Remove "generate bitcoins" from standard client?

[Jered Kenna (TradeHill)]

I think it's time to remove the generate bitcoins option from the standard client or at least put a strong warning on it.

We've had quite a few people say something like "I've run the btc miner for 3 months 6 months or whatever and haven't gotten anything"
I would say for every person that posts and tries to find an answer 10 or more just give up / think it's a scam.

At this point CPU mining is almost pointless. The amount of time it takes to get a block makes the chances of getting one so low.

What do you guys think? Obviously have a client available to CPU mine if for some reason someone wants to.

[1715124669]

1715124669

[1715124669]

1715124669

[1715124669]

1715124669

[Jered Kenna (TradeHill)]

Quoting holy-fire because he said it so well.

"I agree with dsg and stonetz. To summarize the required changes to the site:
 - Remove the suggestion to turn on bitcoin generation.
 - Remove the estimates for the time taken by a CPU to generate a block. They will always be outdated and it's useless information (as Jim said, it already takes several years instead of what's on the site).
 - Emphasize that generation is done on a GPU by a separate software.
 - Add discussion of trading and references for possible places (Forum/Marketplace, otc, mtgox, Coincard+Coinpal...). Especially shops and organizations who consider accepting Bitcoin need information on easy ways to exchange them for fiat currency.

And, of course, the bitcoin generation should be removed from the client. It was nice for a while, but now it does little more than mislead.

Now, for the most important part - who should be contacted about all this? I'm guessing Gavin is one option. Are there any others in control of the site?"

[stakhanov]

I know I'm the minority here, but I think a big warning is enough.

I like the generation function of the default client. I'm not into mining, but I still like the opportunity to get lucky someday. Also, it reminds people of the existence of mining, which is good for the network. Mining should not only be in the hands of a few specialists.

BTW, why not add GPU generation in the default client? I know it will be a lot of work to support all platforms, but it doesn't have to be done overnight.

[Jered Kenna (TradeHill)]

I know I'm the minority here, but I think a big warning is enough.

I like the generation function of the default client. I'm not into mining, but I still like the opportunity to get lucky someday. Also, it reminds people of the existence of mining, which is good for the network. Mining should not only be in the hands of a few specialists.

BTW, why not add GPU generation in the default client? I know it will be a lot of work to support all platforms, but it doesn't have to be done overnight.

Good point, I'm torn between remove it and a big warning.
If you could add the GPU to it that would be nice. Hopefully someone can say how easy or hard that is to do.

[dsg]

Adding GPU mining would add a lot of dependencies (OpenCL framework, etc) to the bitcoin client that 99% of people will not need. Therefore it makes sense to split mining into it's own application with those dependencies.

Also, remember that the bitcoin program is a financial program. It needs to be secure. The more bells and whistles added, the larger the attack surface and the more danger of subtle bugs creeping in.

Of course, mining should be explained - but there's no need to tease every user with the option when a tiny fraction of them will be using it. Those users can easily fetch one of the many miners available.

[Meni Rosenfeld]

QUoting holy-fire because he said it so well.

Thank you . But it will look better if you use the standard quoting format.

[Jered Kenna (TradeHill)]

QUoting holy-fire because he said it so well.

Thank you . But it will look better if you use the standard quoting format.

Not sure how to quote post to post... I can give it a try.

[Meni Rosenfeld]

QUoting holy-fire because he said it so well.

Thank you . But it will look better if you use the standard quoting format.

Not sure how to quote post to post... I can give it a try.

You don't need to use a forum feature for this, you can do it manually. You can use

Code:

[quote] or
[quote=Holy-Fire] or
[quote author=Holy-Fire link=topic=4786.msg70546#msg70546 date=1300887416]
and end with
[/quote]

Where the one that contains a link was copy-pasted from generating a quote in that thread.

[theymos]

It should be removed, but Bitcoin should continue to provide getwork, and a reference miner should become part of the project. People will get used to the fact that mining requires downloading another program.

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

[Jered Kenna (TradeHill)]

It should be removed, but Bitcoin should continue to provide getwork, and a reference miner should become part of the project. People will get used to the fact that mining requires downloading another program.

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

I was thinking about that too but won't someone who essentially never solves a block have a near 0 effect on the network?
I'd guess that if you solve a block every 6 months on average (or worse) you won't speed it up / affect other miners you'll just waste energy. Is there any benefit to it at all? Assuming you're not in a pool of course.

[MacRohard]

It should be removed, but Bitcoin should continue to provide getwork, and a reference miner should become part of the project. People will get used to the fact that mining requires downloading another program.

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

That doesn't make any sense. Surely all hash/s help the network.. why should anyone be concerned about pushing other miners out of the market?

[MacRohard]

It should be removed, but Bitcoin should continue to provide getwork, and a reference miner should become part of the project. People will get used to the fact that mining requires downloading another program.

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

That doesn't make any sense. Surely all hash/s help the network.. why should anyone be concerned about pushing other miners out of the market?

Infact.. I think the network would actually be better served by hashing power being as widely distributed as possible - not concenrated in the hands of a small number of miners.

[barbarousrelic]

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

Could you explain this a bit further? I don't see how raising the difficulty is anything but a good thing. Also if people mining on their CPUs don't have a real chance of mining a coin, then I don't see how they will be any real negative influence on miners.


It's not low-scale miners' responsibility to make things easier for the high-scale miners.

[Jered Kenna (TradeHill)]

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

Could you explain this a bit further? I don't see how raising the difficulty is anything but a good thing. Also if people mining on their CPUs don't have a real chance of mining a coin, then I don't see how they will be any real negative influence on miners.

Yeah I'm lost too, seems it should either help or have no effect.

[Meni Rosenfeld]

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

Could you explain this a bit further? I don't see how raising the difficulty is anything but a good thing. Also if people mining on their CPUs don't have a real chance of mining a coin, then I don't see how they will be any real negative influence on miners.

Yeah I'm lost too, seems it should either help or have no effect.

I don't know if that's what theymos had in mind, but there's the environment to worry about. Using power for an efficient miner that makes a real contribution to network security is one thing, but wasting it on CPU mining is another.

[Jered Kenna (TradeHill)]

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

Could you explain this a bit further? I don't see how raising the difficulty is anything but a good thing. Also if people mining on their CPUs don't have a real chance of mining a coin, then I don't see how they will be any real negative influence on miners.

Yeah I'm lost too, seems it should either help or have no effect.

I don't know if that's what theymos had in mind, but there's the environment to worry about. Using power for an efficient miner that makes a real contribution to network security is one thing, but wasting it on CPU mining is another.

Agreed. At this point I'm more concerned with them giving up than burning an extra 100grams of coal but if you multiply that times 100000 it would be huge. The thing is and I could be wrong but it seems that CPU mining is so inefficient that even if you were using the servers at work etc you would be working really hard for what a GPU could do easily.

P.S. fixed your quote above, thanks for the tip

[tcatm]

I made a patch that removes the miner (not getwork): https://github.com/bitcoin/bitcoin/pull/132

[theymos]

I was thinking about that too but won't someone who essentially never solves a block have a near 0 effect on the network?
I'd guess that if you solve a block every 6 months on average (or worse) you won't speed it up / affect other miners you'll just waste energy. Is there any benefit to it at all? Assuming you're not in a pool of course.

