One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month • Hacker News Board

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#161

Post by alphydan » Tue, Jan 23, 2018, 12:56 AM UTC

whatshisface wrote: post #155

lalaland1125 wrote: post #149

Earlier quoted context omitted.

It is impossible to create a "useful" PoW. The issue comes down to incentives. As bitcoin is a decentralized, trust-less system, there needs to be constant and powerful incentives to prevent miners from cheating (by creating forks, or executing any number of other attacks). The current main incentive against cheating is that miners will lose their invested PoW (compute time + electricity) if they cheat and get their…

In that case, you'd want everybody running profitless basic research scientific computations for other people's research teams. Eventually we may be able to buy cloud compute by inventing a new coin with a proof of work useful only to us, buying some Etherium to back it with (so that there's some initial value to mining it), and then letting speculators put it in there currency baskets to give it long-term value.

> profitless basic research

May not scale in 20 years. Useless research can become very valuable in a decade or two.

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#162

Post by nimos » Tue, Jan 23, 2018, 12:56 AM UTC

jen729w wrote: post #105

nimos wrote: post #54

Earlier quoted context omitted.

The % of the network they would control would be too high maybe? Fundamentally the value of their product is based on the price of bitcoin. A single miner that gets close to 50% might be seen a a huge risk and drive down the price.

Let’s say some party/group gets 51%. Now they “own” the blockchain. Doesn’t its value drop to 0, thereby making the whole effort pointless? I’m not going to buy BTC (or whatever) from that system, therefore the owners take no cash from it. Yeah they have all the BTC, but so what? Everyone loses. What am I missing?

That's my point exactly. They sell them to other people to ensure that doesn't happen.

Now I'm not actually sure if they produce enough for that to be relevant. But my understanding is they dominate the asic market and asics are the only thing that are profitable for mining bitcoin. It could be if they didn't sell and mined themselves they would have enough mining share to cause people to lose faith in the currency and crash the market.

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#163

Post by amalag » Tue, Jan 23, 2018, 1:00 AM UTC

KaoruAoiShiho wrote: post #144

amalag wrote: post #98

Earlier quoted context omitted.

PoW will go away. Noone will entrust their wealth to miners. There are plenty of other ways to achieve consensus and coins are moving away from PoW. But we are still not done with the Bitcoin bubble.

I have a feeling PoS will not be seen as credible as PoW.

There are still more consensus algorithms that PoS and PoW.

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#164

Post by amalag » Tue, Jan 23, 2018, 1:00 AM UTC

gambiting wrote: post #106

amalag wrote: post #99

Earlier quoted context omitted.

Do you see people printing banknotes out of feathers or some other randomly expensive material? No, and similarly mining bitcoin is some randomly expensive process. The process of securing a "coin" can be done very cheaply with minimal computing cost and the world will shift to that.

The issue is not that the process has to be expensive. You are correct - we could generate billions of crypto coins within couple hours and beat bitcoin right away. But people want bitcoin. The same why people are willing to pay the price for a labour-intensive painting that took years to make over an identical reproduction that took 5 minutes to print.

This is now shifting the goalposts. Before it was PoW is necessary to secure the network. Now it is we don't need PoW but it is artisanal.

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#165

Post by wmf » Tue, Jan 23, 2018, 1:11 AM UTC

paulie_a wrote: post #141

wmf wrote: post #113

Earlier quoted context omitted.

The movie plot here is to secretly get 51% of the hashrate, cash out a billion-dollar fraud, then walk away. In reality it's unlikely to happen.

If that were possible, it would be a one time thing I am guessing. But it's possible to short bitcoin now I believe. So ruining the network/reputation could be profitable? I am speculating and don't know for sure.

You probably can't short a large enough amount to make it worthwhile. Also, due to Murphy's Law the price would probably go up.

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#166

Post by Neeek » Tue, Jan 23, 2018, 1:12 AM UTC

banachtarski wrote: post #66

whytaka wrote: post #5

Earlier quoted context omitted.

Distribute porn. Play video games. Or some other thing that has subjective value that you might not agree with.

Not OP but that seems more useful. Even after digging for examples to prove a point, it's telling that this is all you can come up with.

