Topic: [ANN] A public company will build a huge Bitcoin Mining Operation (ASIC). - page 2. (Read 27070 times)

yeah i get that but guess what happens as soon as they drop out? you failed to mention it too.

I would like to thank everyone who responded in this thread and elsewhere, be it compliments, suggestions, criticism (constructive or otherwise).

At the moment I am having a series of meetings with investors, accountants, lawyers and other "professionals" trying to figure out subtle details of "the plan". As soon as there is some better clarity on our exact plans it will be announced. Until then I will be mostly quiet.

Thank You all, again. Your feedback both positive and negative is very much appreciated.

i'm speaking on the straight math, no assumptions, no predictions.  nobody knows the full impact if this company ever launches.  nobody knows what gpu miners might do (like selling their gpu's and buying fpg's) or if a retail asic option would become available, so safest not to make any assumptions.

as to people generally realizing that you have to double current hashrate to gain 50% of the hashing power, numerous and repeated postings on these forums speak otherwise.  you most often find it in a statement like, "well if he got 1.2TH hashing, he'd have 10% of the network.".

This is what makes a private ASIC farm extremely risky. Building ASICs to sell to the public on the other hand is far more profitable and much less risky.

It's the kind of thing I expect people to realize without mentioning it explicitly, and it doesn't change my point.

This isn't right either. The answer is somewhere in between. Take an extreme case...

If 50% of the network is on the borderline of dropping out (at the threshold of profitability), then you can look at the current hashing power and do a straight percentage. i.e. if there are 10 TH of miners to start out with and at least 5 TH of miners cannot tolerate any difficulty increase without dropping out, then you will only need 5 TH to 51% the network.

In reality, it is not reasonable to assume that 50% of the network is just at the threshold of quitting. It is also not reasonable to assume that difficulty can double without leading to any exit from incumbent miners. To 51% a 10 TH network, you need to add an amount of hashing power somewhere between 5 and 10 TH.

He is describing an equilibrium effect which significantly decreases the amount of hashing power needed to carry out an attack. For example,  if there are 10 TH currently, you won't need to add another 10 TH to carry out a 51% attack. 7 TH could be enough if increasing difficulty to 14 TH level causes 3 TH of miners (30% of the starting amount) to drop out. That is what he is saying.

The actual numbers are hard to estimate, however it is certain that you won't need to create a full new 10 TH to 51% an existing 10 TH network. True attack costs are less than people naively assume.

don't forget that to gain x% of the network hashing power, you can't look at the current hashrate, and do a straight percentage.

for example, to gain 50% of the network hash power, someone would have to have a hashrate equal to the current entire network.

The beauty of bitcoin is that if such an operation causes panic amongst the majority of hashing participants then acceptance of change namely in the hashing algorithm will be wide spread thus a new version of bitcoin would become mainstream... as most everyone has no access to asic tech then I'd suspect thats exactly what would happen.... interest as asic cannot really be re-tasked to another job that makes for a massive risk to investors.... I'd not want to bet a lot of money that massively relies on guessing the outcome of human nature :-)

point is, you said 'difficulty will double' causing lots of miners to drop out, but failed to mention this would cause difficulty to drop again.

... and that will be the day a new alternative cryptocurrency will be born, so they can participate again.

If you have a warehouse full of toys, but no one to play with them, do you have any toys at all?

Wow, some economic reasoning. I likey!

Watch out, the sky is falling!

No, they wouldn't. We've already seen such a drop, and the difficulty is still well below what it was about 7 months ago. It will fall sharply and then level out before it starts going up again very slowly (unless it kills bitcoin completely).

and then difficulty will drop again, and lots of miners will start mining, and then...

100% is very unlikely if he doesn't use the >50% to invalidate everybody else's blocks. 70% is probably fairly easy, though. If he adds the same as the current amount of hashing power the difficulty will double, and that will make a lot of mining operations unprofitable. Lots of miners will stop mining, and then his market share will increase to 60 or 70% without him needing to add more hardware.

This will probably also lead to fear about the future of Bitcoin. The price will fall, and then it will become unprofitable for even more miners, and his market share will be even larger. On short, I think anything more than 15-20% will put the future of Bitcoin in great danger.

Sure he is.

Did you not know that everybody on this forum is me

In fact, it is just you ( terrytibbs ) and my 10 000 sockpuppet accounts on here ! It is the Matrix !

Not everybody that puts spaces in their lines is me ...

If the other currency also uses proof-of-work, like bitcoin. It would be a dead letter. Monopolist could control both simultaneously with negligible additional cost.

*Bit*coin is based on SHA cryptography and this is used in a LOT of other rather 'normal' and rather 'national security sensitive' applications.

If the hashpower required to build the BTC economy into the multi-billion dollar size you're envisaging was actually *built* - i.e. your company built a vast number of highly powered specifically-optimised SHA processors...

...I'd imagine that you'd swiftly come under the scrutiny of the 'No Such Agency'. Not being immediately alarmist, but the current implementation depends on a technology that is also used to protect the security of mobile phone communications, etc. It's widely used and even little old *me* knows that my little FPGA array (less than 50) plus my GPU-based 9 Ghash could be useful for cracking mundane, everyday encrypted communications.

Think about it. If our combined intense focus on optimising the SHA algorithm finds a flaw in it, a LOT of commercial encryption infrastructure comes crashing down. That's highly unlikely, but if you end up rivalling the spooks for brute-force cracking power, you're not going to be ignored. I'd imagine that the spooks already *have* large-scale brute-forcing equipment for all commercial encryption technologies (it'd be insane for them to *not* have, certainly in the toxic 'surveillance society' the USA and the UK have developed). In business, it's wise to choose one's adversaries carefully. I'd not want national security agencies (especially not from psychopathic governments like the USA) as my main rivals...