If the idea spreads that mining helps the Bitcoin network, then tens of thousands of people might turn on generation. Many of these people won't ever win a block, but if some do, then the network will be affected. Every time an unprofitable miner wins a block (or whenever they contribute to a pool), the difficulty will go up.

When the difficulty goes up, the least efficient miners are pushed out of the market unless they are volunteers. Volunteers therefore take up a greater and greater portion of the network's total CPU. This is bad for at least two reasons:

• The network becomes less efficient, using more energy than it needs to.
• "Amateur" miners are not able to respond to threats as quickly as professional miners. They're probably not running the most recent version of Bitcoin, and even if they are, professional miners can make changes to Bitcoin without a new release. The situation is better when the amateur miner is part of a pool, but if the pool goes rogue, the amateur miner will probably not know about it.

It's also really going to irritate me if I see propaganda saying, "Do your part: mine Bitcoins!" or something like that, when the network is perfectly capable of running without volunteers.

[Jered Kenna (TradeHill)]

I was thinking about that too but won't someone who essentially never solves a block have a near 0 effect on the network?
I'd guess that if you solve a block every 6 months on average (or worse) you won't speed it up / affect other miners you'll just waste energy. Is there any benefit to it at all? Assuming you're not in a pool of course.

If the idea spreads that mining helps the Bitcoin network, then tens of thousands of people might turn on generation. Many of these people won't ever win a block, but if some do, then the network will be affected. Every time an unprofitable miner wins a block (or whenever they contribute to a pool), the difficulty will go up.

When the difficulty goes up, the least efficient miners are pushed out of the market unless they are volunteers. Volunteers therefore take up a greater and greater portion of the network's total CPU. This is bad for at least two reasons:

• The network becomes less efficient, using more energy than it needs to.
• "Amateur" miners are not able to respond to threats as quickly as professional miners. They're probably not running the most recent version of Bitcoin, and even if they are, professional miners can make changes to Bitcoin without a new release. The situation is better when the amateur miner is part of a pool, but if the pool goes rogue, the amateur miner will probably not know about it.

It's also really going to irritate me if I see propaganda saying, "Do your part: mine Bitcoins!" or something like that, when the network is perfectly capable of running without volunteers.

That makes complete sense. You're probably going to have some "grandma's basement" type of miners killing the efficiency and running some pro miners out too but I dont' see if having the same effect that an army of CPU's would. Thanks for taking the time to break that down for us.

[jgarzik]

We should remove the option, but not the miner.

If we remove the CPU miner from bitcoin, then people starting alternative block chains will not have a self-sustaining system.

Just disable it from the GUI, so that new users won't find "Generate coins?"

[theymos]

We should remove the option, but not the miner.

If we remove the CPU miner from bitcoin, then people starting alternative block chains will not have a self-sustaining system.

Just disable it from the GUI, so that new users won't find "Generate coins?"

People starting alternative block chains can download the official getwork miner.

[jgarzik]

We should remove the option, but not the miner.

If we remove the CPU miner from bitcoin, then people starting alternative block chains will not have a self-sustaining system.

Just disable it from the GUI, so that new users won't find "Generate coins?"

People starting alternative block chains can download the official getwork miner.

Yes -- that's what they're doing right now.  So why add an extra step.

[MacRohard]

I was thinking about that too but won't someone who essentially never solves a block have a near 0 effect on the network?
I'd guess that if you solve a block every 6 months on average (or worse) you won't speed it up / affect other miners you'll just waste energy. Is there any benefit to it at all? Assuming you're not in a pool of course.

If the idea spreads that mining helps the Bitcoin network, then tens of thousands of people might turn on generation. Many of these people won't ever win a block, but if some do, then the network will be affected. Every time an unprofitable miner wins a block (or whenever they contribute to a pool), the difficulty will go up.

When the difficulty goes up, the least efficient miners are pushed out of the market unless they are volunteers. Volunteers therefore take up a greater and greater portion of the network's total CPU. This is bad for at least two reasons:

• The network becomes less efficient, using more energy than it needs to.
• "Amateur" miners are not able to respond to threats as quickly as professional miners. They're probably not running the most recent version of Bitcoin, and even if they are, professional miners can make changes to Bitcoin without a new release. The situation is better when the amateur miner is part of a pool, but if the pool goes rogue, the amateur miner will probably not know about it.

It's also really going to irritate me if I see propaganda saying, "Do your part: mine Bitcoins!" or something like that, when the network is perfectly capable of running without volunteers.

I don't mine at all myself as I don't like the extra noise from my GPU fan.. not worth it for the profits I could receive from mining. If anyone asks I'd tell them not to bother too.. however,

I do find your position troubling. You seem to be making up bogus reasons to justify taking an action that *is* detrimental to the integrity of the network for the sake of protecting the profits of professional miners. If not profits.. you at least want to concentrate control over the network in the hands of a smaller subset of the participants.. I'm not sure that's a good thing. It allows flexibility, but that cuts two ways.

A large number of unconnected and uninterested parties contributing to the network seems better as they would be much harder for somebody with an agenda to influence than a small (much much smaller - hundreds instead of thousands of participants) cabal of professional miners and mining pool bosses.

[theymos]

A large number of unconnected and uninterested parties contributing to the network seems better as they would be much harder for somebody with an agenda to influence than a small (much much smaller - hundreds instead of thousands of participants) cabal of professional miners and mining pool bosses.

(I also don't mine.)

Bitcoin was always designed to be supported by a small backbone of powerful network nodes. Network/disk/CPU requirements will eventually limit the number of actual network nodes. The wiki article on scalability estimates a network connection of 8 gigabits per second will be required for full network nodes on a VISA-sized Bitcoin network.

Since there will be only a few hundred backbone network nodes, I would prefer them to be professional miners rather than pools supported by amateur volunteers. I want "regular users" to forget about mining now so there is no demand for pools in the future.

[Jered Kenna (TradeHill)]

A large number of unconnected and uninterested parties contributing to the network seems better as they would be much harder for somebody with an agenda to influence than a small (much much smaller - hundreds instead of thousands of participants) cabal of professional miners and mining pool bosses.

(I also don't mine.)

Bitcoin was always designed to be supported by a small backbone of powerful network nodes. Network/disk/CPU requirements will eventually limit the number of actual network nodes. The wiki article on scalability estimates a network connection of 8 gigabits per second will be required for full network nodes on a VISA-sized Bitcoin network.

Since there will be only a few hundred backbone network nodes, I would prefer them to be professional miners rather than pools supported by amateur volunteers. I want "regular users" to forget about mining now so there is no demand for pools in the future.

How do you define professional vs amateur in regards to bitcoin mining at this point and in the future? You could explain it in rigs or hash per second or eficiency or acountability or whatever works for you

[theymos]

How do you define professional vs amateur in regards to bitcoin mining at this point and in the future? You could explain it in rigs or hash per second or eficiency or acountability or whatever works for you

People who are mining for profit.

[barbarousrelic]

How do you define professional vs amateur in regards to bitcoin mining at this point and in the future? You could explain it in rigs or hash per second or eficiency or acountability or whatever works for you

People who are mining for profit.

Doesn't everybody, big or small, mine with the hope that they will profit?

