ASIC = The end of decentralized mining

[jjshabadoo]

It is unbelievable that a community of relatively intelligent people cannot see how ASIC will ruin mining profitability for everyone and only make the manufacturers rich.

There is no financial model that can make these work due to their incredibly low marginal operation cost and ridiculously high hashrate.

The ASIC manufacturers would have to control the rate at which these are sold to stand any chance of miners getting a return on their investment and no business is going to do this after spending such a huge initial investment. Especially since after they sell the first several units at high prices, making more is so cheap they can just start unloading them at rock bottom prices.

The early ASIC buyers will never turn them off due to their low operation cost even though they stand ZERO chance of recouping their investment.

This is just standard economics and common sense.

I hope the bitcoin DEVS change the protocol to make ASICS obsolete. It will save the irrational potential buyers from themselves and the rest of us along with it.

[Garr255]

Subed. I'm interested in counteropinions.

[Inspector 2211]

Exactly my thoughts.
Automobiles should never have been allowed, because their success caused the demise of livery stables, horse whip manufacturers and horse buggy manufacturers.
Or maybe automobiles should have be allowed after all, but the ordinance that a man with a warning bell walks in front of them should never have been rescinded.

[Bitcoin Oz]

The first ASIC company that brings them out will have a bit of a hardware monopoly for awhile simply because the cost to implement an ASIC is so high.

Thats why I hope its the open asic team that does so.

[notme]

The first ASIC company that brings them out will have a bit of a hardware monopoly for awhile simply because the cost to implement an ASIC is so high.

Thats why I hope its the open asic team that does so.

BFL is better capitalized and thus more likely to win the race.  Things might face some centralization in the beginning, but soon enough there will be 5-10 vendors.  SHA256(SHA256(X)) is not that complicated compared to many other hardware designs.

[Vladimir]

1. Devs will not change the algo in order to influence ASIC production because it would be a very stupid thing to do and devs are not stupid (well... mostly).

2. ASIC mining is not the end of decentralised mining, it is the beginning. It will make the mining ultimately much more decentralised than it is now, perhaps with exception of short transitional period.

3. ASIC mining proliferation will make Bitcoin much more resilient to 51% attacks and much more secure overall. A few years from now it likely that most governments will not be able to mount a successful computational attack on Bitcoin and before long even the most powerful nations will not have such capability either. Thanks to mass proliferation of various hardware incorporating Bitcoin mining ASIC's.

4. There will be 2 generations of ASIC hardware coming relatively quickly one after another (perhaps with 18-24 month pause), after that it is Moore's law (finally).

5. FPGA investors will regret buying FPGA's instead of just buying bitcoins. Particularly those who are just buying rigs now.

6. Whether initial  ASIC investors will recoup their investment or not is rather not clear.

7. BFL is not the favourite in the ASIC race.

8. Embrace the future.

[notme]

7. BFL is not the favourite in the ASIC race.

Of course their competitor would say this .

[Garr255]

3. ASIC mining proliferation will make Bitcoin much more resilient to 51% attacks and much more secure overall. A few years from now it likely that most governments will not be able to mount a successful computational attack on Bitcoin and before long even the most powerful nations will not have such capability either. Thanks to mass proliferation of various hardware incorporating Bitcoin mining ASIC's.

Government X can buy as many FPGAs and GPUs now in proportion to the network hashrate as they will be able to when everyone has ASICs, and for about the same price, no?

5. FPGA investors will regret buying FPGA's instead of just buying bitcoins. Particularly those who are just buying rigs now.

That's me

[mrb]

Absolutely 100% agree with Vladimir's thoughts. ASIC mining is good for Bitcoin.

[friedcat]

3. ASIC mining proliferation will make Bitcoin much more resilient to 51% attacks and much more secure overall. A few years from now it likely that most governments will not be able to mount a successful computational attack on Bitcoin and before long even the most powerful nations will not have such capability either. Thanks to mass proliferation of various hardware incorporating Bitcoin mining ASIC's.

