This point will always make sense. The 3-month average of the price/diff ratio has always been either stagnating or decreasing over time (sometimes decreasing a lot during crashes), since Bitcoin was first traded on exchanges. This spans the CPU, GPU, and FPGA mining ages.

If you don't know that, then you don't know what you are doing by investing in any mining operation.

Mausini keeps talking about "in the future when operation costs will be high". But at that time, ASICs will be widely available (since competing ASICS will be what causes ASICMINER to have higher operation costs), therefore, again, miners will be able to quickly respond to increases of the exchange rate by deploying new mining capacity.

In principle going from 110nm to 65nm reduces power consumption by 65% (1-(65/110)**2  because power efficiency is proportional to the inverse of the square of the feature size).

But the process node is not everything. Consider ASICMINER for example: they have chips that are more power efficient than Avalon, despite the worse process node (respectively 130nm vs 110nm). ASICMINER does 167 Mhash/Joule vs Avalon doing only 150 Mhash/Joule.

Is this sarcasm? Because I am local to you (Los Angeles), and I would gladly come pick up your Avalon for 50 BTC

When the BTC exchange rate goes up, expected future profits increase only for the very short term. When it happens, other miners add capacity and the price/difficulty ratio returns to its previous level in about 2-3 months. If you look at the history of the price/difficulty ratio, you clearly see this effect and this lag of the difficulty level of at most 2-3 months (more often 1-2 months). This is why I keep telling you that, no, an increased BTC exchange rate cannot significantly reduce the long-term relative operating costs of ASICMINER.

Thanks TechCF!


Four. But I only have three left.

And I explained to you why it will not affect them: https://bitcointalk.org/index.php?topic=99497.msg1624754#msg1624754 But you ignored this post (because your reasoning breaks down?).

(I have been mining and investing in the Bitcoin market since Dec 2010. I have written tools to archive historical Bitcoin economic data and model various investment strategies. I know very well how this market responds to the economic factors we have discussed.)

Ah, I see what you mean, and what you are missing is that as the operating costs increase, the share's fair value would decrease as well, cancelling the later effect of the shares regaining value as the exchange rate increasing.

Example: let's assume the share's fair value is 0.6 BTC today. If the relative operating costs were to increase from 3% to 50% due to increased Bitcoin difficulty, the profit margin would halve from 97% to 50%, thereby reducing the share's fair value in half, from 0.6 to 0.3 BTC. Later, if the BTC exchange rate increases and reduces the operating costs from 50% back to 3%, the share's fair value would double, from 0.3 to 0.6 BTC, and we are back to its original value! This is my whole point: today share's fair value is pretty much as high as it can ever be, because relative operating costs are as low as they can ever be.

I already told you that if BTC went up, relative operating costs can at best go from ~3% to ~0%. This is insignificant.

If BTC doubles from $50 to $100, a share at 0.6 BTC already implicitly gains value ($30 -> $60 = doubling). You seem to think there is an economic reason which would further increase its fair value, say to 1.2 BTC ($30 -> $120 = quadrupling). That makes no sense. I do not know how to explain it more simply to you.

I finally got an email reply from Tom. He offered me a Western Union transfer for half of the amount he owes me. (And the other half in BTC later.)

I sent him my contact info. We will see if he proceeds with actually sending the WU transfer...

opentoe: this an Actel ProASIC3 chip. And despite the marketing name, this is an FPGA, not an ASIC (so I was right, again): http://www.actel.com/products/pa3/default.aspx Based on the VQFP-100 package it has got to be an A3P030, A3P060, A3P125, or A3P250.

It is absolutely not made specifically for Bitcoin mining. There is no way whatsoever that this generic 130nm FPGA beats the power efficiency of the 40nm Arria II FPGA (BFL Mini Rig) or 45nm Spartan6 FPGA (in pretty much all other Bitcoin miners).

Sorry for the false accusation of attempting to scam. It turns out you were merely ignorant of what hardware type you had.

poly: I will win because I did not bet they will ship all units, but merely 10 or more


Wrong? This does not contradict what I said. In fact, thanks for bringing this point: because relative operating costs are guaranteed to increase in the future (mining will never stay this much profitable), they should cause the company's worth (hence the fair value per share) to decrease over time. More precisely, if the BTC price/difficulty ratio stays constant, the company's worth will stay constant. But if difficulty increases faster than the BTC price (which is inevitable as miners transition to more and more efficient ASICs), the company's worth will naturally decline.

Which goes back to my original point that the current high share price (IMHO) can only be justified if ASICMINER develops more efficient chips to maintain profitability, and high dividends.

See my response to iCEBREAKER below.