[Meni Rosenfeld]

How do you define professional vs amateur in regards to bitcoin mining at this point and in the future? You could explain it in rigs or hash per second or eficiency or acountability or whatever works for you

People who are mining for profit.

Then I'd say you mean by "professional" something else than most people. I mine for profit, but I certainly am not a professional miner. As long as mining can be done at home, scalable professional mining will have less profit margin than amateurs who leverage existing infrastructure.

[Jered Kenna (TradeHill)]

How do you define professional vs amateur in regards to bitcoin mining at this point and in the future? You could explain it in rigs or hash per second or eficiency or acountability or whatever works for you

People who are mining for profit.

Then I'd say you mean by "professional" something else than most people. I mine for profit, but I certainly am not a professional miner. As long as mining can be done at home, scalable professional mining will have less profit margin than amateurs who leverage existing infrastructure.

Maybe as a business sounds better than for profit.

existing infrastructure won't be more profitable if it comes down to electricity costs and your GPU gets x hashes for 1watt and the other guy who bought more efficient hardware gets x hashes for .7watts

I don't see how the amatures are going to beat the professionals unless they are using their parent's electricity etc and then they're not really winning just being subsidized, you get what I'm saying. Of course if you somehow get free or super cheap electricity another way then maybe but if you're putting that much effort in then I'd call you professional.

[Meni Rosenfeld]

How do you define professional vs amateur in regards to bitcoin mining at this point and in the future? You could explain it in rigs or hash per second or eficiency or acountability or whatever works for you

People who are mining for profit.

Then I'd say you mean by "professional" something else than most people. I mine for profit, but I certainly am not a professional miner. As long as mining can be done at home, scalable professional mining will have less profit margin than amateurs who leverage existing infrastructure.

Maybe as a business sounds better than for profit.

existing infrastructure won't be more profitable if it comes down to electricity costs and your GPU gets x hashes for 1watt and the other guy who bought more efficient hardware gets x hashes for .7watts

I don't see how the amatures are going to beat the professionals unless they are using their parent's electricity etc and then they're not really winning just being subsidized, you get what I'm saying. Of course if you somehow get free or super cheap electricity another way then maybe but if you're putting that much effort in then I'd call you professional.

Electricity is far from being the only cost of mining. There's depreciation of hardware + interest on the capital to acquire it, rent for the physical location of the servers, making sure that physical location has a wide enough connection to the power grid (a typical apartment has much more room for mining rigs than electrical capacity to power them), salaries for technicians, insurance, "businessy" expenses like accounting and legal, and so on.
I'm not assuming an amateur has free electricity, but I am assuming they have existing hardware or at least a computer for which they buy a dedicated GPU, a place to put it, some extra electrical capacity, and are willing to put in some maintenance time as a hobby activity. I don't think even a x2 per-watt performance will allow professionals to compete, and they won't have that because they'll probably use the same GPU/dedicated hashing device as the amateur.
Don't get me wrong, mining probably will be dominated by professionals, because amateur mining is not scalable.

[gohan]

I'm also divided. One side of me says:

• Rename "Generate Coins" to "Help Network" and don't even write about a reward in the pop-up.
• Put a pop-up warning saying that it's better to use a more efficient miner (power-consumption-wise) to help the network if the user has a graphic card.
• Improve miner (at least add SSE2 support, etc.).

I think having this option in the default client is a good thing for the network.

The other side of me says the on-board miner won't ever be as efficient as standalone miners and it will be a waste of resources. Plus, I'm guessing none of the alternative clients does/will have a built-in miner. I still voted for "keep it with a warning" option.

[Jered Kenna (TradeHill)]

I'm also divided. One side of me says:

• Rename "Generate Coins" to "Help Network" and don't even write about a reward in the pop-up.
• Put a pop-up warning saying that it's better to use a more efficient miner (power-consumption-wise) to help the network if the user has a graphic card.
• Improve miner (at least add SSE2 support, etc.).

I think having this option in the default client is a good thing for the network.

The other side of me says the on-board miner won't ever be as efficient as standalone miners and it will be a waste of resources. Plus, I'm guessing none of the alternative clients does/will have a built-in miner. I still voted for "keep it with a warning" option.

As it is there is such a significant difference between the two. I imagine as 5970 and then 6990 prices fall the gap will be even larger and with increased mining / difficulty CPU mining will be even less efficient. There will be a point when it's beyond pointless to CPU mine. I would guess as it stands now you could spend enough on electricity to buy an ok GPU before you get a block. In 3 or 6 months it will be even worse.

[Anonymous]

It should be in an "advanced"  tab  with clear warnings about expected returns.

Its almost false advertising.

[CryptikEnigma]

I think and voted that there should be a clear warning that you will not generate a block for years.

[gohan]

I think and voted that there should be a clear warning that you will not generate a block for years.

I still think that not even mentioning block generation is much better, if the option will stay. It should be like running Seti@Home, purely to support the network.

Though as stonetz said, it will become pointless at some point, if it isn't already, so maybe we can advertise some easy to use GUI GPU miner in the download pages or the client itself, so that the clueless user can run the miner without any hassle together with the client (then server mode would need to be on by default or be able to be configured through GUI).

[Meni Rosenfeld]

I think and voted that there should be a clear warning that you will not generate a block for years.

I still think that not even mentioning block generation is much better, if the option will stay. It should be like running Seti@Home, purely to support the network.

Though as stonetz said, it will become pointless at some point, if it isn't already, so maybe we can advertise some easy to use GUI GPU miner in the download pages or the client itself, so that the clueless user can run the miner without any hassle together with the client (then server mode would need to be on by default or be able to be configured through GUI).

It's already pretty much pointless. There's no sense in deluding newcomers into thinking that by taking the effort to waste power on this they are supporting the network in a non-negligible way. Humans' incentive structure is weird, if you tell them something will "contribute" they won't bother to ask "how much does it contribute?".
On the other hand, when a person contributes something insignificant (or believes he does) it causes him to be more responsive to making substantial contributions later on. This is an argument in favor of including the "feel good" button, but not strong enough in my opinion.

[genjix]

yeah I left it on a few times before (running on CPU) for battle hardening the network. Like seeding torrents to give to charity.

[da2ce7]

yeah I left it on a few times before (running on CPU) for battle hardening the network. Like seeding torrents to give to charity.

What would be better is to remove mining from the Bitcoin Client, and create a super user friendly mining software that automatically takes advantage of the CPU or GPU (if available).  Create a API so that this mining program so users can get statistics about the rate of Bitcoin being generated directly from their favourite pool.

[gohan]

Create a API so that this mining program so users can get statistics about the rate of Bitcoin being generated directly from their favourite pool.

Don't mind me if I am overthinking this but I don't think pooled mining should be encouraged officially. The only advantage of regular users not "forgetting" (as theymos put it) about mining, is to make the network more resilient by having fewer weak spots. Eventually the number of network nodes will have to be limited to powerful operators, but until we have a multitude of such operators, solo mining can be encouraged. Though maybe if it's made easy as a click for the regular user, there will be many more pools and my argument would become invalid.

[Jim Hyslop]

If we remove the CPU miner from bitcoin, then people starting alternative block chains will not have a self-sustaining system.

Wait, what? Alternative block chains?

