rate, blocks now average 15 minutes to solve, so the miner can wait 504 hours before rejoining.

Over the course of 728 hours of this strategy, had he mined continuously, he would have mined 72,800 BTC while using 728 units of electricity consumption (a ratio of 100 BTC/uec) while by stopping after the difficulty adjustment, he mines 33,600 BTC for 224 units of electricity consumption (150 BTC/uec).  Depending on the relative costs of electricity versus the cost of assembling a rig with a given hashing power, this may even be more profitable after taking depreciation into account (and remember that computing power is generally decreasing in price while energy is generally increasing in price... it may not be now that electricity is a bigger expense for a miner than depreciation but it seems a fairly safe bet that it will be at some point).

perhaps.

as i recall, Moore's Law was cited many times during the development process for Bitcoin.

i think that a vigorous expansion of network hashing power was clearly anticipated.

GPUs are of no concern.  The system was ingeniously designed to adapt.  What worries me is that if mining becomes unprofitable, miners will leave and the network will lose it's security.  Heck, I have my biggest mining rig offline right now, and I am not the least worried. 

Provides goods or services that people are willing to pay for in bitcoin.

I'm kinda looking forward to the reward cut in end-2012. It will mark the point where half the coins were mined, and thus Bitcoin mining as a thing of the past rather than future.

If Bitcoin ever becomes something that's traded rather than just hoarded, initial distribution will no longer play a significant role. If a Bitcoin is mined once, then traded 10 000 times, the spending of miner wealth makes up for 0.01% of the market size. Unimportant.

1: That's not really true except for when GPU mining first became available, but I won't argue the point because it is irrelevant to my argument.

2: No, the person or persons who invested in the mining rig is getting a slightly larger reward than before for strengthening the network which is something anyone can do by investing in some hardware.

3: The EASY way?! It's easy right now compared to what bitcoins were designed to become.  Besides mining is just a small aspect of bitcoin which is designed to distribute bitcoins to people who are willing to spend the time, money, and effort to strengthen the bitcoin network.

4: The opportunity is in buying some bitcoins and getting into the bitcoin community before it really takes off.  Bitcoin is not about earning dollars, its about starting a whole new currency based upon some simple, time-honored, and proven principles.

5: If the people really want to mine bitcoins, they can use there current graphics card and join a pool.  Otherwise they can buy some, or get some from the bitcoin faucet.  But the best way for anyone to get any type of money is not to have it given to them, but to EARN it.

6: YES, Bitcoin is NOT centered on mining, you are missing the whole point of Bitcoin.

Sorry if this sounds a little harsh, but I wanted to get my point across. 

I suspect Satoshi and others did not expect mining to be conducted on GPUs. With miners each virtually owning supercomputers with regards to bitcoin, far fewer miners are needed to drive the difficulty way up. I think this is causing two issues. First, because mining requires fewer people to drive the difficulty up, bitcoins are distributed to fewer miners. Second, because fewer miners receive bitcoins, fewer people are able to get them the easy way. Third, with fewer people receiving bitcoins directly and the bulk of them going to so few, they are seen as objects of curiosity, and not objects of opportunity by would be cpu miners.

There needs to be a solution to get bitcoins in the hands of more people. Even small amounts of bitcoins spread around would create awareness and usefuless for them. I realize that they can be exchanged by financial institutions, but it is too inconvenient for most people to trade them that way. First, we need easier ways to spend bitcoins. I think namecoin can be used as a  vehicle to create virtual and anonymous documentation (such as invoices and receipts) for bitcoin commerce. Second, we need a safe wallet solution for people to store them. Third, we need smartphone apps created that can be used on even the most basic phone OS with internet access. These conditions can help spread the usefulness of bitcoins. Am I missing anything?