Avalon ASIC Bitcoin Miner – Prices for Batch #2 | Hacker News Reader

Avalon ASIC Bitcoin Miner – Prices for Batch #2

launch.avalon-asics.com

28 pointsDiabloD361 comments

It would seem to me that a winning strategy would be to make these units, sell them, and ship them.

In the delays for step 2 & 3, have them sitting around plugged in mining, paying for their use. The final sale defrays the cost of building the thing, and you effectively get free Bitcoins--right before the market for mining tanks due to your dispersal of mining ASICs.

Is this reasonable?

It seems strange to even sell these units at all right now. If they really can pay for their (retail!) value within a week, the real money is in running them, continuously, until the difficulty gets so high that they can only pay for themselves within, say, 6mo or more.

It may be that the competition has driven manufacturers to attempt to sell so they can offload their stock. But Avalon claims that they are made-to-order.

If it were me, I would've built 1000 of these, made them pay for themselves and then some, then started dumping as soon as it looked like a competitor was about to launch. And again, this sale is too early for that to make sense.

They cannot afford to make 1000 of them and keep them for themselves. They need the money up-front, and to get that they need to sell them in advance.

Yeah, this. They're crowdsourcing the funding for the NRE and associated costs of spinning an ASIC. They're also diversified a bit by being both miners, and shovel-merchants.

That must be it. The money is in keeping a significant portion of what they make for themselves, and using the sales to fund production of even more equipment, for themselves.

Then, eventually, these things will be worth very little but the money will be made.

I would imagine this will drive BTC prices pretty far downward. A few guys with a couple hundreds BTC mining ASICs can monopolize the vast majority of BTC being mined. And given the current price, they might want out soon.

I'm starting to understand it a little better after studying it a bit. I don't think it will drive BTC "prices" down because there is a finite amount of the currency that is possible to be created. I don't think it can create a glut (I could be wring on that). There are a couple of incentives not to hoard all of the computing power.

1. Benevolent users want to avoid the "51%" attack. It could undermine confidence and perhaps cause users to flee.

2. It's possible that we'll see a hashing power arms race where several cycles of ASIC production supersede or render the previous generation unprofitable. While it may be possible to go it alone and win, it's still a risky prospect, for the above reason, and for the risk associated with ASIC development. Therefore, ASIC developers have an incentive to distribute the machines into the hands of BTC users at large, and quickly.

These ASIC miners are going to have a limited profitability lifetime. While it may be years before they are uneconomical in terms of the cost of the electrical power to run them, they could be rendered barely profitable by a faster generation of ASIC based machines, or by an unfavorable turn in Bitcoin exchange rates; which could happen on a time scale of months or even weeks.

You've just described how TV infomercials work. 1. Play the commercial on tv. 2. Sell the product and collect the money. 3. earn interest and find the cheapest manufacturer to do JIT manufacturing in the 6-8 weeks you have to ship the product. 4. Ship the actual product.

Wow, that's actually a lot like Kickstarter.

If they build many machines, and use them all to mine at the same time, aren't they suddenly crashing the value?

I guess if they get a good deal for energy (or need to heat the factory) "soak testing" is a good way to get a bit of money back.

Oh, they very much may be--but that's a problem for the purchaser. Presumably they've already been sold and spoken for, money having already changed hands. A slight delay of a few days in shipping--who's to say?

crashing the value of what?

Bitcoin price is not tied to computational effort spent on mining. The creation rate of new Bitcoins is already constrained and well defined. Producing more machines would just mean Avalon would get a larger portion of the new Bitcoins.

That's the thing - this is not a traditional market but it is possible to 'crash' the difficulty. If enough of these things show up globally they can make mining so difficult for everyone else that they will have to give up.

This is a very unhealthy thing for the BTC market in general as only a few players will have the ASICs to mine profitably, and everyone else will be spending far too much per kW to do any mining at all.

One of the weaknesses of BTC, as far as I understand, is that anyone with a majority of the global hashing power can rewrite new transactions on the blockchain. If regular GPU miners give up and a few players spin up some massive ASICs, we may very well see this become a reality.