[Jim Hyslop]

Since there will be only a few hundred backbone network nodes, I would prefer them to be professional miners rather than pools supported by amateur volunteers. I want "regular users" to forget about mining now so there is no demand for pools in the future.

So what kind of processing power do you see a professional miner having? I saw someone (I can envision his picture, but I can't find it right now to mention him by name) talking about setting up 20 or more GPU units. Is that the level you're talking about?

[jgarzik]

If we remove the CPU miner from bitcoin, then people starting alternative block chains will not have a self-sustaining system.

Wait, what? Alternative block chains?

You didn't think The Chain would be the only one, did you? 

Competition is practically inevitable in this space.

[theymos]

So what kind of processing power do you see a professional miner having? I saw someone (I can envision his picture, but I can't find it right now to mention him by name) talking about setting up 20 or more GPU units. Is that the level you're talking about?

ArtForz is the model for Bitcoin mining companies of the future. He designed his own structured ASIC mining chips, and he somehow gets free electricity. With a few more people like ArtForz, even GPUs will become unprofitable. Then the distinction between professional and amateur miners will be more clear.

Mining companies will have state-of-the-art custom mining hardware. They will have an incredible amount of processing power. I believe they will end up having so much power that even the largest botnets will be unable to compete.

[stakhanov]

Mining companies will have state-of-the-art custom mining hardware. They will have an incredible amount of processing power. I believe they will end up having so much power that even the largest botnets will be unable to compete.

What if they decide to stop playing nicely? For example, they could start requesting very high fees, and since they would be generating almost all the blocks, you'd pretty much have to pay unless you want to wait for weeks.

[theymos]

What if they decide to stop playing nicely? For example, they could start requesting very high fees, and since they would be generating almost all the blocks, you'd pretty much have to pay unless you want to wait for weeks.

The fees of the skipped transactions will accumulate until someone invests enough money to put the transactions in a block and collect the fees. I doubt this will take weeks.

[Meni Rosenfeld]

Mining companies will have state-of-the-art custom mining hardware. They will have an incredible amount of processing power. I believe they will end up having so much power that even the largest botnets will be unable to compete.

What if they decide to stop playing nicely? For example, they could start requesting very high fees, and since they would be generating almost all the blocks, you'd pretty much have to pay unless you want to wait for weeks.

Collusion is difficult with hundreds of competitors.

[Meni Rosenfeld]

So what kind of processing power do you see a professional miner having? I saw someone (I can envision his picture, but I can't find it right now to mention him by name) talking about setting up 20 or more GPU units. Is that the level you're talking about?

ArtForz is the model for Bitcoin mining companies of the future. He designed his own structured ASIC mining chips, and he somehow gets free electricity. With a few more people like ArtForz, even GPUs will become unprofitable. Then the distinction between professional and amateur miners will be more clear.

Mining companies will have state-of-the-art custom mining hardware. They will have an incredible amount of processing power. I believe they will end up having so much power that even the largest botnets will be unable to compete.

Surely there will be companies selling PCI-e dedicated hashing devices to amateurs?

[Jered Kenna (TradeHill)]

So what kind of processing power do you see a professional miner having? I saw someone (I can envision his picture, but I can't find it right now to mention him by name) talking about setting up 20 or more GPU units. Is that the level you're talking about?

ArtForz is the model for Bitcoin mining companies of the future. He designed his own structured ASIC mining chips, and he somehow gets free electricity. With a few more people like ArtForz, even GPUs will become unprofitable. Then the distinction between professional and amateur miners will be more clear.

Mining companies will have state-of-the-art custom mining hardware. They will have an incredible amount of processing power. I believe they will end up having so much power that even the largest botnets will be unable to compete.

Surely there will be companies selling PCI-e dedicated hashing devices to amateurs?

If the market exists they will and if the pro competition is so fierce then it will create that market.

[Jim Hyslop]

So what kind of processing power do you see a professional miner having? I saw someone (I can envision his picture, but I can't find it right now to mention him by name) talking about setting up 20 or more GPU units. Is that the level you're talking about?

ArtForz is the model for Bitcoin mining companies of the future. He designed his own structured ASIC mining chips, and he somehow gets free electricity. With a few more people like ArtForz, even GPUs will become unprofitable. Then the distinction between professional and amateur miners will be more clear.

Mining companies will have state-of-the-art custom mining hardware. They will have an incredible amount of processing power. I believe they will end up having so much power that even the largest botnets will be unable to compete.

Interesting. I had been wondering about designing an FPGA to do the work, but haven't pursued that route to see if it's workable or cost-effective.

[Jim Hyslop]

If we remove the CPU miner from bitcoin, then people starting alternative block chains will not have a self-sustaining system.

Wait, what? Alternative block chains?

You didn't think The Chain would be the only one, did you? 

Competition is practically inevitable in this space.

Well, I figured there would eventually be competitors, but I didn't think they'd have emerged this early. Although, I don't know why I was surprised - it is, as you say, inevitable.

Are there many alternative block chains, and do they call themselves Bitcoin?

[MoonShadow]

If we remove the CPU miner from bitcoin, then people starting alternative block chains will not have a self-sustaining system.

Wait, what? Alternative block chains?

You didn't think The Chain would be the only one, did you? 

Competition is practically inevitable in this space.

Well, I figured there would eventually be competitors, but I didn't think they'd have emerged this early. Although, I don't know why I was surprised - it is, as you say, inevitable.

Are there many alternative block chains, and do they call themselves Bitcoin?

The only alternative beyond a theoretical stage is 'testnet', which is almost exactly like the main blockchain except it exists to get hammered.  I believe that you can join the testnet with the regular client by way of a command line switch, but if it isn't -testnet I wouldn't hazard a guess.

[ryepdx]

So, uh, running away from the philosophical question of alternative blockchains, what if we were to simply have the client do a calculation to estimate how long it'll take a person to generate 50btc on their machine and show that in addition to the hashes per second. That would take care of the "why haven't I gotten any bitcoins yet" posts on the forum.

[Jered Kenna (TradeHill)]

So, uh, running away from the philosophical question of alternative blockchains, what if we were to simply have the client do a calculation to estimate how long it'll take a person to generate 50btc on their machine and show that in addition to the hashes per second. That would take care of the "why haven't I gotten any bitcoins yet" posts on the forum.

I'm not a programmer but that doesn't seem that hard. I'm guessing it would be easy to check current difficulty and compare that vs hash speed. I've heard (and agree with) that the simpler the more secure and better it is though. Still if it popped up and said "with your CPU you will generate your first set of coins in 800days do you wish to continue? click here to find out why"



Of course the warning could just put examples or estimates too.

[Meni Rosenfeld]

So, uh, running away from the philosophical question of alternative blockchains, what if we were to simply have the client do a calculation to estimate how long it'll take a person to generate 50btc on their machine and show that in addition to the hashes per second. That would take care of the "why haven't I gotten any bitcoins yet" posts on the forum.

I'm not a programmer but that doesn't seem that hard. I'm guessing it would be easy to check current difficulty and compare that vs hash speed. I've heard (and agree with) that the simpler the more secure and better it is though. Still if it popped up and said "with your CPU you will generate your first set of coins in 800days do you wish to continue? click here to find out why"

Of course the warning could just put examples or estimates too.