I agree with all your comments but this one. Now if the government wants to launch a 51% attack, they buy GPUs or FPGAs, and in the future they just buy ASICs. The difference lies between how much the government wants to spend and how much the rest of the miners want to. The technology is irrelevant.

[Garr255]

The technology available is irrelevant.

That's what I was trying to say.

[DeathAndTaxes]

I disagree.  It will happen eventually.  As long as ASIC is not deployed it remains the "cheat card" for entities who can pay the upfront cost.

Want to 51% attack Bitcoin.  It would be stupid to buy 100,000 GPUs.  Spend a couple million building an ASIC processor and 51% attack it on the cheap.  Bitcoin will never be safe from deep pockets (govt, banks, other entrenched interests) as long as there is a "buy your way to victory" option.

Good miners having access to ASIC means the "bad miners" have no competitive advantage.  It comes back to a battle of sheer numbers.  It also has the advantage of killing off botnets and rogue system admins.  They simply can't compete.

[Bitcoin Oz]

It depends if one company has ASIC and just mines itself it would be dangerous as opposed to distributing the units around the community.

[Vladimir]

3. ASIC mining proliferation will make Bitcoin much more resilient to 51% attacks and much more secure overall. A few years from now it likely that most governments will not be able to mount a successful computational attack on Bitcoin and before long even the most powerful nations will not have such capability either. Thanks to mass proliferation of various hardware incorporating Bitcoin mining ASIC's.

I agree with all your comments but this one. Now if the government wants to launch a 51% attack, they buy GPUs or FPGAs, and in the future they just buy ASICs. The difference lies between how much the government wants to spend and how much the rest of the miners want to. The technology is irrelevant.

Thank You.

I, however, disagree with you on 3. Let me give you an example.

If, say, NSA would want to attack Bitcoin right now, they will not be buying GPUS's or FPGA's. NSA has it's own foundry and should they want it they can produce large amount of ASIC chips relatively quickly. This means that they have "generation" advantage. They can fight GPU's with ASIC's. This window of opportunity is closing up quickly with mass proliferation of ASIC mining hardware. Meaning they will have no "generation" advantage. This all of the sudden will be not GPU v ASIC fight but ASIC vs ASIC fight.

[friedcat]

I disagree.  It will happen eventually.  As long as ASIC is not deployed it remains the "cheat card" for entities who can pay the upfront cost.

Want to 51% attack Bitcoin.  It would be stupid to buy 100,000 GPUs.  Spend a couple million building an ASIC processor and 51% attack it on the cheap.  Bitcoin will never be safe from deep pockets (govt, banks, other entrenched interests) as long as there is a "buy your way to victory" option.

Good miners having access to ASIC means the "bad miners" have no competitive advantage.  It comes back to a battle of sheer numbers.  It also has the advantage of killing off botnets and rogue system admins.  They simply can't compete.

If, say, NSA would want to attack Bitcoin right now, they will not be buying GPUS's or FPGA's. NSA has it's own foundry and should they want it they can produce large amount of ASIC ships relatively quickly. This means that they have "generation" advantage. They can fight GPU's with ASIC's. This window of opportunity is closing up quickly with mass proliferation of ASIC mining hardware. Meaning they will have no "generation" advantage. This all of the sudden will be not GPU v ASIC fight but ASIC vs ASIC fight.

Yes, I was wrong on my reply. I agree with both of you in this aspect.

[Definit]

rich get richer?


oh... ok.

[jjshabadoo]

Good responses, but can someone actually frame a sound economic argument as to how ASIC will profitable for both vendors and miners who buy them?

It just doesn't work no matter how you slice it.

It is estimated that it takes approximately $1 million to develop ASIC technology for mining correct ? Yet the cost to produce the units after the development phase are extraordinarily cheap.

So let's say a company spends $1 million and develops 50 GH/s units. In order to be competitive with let's say, the mini rig at $15,000 for 25 Gh/s, they sell their ASIC units for $25,000.

