Topic: ASICMINER: Entering the Future of ASIC Mining by Inventing It - page 1346. (Read 3916324 times)

Regarding the die size update to 17mm²

My research, which might not be definitive, shows that large die are incompatible with the MLM process.  Can you comment?


Will you be providing investors with documentation of your arrangements with the foundry?

Could you give us the results of your chat with friedcat?  Inquiring minds want to know.  Thanks

We now have more than 2,000 BTC worth of shares executed as private bulk purchase and about 7,000 BTC worth of shares asked for reservation by potential investors.

So there are still some uncertainties but overall our IPO is doing well.

Thanks for these answers !

One more question: How is the private sale coming along ? The sale on GLBSE is doing very well imo
Will you be able to raise the needed capital before the end of August the way the share sale is going so far ?

Thanks for your questions.

It's already taken into account. We provided the return rate based on a set of possible total network hashrates (therefore difficulties).
You could decide yourself which of them is most possible per month.

It may be feasible, but there are some more facts to consider:

1. When we begin to sell products to customers, limitlessly expansion will harm the buyers' motivation, hence the revenue via selling.
2. If may be better to keep the revenue for R&D of the next-generation of products than expand too much with the original technology, especially that we want to stay in the mining ASIC industry and help securing the network in the long run.

projections are looking good !

I assume the monthly return rate takes into account the increased difficulty ?

An additional question: Why does the company not want to or sees it feasible to expand beyond 50TH/s for their own mining farm ?

Is that an 'effort' / location constraint ?

Thanks

Update

A more detailed estimation about the return is shown online, based on the condition given in the main thread:

https://docs.google.com/spreadsheet/ccc?key=0Al1fvFT7Sd5bdGppUktrYmlIZ3MwRUxhZTk0WEJBd1E

Currently it considers three variables:

1. The electricity fee. (0$/KWh and 0.1$/KWh)
2. The total hashrate of the whole network.
3. Block reward halving. (before and after)

More factors will be taken into consideration.

Thanks for you interest and questions. Would you please PM me your skype id and e-mail address so we could arrange the skype conference and send you the documents you require? Thanks.

I really want this endeavor to succeed and I'm pondering whether to invest some serious money. While I have some issues with the business plan, e.g. I am not thrilled by the idea of self mining, there is one thing much more important.

I would like to be reasonably sure that you won't run off with the money, never to be seen again, or perhaps just pretend to produce the chips. Again, I'm not accusing you of being a scammer, but to me you are just a pseudonym posting some text on a forum board.
Is there anybody on this board who knows you and/or have met you in person? I'm afraid your name being known to glbse is not that reassuring.  How can I be sure that your mystery friends even exist? Would you perhaps be willing to arrange a skype conference call to answer questions and show what you're up to? Anything.

Another way would be to arrange it so that GLBSE locks some of the funds in escrow until a later date. I don't know if that's possible.

Perhaps a start would be some evidence that bitfountain actually exists as a registered company.

If they get around to selling ASIC miners doesnt it make sense to create an ASIC p2pool ?

Update

We collect more questions frequently asked by the community in the past several days here.

Q: What is your fundraising target? How do you decide whether this IPO succeeds or not?
A: The minimum target is 100k$. If the raised funds surpass this number, we will consider that our IPO is successful. However, we would like to raise more (120-140k$) because keeping the budget at minimum is very prone to unexpected scenarios.

Q: So you will send the extra shares at at time?
A: Yes.

Q: How do you achieve so low costs?
A: There are several factors.
  1. 130nm node size. As the mainstream switches to 28nm, the 130nm existed for so long that even many smaller foundries could do it very well. The intense competition of manufacturing in China brings the price of everything down, including ICs.
  2. MLM(Multi-Level-Mask). Compared to full-mask, this technology reduces the cost of mask-set to half with the exchange of increasing the margin cost by about 40%. This is a good deal for us because the margin cost of chips themselves is one of the lowest cost in our budget.
  3. Low EDA license fees and low labor cost in China.
  4. We ourselves did most of the RTL design, optimization and simulation.

Q: Why don't you use Bitfountain for your GLBSE ticker?
A: Because the ASICMINER shareholders (GLBSE investors) have an extra set of privileges upon Bitfountain shareholder (us).

I think p2pool would NOT be advisable for another reason: dynamic range. AFAIK p2pool works best if the miners have equal hashing power. If you add a big guy like an asic miner to the pool, the results may get skewed considerably. p2pool members currently range between 100 GH/s to 100 MH/s. So the smallest guy is waiting for a share 1000 longer than the big guy. If you go beyond 10000 the small guys will wait forever for a share, which defeats the purpose of a pool. (But I read somewhere a discussion of introducing tiers, either geographically or by other means. If they would use hashing power, you could group miners by their hashing power, ideally the integral would be preserved, i.e. each tier has about the same hashing power, higher tiers feat. a few big guys, lower tiers many small ones, but that's not done yet...)