Most people will think that when 800 days elapse they are guaranteed to find a block. This will only lead to frustration. Even if we tell them it's random they'll underestimate its variance. And there's no point in this, CPU mining with a pool is worthless enough, no one in their right mind would want to do it solo.

[MoonShadow]

So, uh, running away from the philosophical question of alternative blockchains, what if we were to simply have the client do a calculation to estimate how long it'll take a person to generate 50btc on their machine and show that in addition to the hashes per second. That would take care of the "why haven't I gotten any bitcoins yet" posts on the forum.

I'm not a programmer but that doesn't seem that hard. I'm guessing it would be easy to check current difficulty and compare that vs hash speed. I've heard (and agree with) that the simpler the more secure and better it is though. Still if it popped up and said "with your CPU you will generate your first set of coins in 800days do you wish to continue? click here to find out why"

Of course the warning could just put examples or estimates too.

Most people will think that when 800 days elapse they are guaranteed to find a block. This will only lead to frustration. Even if we tell them it's random they'll underestimate its variance. And there's no point in this, CPU mining with a pool is worthless enough, no one in their right mind would want to do it solo.

Well, I do it solo.  So your probably right that no one in their right mind would do it.

Of course, I do it (in part) to heat my study.  And no, I have never solved a block.

[Garrett Burgwardt]

So, uh, running away from the philosophical question of alternative blockchains, what if we were to simply have the client do a calculation to estimate how long it'll take a person to generate 50btc on their machine and show that in addition to the hashes per second. That would take care of the "why haven't I gotten any bitcoins yet" posts on the forum.

I'm not a programmer but that doesn't seem that hard. I'm guessing it would be easy to check current difficulty and compare that vs hash speed. I've heard (and agree with) that the simpler the more secure and better it is though. Still if it popped up and said "with your CPU you will generate your first set of coins in 800days do you wish to continue? click here to find out why"

Of course the warning could just put examples or estimates too.

Most people will think that when 800 days elapse they are guaranteed to find a block. This will only lead to frustration. Even if we tell them it's random they'll underestimate its variance. And there's no point in this, CPU mining with a pool is worthless enough, no one in their right mind would want to do it solo.

Well, I do it solo.  So your probably right that no one in their right mind would do it.

Of course, I do it (in part) to heat my study.  And no, I have never solved a block.

You'll find one eventually, and be really confused

[Ian Maxwell]

I'd be fine removing it from the GUI, and having a proper installer that offers to install separate mining software as well.

Either that, or renaming it something like "Help process transactions" and not even indicating that this might be profitable somehow. Folks who know about things like fold@home might be interested just out of charity.

Edited to add: The first option is probably better from a conservationist standpoint. When we already have plenty of nodes processing cheaply on GPUs, adding CPUs to the mix mostly serves to waste energy.

[Garrett Burgwardt]

Either that, or renaming it something like "Help process transactions" and not even indicating that this might be profitable somehow. Folks who know about things like fold@home might be interested just out of charity.

The thing is, they won't be helping. Unless you can solve a block you're literally just wasting power. And given how long it would take a CPU to find a block, you're really just not helping at all.

[Pieter Wuille]

A priori you are helping even with the smallest amount of hashing power, since on average your expected number of blocks found is higher than zero (even though it may be 0.001/year).

A posteriori, if you haven't found anything (the most likely case), you have wasted your power. That doesn't mean it was meaningless, since you always had a chance to find something.

I'm just trying to debunk that it isn't true that having a very large chance for not finding anything necessarily implies completely wasted power.

Concerning the real topic of the discussion however, I believe we shouldn't give people false hope that they can generate coins in any serious amount without investing in mining. I'm in favor of removing mining functionality from the main client, but providing a separate reference getwork-based CPU miner in the source distribution.

[Ricochet]

I'd normally be in favor of a large warning, but we are talking about preparing for an onslaught of average clueless users here.  The average user is not going to read a warning of the length that would be required to sufficiently describe why it's a bad idea, so they'll click OK and still get angry when they don't generate anything.  Thus, keep the text of the menu option but make it pop up a very short message.  Something like "You are unlikely to generate a block in any realistic amount of time on a standard CPU.  If you do still wish to enable generation, download the separate mining client at www.example.com"

Personally I've mostly stopped mining entirely.  The difficulty is so high that even with a pool, using my GPU and both CPU cores, it would take multiple days to earn 0.20 BTC.  Even when my electricity is free, that's not worth the additional strain on my computer's components.

[khal]

In case of network hashing attack, having a miner available from the default client that can (be) activate(d) in case of need may be usefull.
Even if hash rate is small, multiply it by all bitcoin users in the world, that are using bitcoin like a money and not for mining, and i guess it is significant.
Miner could be CPU or GPU, be i think there should be one available simply, whatever name it has, integrated or binary independant.

[carp]

I'd normally be in favor of a large warning, but we are talking about preparing for an onslaught of average clueless users here.  The average user is not going to read a warning of the length that would be required to sufficiently describe why it's a bad idea, so they'll click OK and still get angry when they don't generate anything.  Thus, keep the text of the menu option but make it pop up a very short message.  Something like "You are unlikely to generate a block in any realistic amount of time on a standard CPU.  If you do still wish to enable generation, download the separate mining client at www.example.com"

Personally I've mostly stopped mining entirely.  The difficulty is so high that even with a pool, using my GPU and both CPU cores, it would take multiple days to earn 0.20 BTC.  Even when my electricity is free, that's not worth the additional strain on my computer's components.

I agree. Personally my thought on leaving the code in there, perhaps as an option through the CLI/RPC is that others are right about other block chains but, also, a reference implementation of mining is not a bad idea to keep around.

I was thinking, the client has the current difficulty, and it knows how many hashes/s its generating, why not have it tell you how long it is likely to be between blocks on average for you?

That should be unambiguous. Then add a little warning popup that has a link to some contrived forum post or even a FAQ about mining and CPU vs GPU and that there are other clients out there for mining etc.

[Meni Rosenfeld]

A priori you are helping even with the smallest amount of hashing power, since on average your expected number of blocks found is higher than zero (even though it may be 0.001/year).

A posteriori, if you haven't found anything (the most likely case), you have wasted your power. That doesn't mean it was meaningless, since you always had a chance to find something.

I'm just trying to debunk that it isn't true that having a very large chance for not finding anything necessarily implies completely wasted power.

Their contribution to network security is dwarfed by their damage to the environment, and the electric bill required would be orders of magnitude more effective if contributed elsewhere.

[MoonShadow]

A priori you are helping even with the smallest amount of hashing power, since on average your expected number of blocks found is higher than zero (even though it may be 0.001/year).

A posteriori, if you haven't found anything (the most likely case), you have wasted your power. That doesn't mean it was meaningless, since you always had a chance to find something.

I'm just trying to debunk that it isn't true that having a very large chance for not finding anything necessarily implies completely wasted power.

Their contribution to network security is dwarfed by their damage to the environment, and the electric bill required would be orders of magnitude more effective if contributed elsewhere.

Nonesense.  Most of the individual miners use the waste heat to contribute to their home heating, and many would be contributing to Folding@Home or SETI if they were not running Bitcoin. 

[Meni Rosenfeld]

A priori you are helping even with the smallest amount of hashing power, since on average your expected number of blocks found is higher than zero (even though it may be 0.001/year).

