Butterfly Labs ASIC Gear Potential

[mc_lovin]

this can't be real.. they are going to kill GPU mining altogether!  at those prices..  3.5 GH/s for $150 is like getting 6990's for $40 and they run on a little solar panel.  This has to be a joke.  They would ruin bitcoin by releasing something like this.

[rjk]

this can't be real.. they are going to kill GPU mining altogether!  at those prices..  3.5 GH/s for $150 is like getting 6990's for $40 and they run on a little solar panel.  This has to be a joke.  They would ruin bitcoin miners that spent too much on GPUs by releasing something like this.

FTFY

[imsaguy]

this can't be real.. they are going to kill GPU mining altogether!  at those prices..  3.5 GH/s for $150 is like getting 6990's for $40 and they run on a little solar panel.  This has to be a joke.  They would ruin bitcoin miners that spent too much on GPUs by releasing something like this.

FTFY

Pretty much.  The people said the same thing when it went to GPUs from CPUs.

[mc_lovin]

Man, the difficulty is going to get so high..  I could see it hitting 100 TH/s by mid-2013 if they are releasing 1 TH/s units at those prices.

[davidspitzer]

Idiot, Amazon dont make money on selling hardware. They make money on services ....aka selling books thro Kindle.

Appl doesnt make money on your I-products... They make money on App Store.

BFL loves idiots like you.

/thread

What's your point? BFL doesn't have an app store, they only sell hardware.

My point is exactly that.

As someone pointed out, it does not make any financial sense for such price point.

Its not like they can have trillions chips to ship.

Anyone thinks the low price point is to completely conquer the market is idiot

because they can conquer the market selling at 95% of the competition.

remember those are prices for SPECULATED performance specs.  Look what happened to the single and minirig

There are 2 main way to conquere the market, only one is used but never both:
+ Lowest price point
OR
+ Most technology breakthro in your products line.

So yes, they're bluffing to fool idiots.

Apparently based on your posts you consider everyone to be an idiot.


In case you were curious:

ad hominem
You attacked your opponent's character or personal traits instead of engaging with their argument.
Ad hominem attacks can take the form of overtly attacking somebody, or more subtly casting doubt on their character or personal attributes. The desired result of an ad hom attack is to undermine one's opponent without actually having to engage with their argument or present a compelling argument of one's own.

You got reading comprehension? I dont believe any of their bluff. And i'm sure i'm not the only one.

You, on the other hand .....

I'm engaging in a thought discussion  - your reply does not add anything substantive to the discussion. This thread discusses a hypothetical - Namely the possible opportunity associated with potentially game changing hardware. There are many variables and unknowns and I value the insights of people on the board to help consider many of the possible permutations. Calling everyone an idiot, while it may be cathartic for you, is not very constructive to the goal of the discussion and unwelcome. If you have some thoughtful comments to make I encourage you to do so. You are correct though. BFL may not be able to deliver or they might have over estimated the capability or the price point they can offer it at. Those are yet to be determined.

The question that I pose is, if it comes to fruition, what is the opportunity?

[davidspitzer]

I am going to grab some lunch  - discuss amongst yourself till I get back

[Vladimir]

If they really going after full custom ASIC's with a price tag of about 30 million $ they would need to sell only about 1000 of those 1 THps rigs to break even.

If so and if they really will ship it to multiply users in one go this could be the best thing that has ever happened to Bitcoin.

Prepare to start measuring Bitcoin hashing power in Peta Hashes.

When these things come out, I'll have an ISO 27001 certified company ready to host these babies in ex nuclear NATO bunkers, with physical security up to military standards and with DDOS mitigation plans in place. Give me a whistle when you are ready.

[cmg5461]

I'm in the process of making a spread sheet to depict ROI for this device.  Looking great already!

[gyverlb]

If they really going after full custom ASIC's with a price tag of about 30 million $ they would need to sell only about 1000 of those 1 THps rigs to break even.

If so and if they really will ship it to multiply users in one go this could be the best thing that has ever happened to Bitcoin.

Prepare to start measuring Bitcoin hashing power in Peta Hashes.

That can't work : there's not $30M generated quickly enough in the Bitcoin economy to absorb this.

Let's talk numbers :

• given the performance numbers claimed, the current network hashing speed, with only $300k-$400k sold, BFL is wiping the competition (GPUs won't be profitable, FPGAs will need more but they can't represent a big part of the network speed in this scenario).
• after that each new unit sold diminishes the reward per unit (assuming constant $/BTC)
• to have an upper limit to how much units can be sold to miners actually profiting from them, simply compute the total amount of $ to be earned in a year mining. It is (at a $6.5 BTC and 25 BTC/block) 8.5 million dollars.
  If you want to be profitable in less than a year using ASICs (which you should if you worry about BTC volatility and the warranty on your coffee warmer/minirig) you should assume that everyone is doing like you and mining with ASICs. If BFL sells for more than 8 million dollars of ASICs, mathematically you can't be profitable in less than a year (that's not even factoring the electrical/maintenance costs in). I'd guesstimate the limit is at $4M worth of units sold. If BTC value is stable, there's only a marginal market for ASIC after this amount sold.
  

So if BFL sells more than several million dollars of ASICs their customers are screwed, if they sell less than their initial investment they are screwed (and their pre-ordering customers waiting to get their units too). That's not very comfortable for an healthy market.