So, to answer your question directly; Yes, I reckon *Bit*coin will remain 'garage' - maybe 'big garage' as well as small enterprise - but not trillion-dollar alternative currency. Purely because of the implementation. Maybe I'm not as ambitious as you - I just want to run a macro hedge fund in the City - but I see Bitcoin as a stepping stone to future cryptocurrencies and what will finally be *accepted* by all players as means of value exchange. For the now, I can see myself making a few thousands or tens of thousands if I'm lucky. Worth it for teh lulz.


Where it *really* shines is as getting in early for what will succeed it. As a thought experiment, let's say that the dependence on a commercially widely-used cryptographic function, and the prospect of companies like yours building huge dedicated silicon that could brute-force said widely-used function, frightens governments into either interfering technologically, legally attacking companies like yours, or illegally attacking the BTC economy (bitcoin users are terrorists! etc. ). The rational response for the Bitcoin network would be to replace the core crypto function with something else - the P2P infrastructure and decentralisation has already been proven to largely escape government control. Now, replacing the function used has immediate impact on the money supply, aka the miners. If the chosen function is somewhat like Litecoin's, then GPU miners (at least the AMD ones) become less competitive. I haven't studied enough to analyse whether the popular Spartan-6 FPGAs could feasibly compete with cheap Intel CPUs on this particular function, but I make the general observation that an FPGA miner has more chance of maintaining his/her share of total hashpower if the core crypto function changes - just write a new bitstream (easier said than done, lol). However, all investment in ASICs would be lost if the core function had to change. You can't run the scrypt function on dedicated SHA silicon (IANAE - I'd greatly appreciate any real engineers correcting my mistakes) - and without its legal, original (bitcoin mining) application for your huge SHA cracker warehouse... your company is going to need to *find* an application for all that hashpower to stay in business. I can't think of many legal options, short of contracting to a government to spy on its citizens' GSM comms. I'm not sure your personal ethics would be comfortable with this.

I'm not trying to spread FUD at you, Vladimir. I got a glimpse of your character back in the MMC days, I like you and I wish you the very best. Also, you're not the only one desperate to acquire dedicated silicon to gain the hashpower advantage, it's just that most of the others either keep quiet or don't have your marketing skill. I see you as more of a Kim Schmitz/Dotcom of Bitcoin - and I'm referring to his positive traits, take it as a compliment. However, it's very early days for global cryptocurrencies, and locking your investment down by building single-function silicon is very high-risk indeed. Remember that the Bitcoin design allows for alternative core crypto functions (otherwise things like Litecoin wouldn't exist so easily) - and I doubt that this architecture came about by chance. My analysis is that flexibility is essential - the architecture of "%coin" is flexible, so to dominate the money supply, one must also be flexible. Indeed, given the threat of %coin to fiat currencies, the risk of Bitcoin needing to switch its core crypto function is non-negligible, unless one takes the pedantic view that the 'Bitcoin' name equals its crypto implementation, and a change would not be 'Bitcoin', as per Litecoin etc. - but I'm rambling off the point.


To extend my answer to your question, however; No, I do not think that "%coin" will remain 'garage'. I truly believe that some form of decentralised cryptocurrency will persist, that there will be multiple blockchains and one dominant form. Right now, Bitcoin is the dominant form. However, because this is truly a disruptive technology, and disruptive against *traditional finance* - whose puppet-masters control both global banks *and* government policy, it won't simply take control of a largely-empty market like Microsoft and Google did. Largely-anonymous, distributed money transfer goes against the fundamental reason why the elite (I won't get into names or theories) have power. Hence, rat-like, it must evade centralised ownership and control. It's already been designed this way. But at some point, that internal flexibility may be required. Basing the money supply on farms of ASICs is not coherent with the architectural philosophy, and not only exposes a significant point of failure, but in the most paranoid scenario, puts you and your investors at risk. Again, this isn't FUD directed at you, but *anyone* setting up dedicated silicon with the current crypto implementation aimed at *dominating* the money supply.

If I could be sure that the market-leading FPGAs we currently use for BTC mining were flexible enough to be efficient miners for the most likely alternative (i.e. academically-sound, field-proven) crypto functions, then I'd consider investing in a 'proper' FPGA farm (my own FPGA farm is, as you say, 'garage') with both admin staff and expert engineers on hand to write bitstreams as required. However, I don't know enough yet to take that risk, and I don't have the capital to have enough skin in the game to be taken seriously. Maybe you do?


Lastly, and apologies for such an expansive reply, but the big question that hasn't been answered is how governments, especially the Western ones with enormous debt and no way out but inflation or default, react to people transacting in a currency they cannot control or tax. There's been alarmist, hysterical media reaction (e.g. Silk Road yadda yadda), but no obvious government response yet. That's why I'm advocating flexibility. Being slightly flippant, you may be wasting your money building an ASIC from scratch - the spooks certainly already have such silicon, so finding where the design (or copies of the design) are kept may save some cash. Or even hacking government systems to mine Bitcoins - now that would be exquisitely ironic, turning the centralised surveillance systems against themselves to provide money supply for a distributed network...


I applaud your ambition though Vladimir. Please consider constructing share classes that pay dividends out in BTC though, and make that clear from the start... a company that clearly needs to keep the BTC economy expanding to pay its investors is a DAMN sight more friendly to the BTC economy than one that takes in fiat, mines all the BTC, and sells them all to pay out in fiat. The value of the BTC will crash in such a circumstance, as I'm sure you're aware. I'd buy shares that pay out in BTC - I'm aware this is a private enterprise and you're currently gauging community reaction, but I'd appreciate being 'on the list' for small investors when you *do* finalise the corporate structure.

It's times like these that make me think we could do with a BTC corporate bond market...

Are you bulanula?