A posteriori, if you haven't found anything (the most likely case), you have wasted your power. That doesn't mean it was meaningless, since you always had a chance to find something.

I'm just trying to debunk that it isn't true that having a very large chance for not finding anything necessarily implies completely wasted power.

Their contribution to network security is dwarfed by their damage to the environment, and the electric bill required would be orders of magnitude more effective if contributed elsewhere.

Nonesense.  Most of the individual miners use the waste heat to contribute to their home heating, and many would be contributing to Folding@Home or SETI if they were not running Bitcoin. 

In most cases, A/C requires less power for a given level of heating.
If CPU folding/seti is worthwhile, people shouldn't mine instead. If it's not, people should be educated that CPU folding/seti is a waste of power which is an entirely different issue.
Even if edge cases exist where CPU mining is a positive net good, it is so small as to not be worth the effort.

[fetokun]

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

I'm ashamed to say that I was one of the people thinking this until right now.
I was leaving my client generating in a "help the network and maybe get luck" kind of attitude.

[Meni Rosenfeld]

The thing I'm most worried about is that people will start mining just because they think they're helping the network, and this unprofitable mining will push more efficient and competent miners out of the market.

I'm ashamed to say that I was one of the people thinking this until right now.
I was leaving my client generating in a "help the network and maybe get luck" kind of attitude.

Don't feel bad. Now you'll be much more motivated to help Bitcoin for real!

[MoonShadow]

A priori you are helping even with the smallest amount of hashing power, since on average your expected number of blocks found is higher than zero (even though it may be 0.001/year).

A posteriori, if you haven't found anything (the most likely case), you have wasted your power. That doesn't mean it was meaningless, since you always had a chance to find something.

I'm just trying to debunk that it isn't true that having a very large chance for not finding anything necessarily implies completely wasted power.

Their contribution to network security is dwarfed by their damage to the environment, and the electric bill required would be orders of magnitude more effective if contributed elsewhere.

Nonesense.  Most of the individual miners use the waste heat to contribute to their home heating, and many would be contributing to Folding@Home or SETI if they were not running Bitcoin. 

In most cases, A/C requires less power for a given level of heating.
If CPU folding/seti is worthwhile, people shouldn't mine instead. If it's not, people should be educated that CPU folding/seti is a waste of power which is an entirely different issue.
Even if edge cases exist where CPU mining is a positive net good, it is so small as to not be worth the effort.

I've lived in many places, and have never had an apartment that was not heated by baseboard electro-resistive heating.  The only alternatives that I've ever had was in single family homes I've owned.  I'm sure that I'm not an anomaly.

[nanotube]

i'd suggest to leave it in for 'just in case' or for historical reasons or whatever... but take it off the main menu, and stick the option somewhere far in advanced preferences. and instead of calling it 'generate coins' call it something else, like 'generate blocks' or 'process blocks', and make a note that it's unlikely to net you any coin.

this should pretty much solve the problem of people coming in and saying "omg i've been generating for 3 days and don't have any coins".

though i guess i won't be against removing it completely, either.

[Jered Kenna (TradeHill)]

i'd suggest to leave it in for 'just in case' or for historical reasons or whatever... but take it off the main menu, and stick the option somewhere far in advanced preferences. and instead of calling it 'generate coins' call it something else, like 'generate blocks' or 'process blocks', and make a note that it's unlikely to net you any coin.

this should pretty much solve the problem of people coming in and saying "omg i've been generating for 3 days and don't have any coins".

though i guess i won't be against removing it completely, either.

Also make sure they know coins come when you solve a block and you get 50btc or none because a lot of people are expecting them to trickle in.

[barbarousrelic]

i'd suggest to leave it in for 'just in case' or for historical reasons or whatever... but take it off the main menu, and stick the option somewhere far in advanced preferences. and instead of calling it 'generate coins' call it something else, like 'generate blocks' or 'process blocks', and make a note that it's unlikely to net you any coin.

this should pretty much solve the problem of people coming in and saying "omg i've been generating for 3 days and don't have any coins".

though i guess i won't be against removing it completely, either.

Also make sure they know coins come when you solve a block and you get 50btc or none because a lot of people are expecting them to trickle in.

They probably get that impression from here:

So your wealth is determined by the amount of CPU power you have?

No. There's a constant average rate of new Bitcoins created, and that amount is divided among the nodes by the CPU power they supply.

http://www.bitcoin.org/faq#So_your_wealth_is_determined_by_the_amount_of_CPU_power_you_have

[Akarbb]

I wrote an article about this particular topic and my solution to it.

http://thedailybitcoin.blogspot.com/2011/04/cpu-mining-no-longer-viable-option.html

[jgarzik]

I wrote an article about this particular topic and my solution to it.

http://thedailybitcoin.blogspot.com/2011/04/cpu-mining-no-longer-viable-option.html

GPU mining will not be added to the bitcoin client...  too many annoying dependencies.

[Akarbb]

I wrote an article about this particular topic and my solution to it.

http://thedailybitcoin.blogspot.com/2011/04/cpu-mining-no-longer-viable-option.html

GPU mining will not be added to the bitcoin client...  too many annoying dependencies.

Dependencies?  On a windows machine, the OpenCL drivers would be set up already when you put in a new card.  The GPU miners that exist right now require very little work, in fact Kiv's GUI miner requires almost none.  It seems like all you would be doing would be combining the libraries for the GPU miner into the package for the Bitcoin standard miner.  It might be a slightly larger file, but it would have much more functionality.

[jgarzik]

Dependencies?  On a windows machine, the OpenCL drivers would be set up already when you put in a new card.  The GPU miners that exist right now require very little work, in fact Kiv's GUI miner requires almost none.  It seems like all you would be doing would be combining the libraries for the GPU miner into the package for the Bitcoin standard miner.  It might be a slightly larger file, but it would have much more functionality.

It's definitely not that easy

[Akarbb]

Dependencies?  On a windows machine, the OpenCL drivers would be set up already when you put in a new card.  The GPU miners that exist right now require very little work, in fact Kiv's GUI miner requires almost none.  It seems like all you would be doing would be combining the libraries for the GPU miner into the package for the Bitcoin standard miner.  It might be a slightly larger file, but it would have much more functionality.

It's definitely not that easy

Definitely not easy, but also not impossible. 

[error]

Dependencies?  On a windows machine, the OpenCL drivers would be set up already when you put in a new card.  The GPU miners that exist right now require very little work, in fact Kiv's GUI miner requires almost none.  It seems like all you would be doing would be combining the libraries for the GPU miner into the package for the Bitcoin standard miner.  It might be a slightly larger file, but it would have much more functionality.

It's definitely not that easy

Definitely not easy, but also not impossible. 

OK, let's see your pull request.

[Jim Hyslop]

Dependencies?  On a windows machine[...]

What about the other platforms that Bitcoin supports? You can't focus on one operating system.

[gigabytecoin]

I think it's time to remove the generate bitcoins option from the standard client or at least put a strong warning on it.

We've had quite a few people say something like "I've run the btc miner for 3 months 6 months or whatever and haven't gotten anything"
I would say for every person that posts and tries to find an answer 10 or more just give up / think it's a scam.

At this point CPU mining is almost pointless. The amount of time it takes to get a block makes the chances of getting one so low.