Crashing the value of the machines?

If they're that good and the manufacturers put a load online themselves, then the folks buying one are not buying as big a slice of the pie as they thought they could?

But the amount of computation required to get a new bitcoin is definitely tied to the total computation being applied to mining.

The launch was a disaster again, with most requests timing out, carts going to checkout only to ask for a few USD of payment instead of the full $1500, and a report a of a customer seeing other customer's name.

Avalon has insisted on using a different payment processor for each of their batches, I assume as way of supporting the community. This almost ensures there will be issues with their store every batch.

Great product, lousy store.

What do you do to get rich during a gold rush? Sell shovels to miners.

Would you sell shovels to other people if the shovels mined gold on their own?

The real reason that the ASIC companies take preorders is to pay for the high upfront engineering costs. Additionally, a single entity controlling a majority of the bitcoin network would decrease confidence in Bitcoin, so it is in everyones best interest to spread out the mining devices.

I like the idea of Bitcoin, but it bothers me that the computational power gets spent on such a useless project (hash function calculation). Isn't there anyway that the system of Bitcoin could remain the same while the computational power goes into something like... I don't know... protein folding or weather prediction?

No one seems to have proposed such a system, either for Bitcoin or for earlier proof-of-work systems (designed for fighting spam, that sort of thing). How do you come up with a decentralized secure system with verifiable problems which can come from anyone which they couldn't've computed the answer to in advance or have chosen it to be especially weak but which answer will still be actually useful?

I keep some notes on the topic http://www.gwern.net/Notes#a-bitcoinbittorrent-driven-econom... but so far I haven't seen any proposals which might work.

The hash function is not arbitrary. Its purpose is to put the global list of transactions in an unambiguous sequence that everyone agrees on, to solve the double-spending problem. The inputs to the hash function are transactions.

When a block of transactions is successfully hashed, the miner is awarded new bitcoins, as an incentive to do that necessary work of putting those transactions in order.

I was worried about energy consumption, until I worked out how much the numbers. Starting with an extreme assumption that the bitcoin money supply reaches parity with the dollar by 2021 or so, total power consumption used directly for mining is only a couple gigawatts, about a tenth of a percent of the world's total. I worked that out like this:

First assume the cost of mining equals the value returned. If mining is very profitable, more people will mine and difficulty goes up. If mining loses money, people drop out and difficulty goes down. Overall mining will always tend to be in the neighborhood of breakeven, with just barely enough profit to keep people from dropping out (though that's the equilibrium, and in periods of rapid growth we could be far from equilibrium at times).

Pick a date, figure out how many reward halvings we'll have between now and then, and you get how many coins we award every ten minutes. Multiply by coin price to get value awarded each time. If you assume electricity is half the mining cost, divide that value in half. Then divide by the cost per kWh to get kWh per ten minutes. Multiple by six to get kWh per hour, which is the same as just kilowatts.

And now you know how much power production you need to support the bitcoin economy given your assumptions.

The hash function is completely arbitrary. An infinitesimal amount of the work that goes into a block is actually processing the transactions. After that it's all proof of work, and you can slot in any proof of work.

The problem is that 'useful' calculations aren't suited for extremely fast verification.

I guess that depends how you define it. The only thing it's doing is taking transactions as input, along with a random number, and hashing them. It's just that the result is required to have a certain number of zeros to be considered valid, so you have to do it over and over again with different random numbers.

If you didn't require a difficult computation, anyone could compute a different order of transactions and it would look just as valid. Easier hashing would not accomplish the goal of making a universally-agreed-upon transaction order.

That's why I said you can slot in any proof of work. Anything slow to do but quick to verify will suffice.

Any proof of work with bitcoin transactions as the input will suffice. Generating similar hashes of, say, public astronomical data, wouldn't do the job at all.

You only need to feed in a handful of bytes of bitcoin data. One fake star, or slight perturbations of water molecules in frame 0 of a simulation, would be enough. The problem is that it's difficult to make an algorithm that is both proof of work and actually useful for something.

Many power companies in my area offer free electricity at night to help them shed excess load.

If your power company does that, you should definitely get an ASIC miner and mine at night, since power is the dominant cost.

