Bitcoin Forum

Hi,

60gh per box @ 300 boxes = way more money via bitcoins then selling them at $1299......

just a little common sense says something is not right here...why sell when you could make WAY more by mining them yourself.

60ghashes per box @ $1299 cost = 60000 million hashes = $6100 per month at current bitcoin exchange rate. $6100 per month per box * 300 boxes = $1.8 million.

LOL.

That's how much they could make at the current exchange rate and hashrate over the lifetime of the product. Of course they want to sell them, they make more money up front and let others incur the risk of competition. Then they can spend that money on NRE for better chips or whatever.

There's been a big debate on this. Basically it makes no sense to sell cutting edge ASICs. But there's a lot of self-delusion and wishful thinking that makes a lot of people on here convinced that that they can buy them.

I don't know the exact reasons, but I can think of a couple possible ones.

Your numbers don't account for the rise in difficulty due to ASICs. The income will probably end up being much less than your estimates.

Manufacturing ASIC mining rigs requires a lot of money upfront. A bank is much more likely to loan to someone with pre-orders than to someone that is going to mine bitcoins.

I think if one group was basically in charge of the network it might have a major impact on bitcoins and what they are worth.  Distributing the network goes more towards a free market idea, and a decentralized network.   

Think about this for a moment.  If they did start mining themselves with all of there hardware, and just kept producing more of it for them to mine, would you compete against them with your fpga or gpu?  They would kill the market and the network. 

That all being said if I were in their shoes I would have kept the first unit and turned it on to make a profit so they wouldn't have to be so weird about preorders and preorder funding for the second batch.  If I remember reading correctly the 1 unit given to Jeff makes about $240 a day.  They probably could have mined for a 2+ weeks, and proven they had the product while they were producing the remaining 300.  Once all of them were produced then they could ship all of them out to people at once and not given anyone other then themselves a distinct advantage (aside from geography and how long it takes you to receive something from China).

Many people have asked this. It really comes down to how you want to run your business and what is growth. Do you want to be a miner. Or do you want to be a developer. I am pretty sure I read somewhere that Avalon would like to sell to a wholesaler here in the stats. This way they can focus on production/development and let someone else deal with the orders.

The early ASIC miners will make a lot of money, the mid-pack ones will profit modestly and those late to the game will not even get their initial investment back. I am sure there are still people buying GPUs even today for the purpose of mining. Avalon and other ASIC manufacturers probably expect to be able to continue to make money selling to those late comers.

Why do those latecomers buy? They could be just stupid, or they could be motivated for reasons other than profit. For example, someone who already has tens of thousands of BTC might decide to buy 1 ASIC device simply to do his/her part to protect the BTC network, thereby protecting the value of their pre-existing stash.

As others have said, this income would only be for the first month, as difficulty will skyrocket immediately. So your income will instantly decrease. If you sell the coins on the market, the price will probably decline as there aren't enough buyers for the majority of the coins you're trying to dump now. If you hold the coins and slowly put them on the market (similar to diamonds), you can artificially keep the value of bitcoin really high and get maximum profit from each one sold, however, it means that your wallet will be forever expanding as you can't sell them as fast as you make them.

Additionally, if the network realized that there was 1 entity (especially one that they paid money to for a product) that controlled most of the network, I have a feeling that many people would just walk away. They'd let bitcoin die with their preorders and that'd be the end for them. There wouldn't be any trust (or very little) left in the system if someone took over the network like that. Every new bitcoin company would be scrutinized thoroughly to the point that legitamate companies won't be able to startup.

In effect, they'll be killing their own market. Currently (and probably for several months), they can sell each unit at $1500, which means a constant return and constant profit. Whereas if they mined with each unit then their ROI for each unit produced will take longer than the last. The first batch of units will pay for themselves in a month or two, but the next batch will take 4-5 months, and the next will take 9-10, even with the first units adding their hashes to the later batches. Eventually you'll never make your ROI. Or, you could, as another user said, pass off this issue to customers and let them fight over ROI calculations while you continue to sell at your fixed cost and make your fixed profit.

By "There's been a big debate", do you really mean "I asked the same thing a month ago (https://bitcointalk.org/index.php?topic=134236), and I still havn't learned a thing since then" ?

Why sell when you can mine with it? Because you are selling it at an (almost) unfair price to everyone but the first few customers.

Because during the gold rush, it was the hardware vendors that made all the money, not the miners.

Utterly bogus analogy.

Maybe Avalon (and maybe others) DID mine, enough to cover costs. For me, to see network hash rate spike in the end of December, and in general hashing raising now when we all expect ASICs, it's a bit strange.
Also a little funny is the hashing power of Avalon batches. Around 20TB 1st batch, 40TB 2nd one.
1st batch it's 50% of the network (GPU+FPGA+Avalon 1st batch), 2nd again 50% of the network (GPU+FPGA+Avalon 1st batch+Avalon 2nd batch). This when demand was grossly greater. It's like they took care not to have 51% (and seriously affect trust in bitcoin).

The analogy with a goose (https://bitcointalk.org/index.php?topic=134264.msg1431275#msg1431275) was funny but bogus as well. There is a difference between having a goose and having a concept but no money to realize it.
One has to make a deal (money for the effects). You never know for sure that you will get the bitcoins earned by hosted devices or the device itself or, say, a TV bought by internet - it depends on honesty of the seller and on ability to punish him in case of no delivery.