What do you guys think? Obviously have a client available to CPU mine if for some reason someone wants to.

I just booted up a new computer the other day and it's CPU was getting close the 30 MegaHash/s

Of course it would be unwise for the average joe to leave his computer running for 140 days straight just to get 50 bitcoins, I do not think it is up to us to determine exactly what the public's acceptable level of tradeoff should be in this situation.

[MoonShadow]

It's also important to keep the ability to mine via cpu's in the vanilla client, should there ever come a day when the system is attacked, and the userbase is called upon to mine at a loss in order to defend against an ongoing blockchain attack.

[Jered Kenna (TradeHill)]

It's also important to keep the ability to mine via cpu's in the vanilla client, should there ever come a day when the system is attacked, and the userbase is called upon to mine at a loss in order to defend against an ongoing blockchain attack.

But if they don't (or  1 or 2 lucky ones do) solve any blocks will that even help?

[khal]

Non generating nodes will be numerous. More the difficulty augments and more nodes won't generate coins, because bitcoin will ben used like a money.
Do you think each people who will use bitcoin like a money will try to generate coins if they must wait 2 years ?

Now, imagine 100'000 nodes which can generate 1 block each 2 years (a quad core is currently more efficient than that), this leads to a significant hashing power (pool), even if from an individual perspective it is useless.


But, this argument will disappear if light clients (without the whole blockchain) are more used than the current one, because light clients can't generate blocks if i remember correctly what gavin write in another thread.

[MoonShadow]

It's also important to keep the ability to mine via cpu's in the vanilla client, should there ever come a day when the system is attacked, and the userbase is called upon to mine at a loss in order to defend against an ongoing blockchain attack.

But if they don't (or  1 or 2 lucky ones do) solve any blocks will that even help?

Yes.  Like raindrops form a river.

[MoonShadow]

But, this argument will disappear if light clients (without the whole blockchain) are more used than the current one, because light clients can't generate blocks if i remember correctly what gavin write in another thread.

Although there will be 'lightweight' clients in the future, most likely those will be on consumer devices without the capacity to support a full client.  Future users of Bitcoin are as likely to have a client running on portable devices and home computers.  There is no reason to expect that future users are going to prefer lightweight clients on personal computers.

[divergenta]

What is the point of having it in the standard  client if it's pretty much useless?

[MoonShadow]

What is the point of having it in the standard  client if it's pretty much useless?

It's not useless.

[divergenta]

As far as I know it would take forever to mine anything with just CPU even if you have a good processor so please explain why it's useful if you not have access to some kind of supercomputer?

I mean, doesn't this feature give new users false hopes and will make them waste their CPU on nothing?

[MoonShadow]

As far as I know it would take forever to mine anything with just CPU even if you have a good processor so please explain why it's useful if you not have access to some kind of supercomputer?

I mean, doesn't this feature give new users false hopes and will make them waste their CPU on nothing?

I've already explained it.  Please read the thread.  If you are just thinking about what you personally gain from mining, then it is pointless to you.  If you consider the altruisic value of mining, particularly in a crisis event such as an ongoing attack upon the blockchain, then it's not unwise to have every possible cpu still have the ability to mine.  Even if there is no possibility of profit.

[Jim Hyslop]

But, this argument will disappear if light clients (without the whole blockchain) are more used than the current one, because light clients can't generate blocks if i remember correctly what gavin write in another thread.

That will depend what the lightweight clients do.

You can spend Bitcoins without the block chain, but you must have the entire block chain to accept Bitcoins, otherwise you can't protect yourself against a double-spend.

[divergenta]

I've already explained it.  Please read the thread.  If you are just thinking about what you personally gain from mining, then it is pointless to you.  If you consider the altruisic value of mining, particularly in a crisis event such as an ongoing attack upon the blockchain, then it's not unwise to have every possible cpu still have the ability to mine.  Even if there is no possibility of profit.

Ok, you do have very a good point there. I rest my case.

[MoonShadow]

 but you must have the entire block chain to accept Bitcoins, otherwise you can't protect yourself against a double-spend.

That's not entirely true, as there are other steps that can be taken to mitigate such risks without access to a local blockchain.  Yet, the client does not attempt these other methods at of yet, so your statement is practically true at present.

[Meni Rosenfeld]

But, this argument will disappear if light clients (without the whole blockchain) are more used than the current one, because light clients can't generate blocks if i remember correctly what gavin write in another thread.

Although there will be 'lightweight' clients in the future, most likely those will be on consumer devices without the capacity to support a full client.  Future users of Bitcoin are as likely to have a client running on portable devices and home computers.  There is no reason to expect that future users are going to prefer lightweight clients on personal computers.

If Bitcoin is successful, a personal computer with a standard internet connection will not be able to download all the transaction data. These projections talk about an 8 Gbit/s link. My understanding is that only mining companies will have full nodes.

[MoonShadow]

But, this argument will disappear if light clients (without the whole blockchain) are more used than the current one, because light clients can't generate blocks if i remember correctly what gavin write in another thread.

Although there will be 'lightweight' clients in the future, most likely those will be on consumer devices without the capacity to support a full client.  Future users of Bitcoin are as likely to have a client running on portable devices and home computers.  There is no reason to expect that future users are going to prefer lightweight clients on personal computers.

If Bitcoin is successful, a personal computer with a standard internet connection will not be able to download all the transaction data. These projections talk about an 8 Gbit/s link. My understanding is that only mining companies will have full nodes.

Perhaps only mining companies will have their clients set to run as full nodes; but even if end users don't want their client running in full node mode, doesn't conclude that it won't have the capacity to do so at the user's will.  Those projections are highly speculative anyway, and depend upon the dual assumption that Bitcoin grows to nation-state-economy levels while never developing parrallel systems to mitigate traffic along the way.  Large assumptions, particularly since mitigating systems have already been considered by this forum.

[kiba]

Perhaps only mining companies will have their clients set to run as full nodes; but even if end users don't want their client running in full node mode, doesn't conclude that it won't have the capacity to do so at the user's will.  Those projections are highly speculative anyway, and depend upon the dual assumption that Bitcoin grows to nation-state-economy levels while never developing parrallel systems to mitigate traffic along the way.  Large assumptions, particularly since mitigating systems have already been considered by this forum.

We're also assuming that our internet connection won't get faster.

[khal]

But, this argument will disappear if light clients (without the whole blockchain) are more used than the current one, because light clients can't generate blocks if i remember correctly what gavin write in another thread.

Although there will be 'lightweight' clients in the future, most likely those will be on consumer devices without the capacity to support a full client.  Future users of Bitcoin are as likely to have a client running on portable devices and home computers.  There is no reason to expect that future users are going to prefer lightweight clients on personal computers.

If Bitcoin is successful, a personal computer with a standard internet connection will not be able to download all the transaction data. These projections talk about an 8 Gbit/s link. My understanding is that only mining companies will have full nodes.

- 8 Gbit/s during 1mn and nothing sent during the other 9 minuts.

It's also possible to optimize the protocol somewhat. A solved block could be broadcast as only its header and merkle tree. Nodes would then check the transaction hashes against their own in-memory lists. If a block contains a transaction the node did not see for some reason, it'd then request only those transactions. For the common case of a node with good, 24/7 network connectivity, this would reduce the size of block broadcasts dramatically.