>power is the dominant cost.

That's only true for GPU/PC based miners.

66GH/s @ 600Watts, Power is 1-2% of mining cost at current Difficulty / $Exchange / power cost ($0.12). Here is a link to info about the "free nights" plan. https://www.txu.com/residential/promotions/mass/free-nights....

The ASIC based system is much more power efficient than GPU/PC systems. It's even better than FPGA based systems which are (~10-20x) more power efficient than GPU/PC systems.

If Bitcoin mining economics are able to sustain ASIC production, it will probably render FPGA based systems uneconomical.

Even with just two viable ASIC production runs, I suspect that GPU/PC mining is dead.

Ok I doublechecked and you're right that hardware is the dominant cost with these ASICs. But power accounts for about a quarter of the cost at $0.12/kWh, according to this calculator: http://www.bitcoinx.com/profit/

Thanks for the link, that's pretty cool...wish my utility did that.

I think you may not be reading it correctly. I get much better numbers even with more conservative parameters. I more than doubled difficulty, since this is definitely going to change soon due to the effect of the ASIC machines. I need to figure out how to figure out a real number for that.

Parameters that I changed:

Difficulty 6,968,775 (I don't really think it's likely to double)

Hash Rate 50,000 (since 100% uptime is unicorns)

Electricity rate (USD/kWh) 0.12 (my current rate)

Power consumption (W) 600

Results:

Coins per 24h at these conditions 3.6083 BTC (~3.6*20)

Power cost per 24h 1.73 USD

Revenue per day 71.70 USD (Gross)

Revenue Less power costs 69.97 USD (net)

It still looks very good to me, am I mistaken?

PS, I wish I could take advantage of the free nights, unfortunately, I expect that if this night time load-shedding problem persists for any length of time (a few years), someone will step in and find a way to arb it out, and any big investment in doing that on my part may be lost.

Some might argue that decoupling currency from central banks and government is not a useless project (or a toy problem). With Bitcoin, participants invest their own resources. It's none of my business how private individuals or enterprises peacefully allocate their resources, so I choose to not be bothered by how they do so.

In general, specialized hardware tends to be expensive (niche uses result in poor economies of scale), which makes it difficult to see more widespread adoption, and consequently makes them less usable by researchers.

If these types of hardware end up finding use in a new problem space (like bitcoin mining), then in theory that could make them more popular, and thereby cheaper (as well as increasing the acceleration of the technology, leading to better, faster, hardware for everyone who uses them).

So in theory, even if Bitcoin mining doesn't have any "direct good" that comes from it, it could still be indirectly helpful to the larger scientific computing community.

If Bitcoin holds up over time and fulfills it's full promise of a democratized currency immune to the influence of corrupt governmental power (which is a big if) then I'd argue that spending a lot of dedicated processing power is a small price to pay.

See this StackExchange answer for an answer to this question: http://bitcoin.stackexchange.com/questions/5617/why-are-bitc...

Some issues with "useful" computational work is that it needs to be computationally easy to verify the result (a hash function is but other work may not be), also it may not be crptographically strong so someone could come up with some method to solve the problems much more efficiently than others and make the network insecure.

It's not useless; BitCoin security depends on the majority of hashing power being in the hands of benevolent users. People who invest in ASICs and other mining hardware have an implicit incentive to be benevolent (because a collapse of the system means they won't see a return on their investment). So developments like this strengthen the security of BitCoin.

You're missing his point. It's not that people who buy ASIC rigs are evil; it's that the whole system is burning energy on a toy problem, instead of addressing a real scientific problem that needs compute.

I think you're missing /my/ point, which is that BitCoin is useful as a currency, and these ASICs are strengthening the system, so by extension they are useful too.

If you think the BitCoin project is entirely useless, even as a proof-of-concept, then yes, these ASICs are useless, but that's an entirely different discussion.

> ... toy problem, instead of addressing a real scientific problem that needs compute.

What, like generating Mersenne primes? A lot of these other so-called “real” “scientific” problems have virtually no scientific relevance. I think research into crypto-currencies is just as “real” as finding more optimal Golomb rulers or searching for alien lifesigns in radio noise.