Of course they want to recover their $1 million investment and some profit so they take pre-orders for 50 units. They receive $1.25 million in revenue and let's assume they deliver the 50 units within a 30 day period, once the units are ready to be shipped, to be fair to these customers. Now 2.5 TH has been added to the network within 30 days. Currently the network is approximately 12 TH so we have a 20% difficulty increase right away.

At the current difficulty you get about 900 coins per month with 50 GH at current prices this about $4,000 USD per month minus power costs which by all accounts are extremely small so they are practically irrelevant. This looks good for early adopters as they are thinking 6 months and its all gravy.

Well not so fast because within 30 days, 900 coins became 720 coins with 20% difficulty increase and profit becomes $3200 per month at current pricing. Never mind that hoarding becomes less desirable as people want to get their 25k back(very few people will hoard/save after they plop down 25K) so there is certainly some downward pressure on pricing.

Also ASIC's are unlikely to be deployed before the block reward halves, so 360 coins per month is more like it. For arguments sake lets assume it doesn't matter because coins are now $10 each. Then the ASIC developer sells another 50 units at $20,000 because demand is lower and well hell, they can produce these things for dirt cheap relative to their initial investment. So they pocket $1 million more and are smiling ear to ear.

Unfortunately, network hashrate is now 17 TH for a 30% difficulty increase from 12 TH and people are now getting 630 coins or 315 after the halving. So now at $10 a coin they are getting $3150 per month. Original buyers are approximately $19,000 in the hole and second gen buyers are $16,850 in the hole.

Okay so you say, " Well GPU people are exiting the game en masse " fair enough, but at 50 GH per unit it only takes 240 units to displace the original 12 TH assuming those were all GPU miners.

Bottom line is, demand will slow for the ASIC increasing downward pressure on pricing per unit as people become hesitant to put up 20K, yet the manufacturer has now made a substantial profit so why the hell wouldn't they start selling them for maybe 10k or 5k ?

It wouldn't be out of the question to think that 300 units could be sold in 6 months easily creating a 15 TH network assuming all GPU and FPGA miners leave. Which won't happen because FPGA people also have inconsequential power costs so they stick around hoping to get some return and some GPU people will stick around because they have super cheap electric or don't pay at all.

Once can see that the 25K people will never get their money back as the network would easily swell to 25 TH in 6-9 months.

If competition enters the market this accelerates the process.

Now you can argue that the ASIC people are benevolent bitcoin benefactors and they choose to roll out these rigs in a controlled manner at lets say only 5 per month... AND PIGS START FLYING.

[jjshabadoo]

I don't disagree that ASIC will be good for securing the network, but how does it make rational economic sense as to how they are rolled out? This is the problem. Unless people here think folks will be willing to essentially "donate" their ASIC investment to the network without a hope of getting their money back.

This is absurd since I don't think anyone here can point to too many or anyone at all who has donated 25k to the bitcoin project to date.

The free market works fine if everyone in the market is rational, but miners are often not rational and the rush to become an "early adopter" seems to be an irresistible temptation.  As can be observed from the 6 month interest free loans that many gave to BFL for their singles.

[Bitcoin Oz]

I don't disagree that ASIC will be good for securing the network, but how does it make rational economic sense as to how they are rolled out? This is the problem. Unless people here think folks will be willing to essentially "donate" their ASIC investment to the network without a hope of getting their money back.

This is absurd since I don't think anyone here can point to too many or anyone at all who has donated 25k to the bitcoin project to date.

The free market works fine if everyone in the market is rational, but miners are often not rational and the rush to become an "early adopter" seems to be an irresistible temptation.  As can be observed from the 6 month interest free loans that many gave to BFL for their singles.

What if they dont sell hardware but simply mine themselves ?

[Vladimir]

jjshabadoo, how is any of this different from CPU, GPU and FPGA generations, except in magnitude?