With this "small" optimisation, in the worst case, you will receive/send 2000 transaction/s from/to 50 nodes (2'000*1kB*50) per second + a list of transaction's hashes with a block "header" each 10mn (~50*2000 + ~500). So, in 10 to 100 years, you will need 1Gbps/s duplex to be a supernode, with a single optimisation. Wow, impressive :p

[Pieter Wuille]

I'm not sure what is best. On one hand, I think the "generate coins" option gives both the wrong impression (either "bitcoin = free money!", or "bitcoin = worthless, people can generate it themselves!") and gives people false hope ("what the fuck? it takes 137 years to generate something?! Bye"). On the other hand, the idea of mining may attract people to become involved (like myself...).

I am quite sure though that a "Generate coins" flag in the UI is a bad idea. The only reasonable thing to use it for is testnet, and people who need that, typically don't need a GUI flag. I'm in favor of removing the miner code from the default client, but have an simple reference RPC miner as a separate binary included in the distribution. Ideally, this program should look at the nBits field in the received work units, and give an estimate of expected BTC/time and time/block. Some explanation in a mining.txt file may be useful as well.

The alternative would be leaving miner code in the default client, and either only make it accessible through RPC, or at least rename the option in the GUI ("Verify transactions", or "Support network", or "Generate blocks", but not "Generate coins") and show a warning that it may take ages.

[MoonShadow]

The alternative would be leaving miner code in the default client, and either only make it accessible through RPC, or at least rename the option in the GUI ("Verify transactions", or "Support network", or "Generate blocks", but not "Generate coins") and show a warning that it may take ages.

I'm okay with this.  I would agree that calling it "generate coins" gives the user the wrong idea.  "Support Network" or "Donate CPU Cycles" would be a better option, and it should be buried deeper into the menu and not on the main GUI screen by default.  But I do think that it should remain on the vanilla client.

[deisraeli]

I agree that a "proceed with caution" for mining bitcoins using a CPU vs using a GPU or any other processor should be a notice to those who wish to mine whether they mine for profit or to "help the community by running the Full-Bitcoin client" (Bitcoin.org/en/Choose-your-wallet).

Moreover, I agree that mining should be explained. However, I whole-heartedly disagree that "there's no need to tease every user with the option [mining] when a tiny fraction of them will be using it" (dsg: quote #5). If only a tiny fraction ... will be using it, then what is the problem if every user is teased? Also, I am appalled at "the thing I'm most worried about is people will start mining ... and will push more efficient & competent miners out of the market" (quote #9). Really, these comments explain why in 2014 people believe Bitcoin is a Madoff scam. For example, I have the Full-Bitcoin client and every inquiry or question I pose about solo Bitcoin mining (especially, "getgenerate" and/or "setgenerate <generate> [genproclimit]) I get responses "you won't make money by solo mining - join a pool." Yet, despite Satoshi warning us against third-party sites, it's apparent those responses are from those who have a capitalist centralized mentality because I get those responses despite posting I want to solo mine regardless my intentions. If I am helping the Community then I should be allowed to participate in solo-mining whether I do it for profit, a Satoshi, or do it because I can leave my system running the Full-Client 24/7.

I thought from reading Satoshi's genesis paper that Bitcoin was a peer2peer currency for those who participate in the network. It's people with these centralized ideas based on greed is why government regulation needs to be involved. Hell, if Bitinstant.com was up and running, then I could had easily invested my U.S. Fiat to purchase bitcoins via CVS or Walmart. Just think about the people who lost there. And let's not forget those who lost bitcoins using exchanges that I've read about that were hacked.

Satoshi, if you read this - please post instructions for solo-mining for those of us who are using the Full-Client. I have a AMD graphics card and I want to solo-mine, not merely for profit, but to help the community because I am running Bitcoin-QT from a computer that I can run 24/7. I am not a computer programming geek but I do hold a Master Degree in Business and have computer skills that if I need to create a Bitcoin.config file I can. I just do not understand how to use the functions in the Bitcoin-QT Debug Window:console. However, what's the use of running Bitcoin-QT 24/7 if these capitalists centralized greedy minds you have left in charge keep telling you what not to do instead of how to do it?

[Holliday]

I agree that a "proceed with caution" for mining bitcoins using a CPU vs using a GPU or any other processor should be a notice to those who wish to mine whether they mine for profit or to "help the community by running the Full-Bitcoin client" (Bitcoin.org/en/Choose-your-wallet).

Moreover, I agree that mining should be explained. However, I whole-heartedly disagree that "there's no need to tease every user with the option [mining] when a tiny fraction of them will be using it" (dsg: quote #5). If only a tiny fraction ... will be using it, then what is the problem if every user is teased? Also, I am appalled at "the thing I'm most worried about is people will start mining ... and will push more efficient & competent miners out of the market" (quote #9). Really, these comments explain why in 2014 people believe Bitcoin is a Madoff scam. For example, I have the Full-Bitcoin client and every inquiry or question I pose about solo Bitcoin mining (especially, "getgenerate" and/or "setgenerate <generate> [genproclimit]) I get responses "you won't make money by solo mining - join a pool." Yet, despite Satoshi warning us against third-party sites, it's apparent those responses are from those who have a capitalist centralized mentality because I get those responses despite posting I want to solo mine regardless my intentions. If I am helping the Community then I should be allowed to participate in solo-mining whether I do it for profit, a Satoshi, or do it because I can leave my system running the Full-Client 24/7.

I thought from reading Satoshi's genesis paper that Bitcoin was a peer2peer currency for those who participate in the network. It's people with these centralized ideas based on greed is why government regulation needs to be involved. Hell, if Bitinstant.com was up and running, then I could had easily invested my U.S. Fiat to purchase bitcoins via CVS or Walmart. Just think about the people who lost there. And let's not forget those who lost bitcoins using exchanges that I've read about that were hacked.

Satoshi, if you read this - please post instructions for solo-mining for those of us who are using the Full-Client. I have a AMD graphics card and I want to solo-mine, not merely for profit, but to help the community because I am running Bitcoin-QT from a computer that I can run 24/7. I am not a computer programming geek but I do hold a Master Degree in Business and have computer skills that if I need to create a Bitcoin.config file I can. I just do not understand how to use the functions in the Bitcoin-QT Debug Window:console. However, what's the use of running Bitcoin-QT 24/7 if these capitalists centralized greedy minds you have left in charge keep telling you what not to do instead of how to do it?

There is a decentralized pool model called P2Pool. It gives the network all the benefits of you solo mining, but shares the reward variance reduction of a traditional pool.

Most people say, "don't solo mine" because the chances of you actually finding a block with a single AMD graphics card is so low that you will simply be throwing money away (electricity). I don't think they are being "capitalist centralized greedy minds" so much as they are being realistic.

Anyway, get the best of both worlds while sticking to the decentralized ideals behind Bitcoin. Use P2Pool. (Although, you probably still won't earn anything.)

Here is a guide for P2Pool: https://bitcointalk.org/index.php?topic=153232.0

The actual P2Pool thread: https://bitcointalk.org/index.php?topic=18313.0

By the way, I am a greedy capitalist and I don't like centralization, so maybe relax with the generalizations.