Circular argument is circular.

Bitcoin's proof-of-work system requires a problem that is difficult to solve, has a predictable and adjustable difficulty, and allows solutions to be verified easily. Many hash functions, in Bitcoin's case, SHA-256, meet these requirements. Protein folding and weather prediction do not.

Can someone describe what Avalon's hashrate: 66 Gh/s means in terms of bitcoin production (or better, in terms of money). How many days would Avalon need to work 24/7 to provide 100% ROI ? I don't know much about bitcoin, I would imagine such info should be the first thing a potential customer reads on Avalon's front page.

It's very difficult to predict how much money these will make as it's unknown by the time you get one how many other ASICs will ship. If you look at https://bitcointalk.org/index.php?topic=89685.0 these are preorders for BFL ASIC that people have reported.

So at the current difficulty you would make $220.35 / day with a 66 Gh/s Avalon. If you look at predictions people are throwing around though on what these BFL orders will do to the difficulty (https://bitcointalk.org/index.php?topic=89685.msg1218884#msg...) then your $/day would go down to $11.42.

The question is will BFL ship on time and how many units will they ship. They claim to be shipping their first batch on the week of Feb 10th. This second batch from Avalon is said to be shipping March 5th – April 5th, 2013. So if you believe BFL will ship before then, then the difficulty will look very different by the time you get your Avalon.

I don't think they could put how much money you'd be expected to make, as that fluctuates in relation to the dollar etc.

But I'd be interested to know what 66 Gh/s means in practical terms too!

Just found this site, that calculates your profit http://www.bitcoinx.com/profit/

Edit: This makes me wonder, why would you release this technology to the public.

I would say it's a safe bet that the developers of the machine has been running it for quite a while themselves and have meticulously calculated exactly the breaking point at which these machines will drive up the difficulty enough that it's more profitable selling them than it is running them for BTC.

If Avalon was running 20 TH/s somewhere, it would show up (considering that the entire Bitcoin network is only 25 TH/s).

My 7970, the fastest single GPU for mining, does 675mh/s overclocked to 1125mhz. Thats about 1/100th the speed of an Avalon.

Ah, nothing like a good old duplicate Bitcoin mining 101 thread. http://news.ycombinator.com/item?id=5142021

Once they become prevalent, they'll make money at the same rate GPUs make now, or CPUs made before, or etc. etc.

There are a lot more miners now, so each miner makes less money. Actually the reward was cut in half a couple of months ago, so even in aggregate, they'll make money at half the rate CPU's used to.

It all depends how you define money. The reward halving seems to be correlated with a price spike in the Bitcoin exchange rate, so although fewer Bitcoins are being mined per block, their total value is actually about the same today as it was before the reward halving.

I do agree that there are more miners now, which lowers profits for individual miners.

right now, like ~8 days. It's really impossible to tell though when these things hit mainstream as they will affect the current difficulty so much.

So it's basically a money making machine. Why would you ever sell such a machine for a one-time profit?

The difficulty in mining will increase dramatically once people start using them.

To use the analogy I've been seeing everywhere, during the 19th century gold rush would you rather have been a miner or the guy selling pickaxes?

Suppose gold is evenly distributed everywhere and you are the only one with a pickaxe and everyone else has spoons?

Consider the fact that these pickaxes cost half a million dollars to make and only come in batches of 1000. Also, there is nothing to stop some other folks from building a similar batch of pickaxes.

At this point, I think Bitcoin just has to survive. If it survives long enough, enough people will have heard of them and start to believe that they're money, it will be a defacto currency. So long as there aren't any disasters that kill its credibility, this will happen in a matter of a few to several years.

Someone explain this to me: if I have extra computing power just lying around being unused (and say I have a lot), should I be mining or what?

No. You need specialized ASICs to mine Bitcoin profitably.

Best business is instead of digging for gold - to sell showels to gold diggers.

yeah but when will these things ship??

Sometime in March. They're already shipping their first batch as we speak.

The site claims March 5th – April 5th, 2013.

Reply on news.ycombinator.com