Why do you think they aren't mining themselves? What would make you think they wouldn't claim to make 300 machines, charge preorders work 2x that amount and actually build 600 machines? They need someones money at the start to build them in the first place. No reason to think they aren't mining themselves as well.

You're right!  You should go into business making ASICs and keeping them instead of selling them.

Of course they could do this, then they will risk of disappearing from the business, when the competitors who do not mine by themselves showed up

Imagine that after one month, people discovered that Avalon was mining with those machines themselves, then who is going to take a future order from them again?

Their batch 2 order only get 45 unit sold, could be partly due to this doubt from batch 1 customers, seems no one except the first 2 people have received an unit from them

How much did it cost to design/fab the asic chips?  The $1299 price allows them to make a profit???????? 

So many responses and yet, nobody has given the correct answer...

You have to realize that the Avalon team is not a large company with lots of money. They did not have the $200-300k to cover the NRE costs to design even just a handful of these 110nm chips. They had no choice but to pre-sell 300 devices in order to get funding. And now that the chips are made, Yifu and ngzhang are obviously ethical people who are not going to steal customers' money by deciding to keep the devices and mine for themselves.

hi: yes, 300 units at $1300 does seem to cover the $200-300k NRE cost from TSCM + whatever other overhead they have.

If one company suddenly started mining half the coins, it would most likely cause a large crash in market price. That and Avalon equipment may not be very profitable, if butterfly meets their specs.

If they mined all the new coin themselves, as soon as they hit 51% we would know and would abandon the coin, making it worthless. There is actually a LOT more money in selling hardware. Especially if they get these products to you and your neighbors, relatives, and friends see you are making money, then they will also buy more products. Anyone in the first and second batch is likely going to see very quick ROI, especially if the price increases as the demand remains the same but the supply decreases. Third batch and BFL customers may also see some ROI. Where it's going to get crazy stupid to buy a miner is when people who don't even know what Bitcoin is right now are here and on the forums 3 months from now trying to figure out who to give thousands of dollars to "make them rich".

IMHO, people that don't understand why companies like BFL and BitSyncom are selling the hardware right now seem to believe everyone is rational just like them. People are irrational. You are taking a risk buying miners. For some, this is an easier choice than others. After all, what is 75 BTC when you own thousands?

If you had any question about whether or not old miners were buying Avalons:
http://blockchain.info/charts/bitcoin-days-destroyed-min-month?showDataPoints=false&timespan=all&show_header=true&daysAverageString=7&scale=0&address=

The logical approach for a manufacturer is a split between mining operations and selling units. That pretty much the best way for everyone as well. That covers capital and startup costs and reduces financial risk.  I wouldn't believe any manufacturer that said they were not going to mine or have a sister company/associate mine. Normal business sense would determine that as the path but maybe logic business sense doesn't apply here.

Because it wasn't about the money?

Because Avalon were worried about one of the other ASIC developers creating a monopoly, so they developed their own and low and behold, they are the monopoly! Much to their disappointment.

/thread

                                       ^^^^^^^^^^^^^^
Just read all the arguements and this is the best one.

Keep as many ASICs as possible to mine with and sell just enough to keep the masses onboard and Bitcoin afloat.

This is actually a test for the manufacturer's value. I think one year ago they would never expect the price of BTC has risen to a level that it even becomes attractive for them to mine the coin by themselves. Then the manufacturer who do not mine by themselves and distribute the machine evenly will gain the most orders

It's a bit pitty that bASIC quit the game, otherwise the playing field will be more even to ensure the competition, but I heard that ASICminer is going to deploy 800 unit this month, so the difficulty will rise quickly anyway no matter those manufacturer mine or not, their future order purely depend on the quality of their current delivery

maybe because cash today versus maybe cash in two weeks?

>Bitcoin

>Not about money

Miners risk many things... the network hash rate could explode with 4 vendors releasing ASICs at the same time, the price of bitcoin can crash, theft of hardware, broken hardware, fire/flood, etc.

Since they can still make a sizeable profit with absolutely no risks, why not do it?

PS: If you're ordering a lot of ASICs, make sure to get insurance. It will help you sleep better at night!  ;D

"Why would anyone develop and release a GPU miner for free?  Anyone capable of writing GPU mining code could make a ton of money by mining themselves while everyone else is stuck using CPUs.  A little common sense says that these GPU miners can't be real.  Anyone downloading one is just asking to end up with a wallet stealing trojan."

The world is full of people with all kinds of motivations.

lol..terrible analogy....

Writing code for existing mass market hardware is trivial compared to designing/fabbing hardware that is specialized in doing one task.  

Consider that 300 of these would make a nice profit if mined by one entity.  

The greedy algorithm usually covers most human behaviour..especially in the bitcoin world....only the naive or stupid believe otherwise.    

It appears that after all the reasons listed in this thread most still fall under "The hardware store that sells the shovels makes more money then most of the miners"

I put this link in another thread but i think it belongs here.

http://www.btcmu.info/articles/128-miner-speculation-avalon-s-look-to-the-future.html

Currently I believe that they have to mine for themselves to earn the money to finish the production of the first batch  ;)

They will first make 5 units and then use these 5 units to mine and buy components to build another 10, and then use these 15 units to mine to buy components for another 30, etc...