Here's one, people want to use bitcoin for transactions, as these blocks of transactions fill up, new hashes are needed for each block. The miners are finding these hashes and facilitating the continued use of the blockchain. I'm hazier on this one, but I believe transaction fees go to the finder of the hash? Anyway, wouldn't the value be is the protocol/platform itself, that's what people are buying in to, they want a usable, distributed ledger.

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#167

Post by objclxt » Tue, Jan 23, 2018, 1:13 AM UTC

paulddraper wrote: post #145

blackrock wrote: post #129

The amount of waste of Bitcoin Mining is incredible. It is all about virtual digital mining, which is to race to find the first hash, that has X number of zeros in the front of it. And this burns an incredible amount of electricity. And all for what? Just to show proof of work. Can't there be a more useful usage of cryptocurrency mining to show proof of work? Something that is more relevant to humanity. Like protein…

Same is true of gold, diamonds, ivory, other items whose rarity makes them precious.

Gold and diamonds both have industrial applications. They have intrinsic properties outside of their scarcity that also give them worth (for gold, their corrosion resistance, for diamonds their hardness). Even if they had no aesthetic value they would still be mined and sold, because they have practical uses. Is the same true of Bitcoin? Right now, I don't think so.

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#168

Post by whatshisface » Tue, Jan 23, 2018, 1:22 AM UTC

alphydan wrote: post #161

whatshisface wrote: post #155

Earlier quoted context omitted.

In that case, you'd want everybody running profitless basic research scientific computations for other people's research teams. Eventually we may be able to buy cloud compute by inventing a new coin with a proof of work useful only to us, buying some Etherium to back it with (so that there's some initial value to mining it), and then letting speculators put it in there currency baskets to give it long-term value.

> profitless basic research May not scale in 20 years. Useless research can become very valuable in a decade or two.

That's true, but nobody would be able to bet on it being useful specifically to them. (I'm thinking about simulating LHC stuff, searching astronomical data, the sorts of things where at the time of computation nobody has a clue who will end up benefiting. Protein folding might be off-limits depending on who wants to know how it folds, and why.)

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#169

Post by ww520 » Tue, Jan 23, 2018, 1:41 AM UTC

Animats wrote: post #43

On top of that, Etherium mining is eating the graphics board market. The price of NVidia's higher end boards has doubled in the last six months.[1] Etherium was supposed to require a general purpose CPU to mine, but now it's done on GPUs with large RAM. [1] https://arstechnica.com/tech-policy/2018/01/cryptocurrency-b...

NVDA is doing well. God, that company has nicely leveraged their core product to hit on all the growth cylinders.

Re: One Bitcoin miner is buying 20,000 16nm wafers from TSMC per month

#170

Post by stephen_g » Tue, Jan 23, 2018, 1:43 AM UTC

Ologn wrote: post #118

gambiting wrote: post #88

Earlier quoted context omitted.

Why are banknotes worth anything? They are just pieces of paper with some ink on them, after all. They are "worth" something, because people are willing to do work or trade you stuff for them. The same with mining bitcoin - instead of printing banknotes, you make a "thing" that you can then trade for services and objects, because people want to have that "thing"(this part is crucial). Most importantly, people are wil…

> Why are banknotes worth anything? Until 1971 dollars were worth something because they could be converted into one of the thousands of tons of gold bars the US stores at Fort Knox (and elsewhere). Then a magic wand was waved over the printing presses at the US Treasury and they began to attain value for no reason. This magic has now spread to Bitcoin, Ethereum, and even cryptocoins like Dogecoin, whose creator has…

Convertibility has always been based on a fundamental misunderstanding of money, and always fails because it's a flawed concept (and requires fixed exchange rates, which is always a bad idea).

Money has always been based more on representing debt than actually having intrinsic value (that misunderstanding comes from the barter myth, that pre-money economies had markets that used barder, which there is little anthropological evidence for - see Debt: The First 5000 Years for a pretty good overview).

As for now, the US dollar has a baseline demand because all business and income in the US requires taxes to be paid to the Government in US dollars, and all spending by the Government happens in dollars. The rest of the domestic economy emerges out of that, just as market economies always have. The value of the dollar is effectively ultimately 'backed' by the goods and services produced by the US economy. Herein lies the problem with convertibility - why would we expect that the amount of gold the Fed corresponds to the size of the economy? It won't, so the value will always diverge and hence how the system failed (more than once).