I've doubts about the economical feasibility of significant volume of ASICs right now : if you need to invest more than $0.5M you probably lose, there's not enough value mined to warrant such an investment because the reward of buying hardware after that amount sold will diminish (as there's no more GPU to replace) so your potential customer will start losing motivation and simply ignore your product when you enter the $1-4M volume sold range were there's simply not any money left to allocate to production costs and miner gains (at least currently). You'd have to take a guess at which amount sold (between $500k and $4M) ASICs will become profitable for BFL and at which amount they stop being profitable to miners.

Assuming BFL ASICs actually work, a few early adopters might probably be profitable (assuming the units don't go to fast out of the factory in the hands of customers) and everyone else is screwed (and it only takes 100 SC mini-rig sold to enter in the profitability vanishing zone).

Another possibility is that they don't plan to flood the market and slowly release ASICs to match the BTC value going up. It's quite a bizarre business model because of the risks involved for BFL but the only one that won't hurt miners badly.

And as always : if you bet on BTC's value going up significantly, just buy BTCs and wait for it to stabilize again to resume mining...

[cmg5461]

https://docs.google.com/spreadsheet/ccc?key=0Ai2b4IJuKUZhdE5EVFJkUF95TS1BVjdxaEFsSVlQSkE

you can download the sheet and edit it as you please.

[Vladimir]

gyverlb,

You got me wrong, my friend. My post was not from point of view of a miner. But from point of view of an ASIC manufacturer. All I said is that they need to sell 1000 of 30,000$ rigs 1 Thps each to break even on hypothetical (educated guess) 30,000,000$ costs of full custom chip design and manufacturing. Are you going to argue my arithmetic skills here?

So if you are going to argue with me here you really need to argue the above numbers. I said nothing else except  that 1000 of 1 Thps rigs would bring total hashing power of Bitcoin to PHps range, which is also indisputable. 1000 x 1 Thps = 1 Phps, last time I checked.  Hashing power above Peta Hash is better for Bitcoin than that of 10 Thps and this is indisputable as well.

What you are talking about is an investment case of either investing or not into that hardware. And one thing I know for sure is that I do not know whether it will be profitable or not.

But whoever buys those rigs I will provide them option of having those rigs managed and hosted with both physical and information security done professionally. That's all.

[rjk]

I still wonder if $30 million is too high for full custom, but I guess it would depend on the process node used. For this kind of efficiency, some are assuming very small and expensive nodes like 28nm, but I wonder if it can still be as efficient on a larger and cheaper node such as 65nm. Since nobody has done it before, we really don't have actual numbers to compare with.

[gyverlb]

gyverlb,

You got me wrong, my friend. My post was not from point of view of a miner. But from point of view of an ASIC manufacturer. All I said is that they need to sell 1000 of 30,000$ rigs 1 Thps each to break even

That's what my post was about. I simply noted that at current BTC price, it's unrealistic to think of selling much more than 100 of these.

So if you are going to argue with me here

I won't. In fact I was writing my post before reading yours and included it when I saw the $30M. I probably should have only quoted this part of your post, sorry.

[Vladimir]

fair enough.

BTW, all your numbers are based on one assumption, which is that Bitcoin price will stay at 6.5$ forever. Since this assumption is obviously wrong, there is no need to even look at the rest of your reasoning as it is a logical fallacy already.

Anyway, if you are an investor in Bitcoin, diversify your portfolio, 50% BTC and 50% mining hardware sounds as a good start. Rebalance the portfolio from time to time, maybe add some gold and dividend paying equities and real estate in the mix and you are already doing great. Nothing new here, standard stuff.

[EPiSKiNG]

Agree with Vlad.  This is one of those opportunities to get out of the mining game and still make a decent amount of money back on your hardware, then invest that money in BTC and watch it grow!!

[gyverlb]

fair enough.

BTW, all your numbers are based on one assumption, which is that Bitcoin price will stay at 6.5$ forever. Since this assumption is obviously wrong,

Obviously assumptions are wrong as, you know, they are... assumptions. However the goal of my post isn't to make a clear cut prediction but give some link between the Bitcoin value and the size of the miner hardware market.
For $30M of BFL hardware sold you'll need a BTC price around $50-100 to absorb it without a miner ripoff. But you will have to read the rest of my post to validate this

If you think it will reach this amount when BFL is starting to sell ASICs sure they can sell $30M without their customers losing big, but if it remains in sub $10 territory it will be a bloodfest for sure.

[davidspitzer]

Very interesting discussion. So many possible outcomes. I will be talking to BFL in the coming weeks as I contemplate how to invest my capital. I hope they can deliver on their promises. For every difficulty the right gear at the right time will net profit. The trick is when and what.

[cmg5461]

Very interesting discussion. So many possible outcomes. I will be talking to BFL in the coming weeks as I contemplate how to invest my capital. I hope they can deliver on their promises. For every difficulty the right gear at the right time will net profit. The trick is when and what.

That is correct, but if you miss a cycle, you could be out of your largest chunk of change

[EPiSKiNG]

Very interesting discussion. So many possible outcomes. I will be talking to BFL in the coming weeks as I contemplate how to invest my capital. I hope they can deliver on their promises. For every difficulty the right gear at the right time will net profit. The trick is when and what.

Interested in how that conversation is going to go:

"Hello Butterfly labs, I have $XXX,XXX to spend on your awesome hardware, but I want to be the first person to receive the hardware, and I want you to hold off sending any other hardware out for 2 weeks"

"Sure thing!  But you have to give the money now as a deposit, and it's non-refundable, and we don't have solid specs yet, and we're not sure when we're shipping.   Thanks for your business!"

-EP

[cmg5461]

updated the spread sheet*

https://docs.google.com/spreadsheet/ccc?key=0Ai2b4IJuKUZhdE5EVFJkUF95TS1BVjdxaEFsSVlQSkE