ADDENDUM:  "though large miners have the option to raise their difficulty, and so reduce the impact of their mining on P2Pool's minimum difficulty." (https://en.bitcoin.it/wiki/P2Pool) Which would certainly make sense for an ASIC miner...

OP.  Happy with this endeavor.  Good luck.

IF it is the first ASIC miner in the bitcoin network, it must be in Oct. or Nov., which means that you should put 7200 bitcoin as the daily total output, and this will make a 56% return per week.

And one more thing, if everything goes smoothly and bitcoin exchange rate keeps rally, ASICMINER won't have to spend every bitcoin they raised and it will be given back to the shareholders. (maybe won't have to sell every share of the 200k available.) that will make a even better payback period. Given the potential profit, it is worth to take the risk IMHO.

Thanks for your questions.

We decided to use a different name from the start, because that the ASICMINER shareholders have an extra set of privileges upon Bitfountain shareholders (two of my partners and me).

Yes, it's insanely profitable given that all the given conditions will hold. But "we are first to the market" and the unchanged 15TH/s network hashrate after several months are both strong assumptions.

Before the investors break even, all net profits will be paid to them. So in the first week that's 5,000 BTC minus the electricity fee and place rent.

There are two kinds of failure rates.

The first one is the noise points in mass production. When we get the estimation of 12TH/s we already considered the noise points, or that will be more than 16TH/s on the unrealistic hypothesis that the production process is perfect.

The second one is the failure of making a workable mask-set. It means the waste of the major part of NRE cost. This is the largest risk here and investors should consider this seriously before being in.

The first time that MOORE get boosted will be with or a little after the deploying first batch of our chips. It's like bulk-selling of hashrates from Bitfountain to me myself (I as an individual, am fully responsible of MOORE). So there has to be an approval by shareholders.
Later, if we decide to use MOORE as a future hashrate selling engine, it also needs motion-based approvals.

If at that time, our competitors delivered their products, and the market price of ASIC-based hashrate on the GLBSE market is already settled, we will probably price MOORE at (evolved hashrate at that time with our formula of each share) * (BTC per MH/s) * 1.6. If we are the first to deliver our products, we have to do the price discovery ourselves and what in my mind is keeping the current price of 0.5BTC/share, push the hashrate to 50MH/s per dollar, and fix the hashrate per share based on the BTC/USD exchange rate at that time.

The asics are a quantum leap for mining. So insanely profitable is not unlikely - at least for a few weeks. However, it's a race. The one who has the first working chips will be able to profit the most. Others will just get a fraction of those profits. The real question is the timelag between chip product and mining operation and the rate at which new chips can be added.

At least with ASICMINER the community has an opportunity to participate in the enterprise. AFAIK BFL doesn't offer this - but my guess is they will be doing the same behind the scences and keep all the profits for themselves.

Ok I get it, asicminer is just a glbse asset name representing Bitfountain. It's awefully confusing though. What is the point of choosing a different name if it is one and the same?


If we look at the short term, there are 200,000 public shares that are going to take all the initial profits. The estimated income will be 12TH/s of mining power. Currently the network hashrate is roughly 15TH/s. When 12 TH/s is added you will get 12/27=44% of the daily 25x144=3,600 bitcoins, which is 1,600 BTC/day, or ROI in less than two weeks and a return of 28% per week after that if you're first to market!

If I'm not making any math errors this would be insanely profitable (for a while), and I'm just going to assume it's too good to be true

That is also just gross income in the most optimistic scenario, and I can't find anything about costs. Let's pretend that everything works out and you start mining, and the first week you earn 5,000 BTC. What would the dividend be?


Do you have any estimate of the failure rates? Also, are they somewhat independent between runs, i.e. can we expect that if the failure rate is 10%, say, the probability of all 12 wafers being bad is very small, or is it more likely that if one fails all fails?

I guess my question is what is the likeliest scenario: 1) You get anywhere from 0 to 12 TH/s in the first batch or 2) either 0 or 12 TH/s?


What exactly will MOORE get, and when and at what price?

Thanks. Though some information will be only kept for board members, our progress and milestones will be updated to the public for anyone to see.

Answer to the first question is: our chips are made with I/O simplicity in mind, so there is nothing serious hampering the compatibility on future versions. But as you said, this stage is early and future compatibility is not on the highest priority when making choices.

The second one: We are not planning to disallow that.

There will be. There will also possibly be a small number of free devices for the community. But both needs to be approved with the motions by all shareholders.

Ok im about to be in the market for more mining gear and im going the ASIC way. My faith in BFL being done in time is very low. With that being said I would love to have a choice, and not forced to go with one company. I wish you well in your venture and will be following very very close to progress.