Except there is no room for relative costs to go down. Assuming 0.10/kWh (worldwide average), ASICMINER's electrical costs currently represent only 0.28% of their BTC revenues (friedcat disclosed an efficiency of 167 Mhash/Joule). Assuming their other operating costs are 10x higher (salaries, data center space, etc), they represent only 2.8% of their BTC revenues.

So if BTC were to appreciate a lot today, making operating costs completely negligible, their profit margin and dividends would at most be increased by ~3%...

Your argument works for gold mining companies, because their relative operating costs are a lot higher than ~3%.

No. They earn BTC, but they are also priced in BTC to start with. Therefore if BTC doubles in value, then both their earning and price already implicitly double in USD value, hence preserving the value proposition of the shares. There are no economic reasons for their numerical BTC value to further increase on top of that change.

Another way to look at it is that the same share priced "X" BTC will return the same "Y" BTC of dividends regardless of the USD/BTC exchange rate. Variation of the exchange rate alone is insufficient to rationally cause "X" to increase while Y" remains constant. In fact, irrational investors who would buy more and more shares for no reason (pushing X up) would cause the dividend yield (Y/X) to decrease therefore making investing in ASICMINER less and less attractive compared to just holding BTC.

Yes I do take it into account: my point is that investing 0.6 BTC in Bitcoin may be more profitable than investing 0.6 BTC in ASICMINER. Because if BTC appreciates to $1000, then at the end of the day 0.6 BTC will be worth $600, whereas your ASICMINER share will be worth $250 (assuming my hypothetical example where the share is worth close to nothing after having returned only 0.25 BTC in dividends).

So, in your investment decisions, you need to compare them against holding BTC. If an investment does worse than merely holding BTC, then it is not an optimal investment.



I think that because friedcat has communicated very little in terms of future plans. What is his vision for growth beyond the next 12 months? What technical strategy will he adopt to compete with more efficient miners? Will he develop more efficient chips? Does he have the expertise in-house or will he have to hire it? Will he relocate the miners to a location with cheaper electricity? Will he have to hire foreign employees to manage them, or rent expensive oversea colo space? Will he consider building his own data center like Tytus (for his 100 Th/s farm which will be located in Wenatchee where power is only $0.02/kWh). Etc.

No, the 0.25 BTC return that I estimated does average out the period of time after and before shareholders have recouped their IPO investment. In other words, ASICMINER will represent 10% of the global hashrate on average, from the moment they launched until the end of 2013 (IOW right now they are at 15% and it will go down to 5%).


No, that would not be a 10% profit. What I explained is that if you buy a share at 0.6 BTC today, and recoup 0.25 BTC by the end of 2013, and if after that the share's worth is close to 0 (because by that time ASICMINER will generate negligible profits), then you have not made any profit. You have lost 0.35 BTC per share.


Even if you make power free or ignore all operating costs, ASICMINER won't generate significant profits: if the difficulty increases by 30x by early 2014, which is a reasonable estimate, then one share (assuming they are at 50 Thash/s) will generate merely 0.014 BTC per month ignoring operating costs. Assuming you buy a share at 0.6 BTC today, and assuming you recoup about half the investment by early 2014 (by my estimate in my previous post), you have 0.3 BTC remaining to be recouped, and at 0.014 BTC per month, at a constantly increasing difficulty, you will basically never do it...

The only shareholders who will make money are those who bought low (0.1-0.3 BTC, eg. those who went in the IPO), or those who manage to buy & sell at the right time (the share may be volatile enough to have lows & highs despite being overpriced). That, or you may make money if ASICMINER's future unknown plans about scaling beyond 50 Thash/s with more efficient chips turn out to be at least moderately successful. Maybe they will but that is not certain at all.

http://www.ddrdrive.com - it was designed for ZFS ZIL, and the RAM is backed by SLC NAND flash. It was priced $2000 when it launched ~3 years ago. The price has probably gone down a lot, but it is still way more than 10x the price per GB than system's RAM.

Part of the reason prices are so high is because there are so few vendors, so there is practically no competition.

I predict that opentoe will ask whoever is interested in a board to "just pay for shipping" to get the "free" board and then somehow not receive it or receive something that turns out is not a Bitcoin mining board

Your math completely fails to take into account how quickly difficulty is going to rise.

Look, even friedcat estimates that ASICMINER will average only 10% of the global hashrate over the year. It means each share will bring: 25 (coins/block) * 6 (block/hour) * 730 (hours/month) * 9 (months) * 10% / 400k (shares) = merely 0.25 BTC of revenue between now and Dec 2013 (1/10th of your "2.22 BTC" estimate!)

Plus the 0.25 BTC of revenue per share assumes 100% of the mining revenues are going to be distributed to shareholders. We know this is not going to be the case given friedcat's plans (he will reduce the dividend).

And in 2014, all bets are off. ASICMINER's efficiency (130nm chip) is relatively bad compared to other ASICs (55-65nm). They are not hosted in a country with low electricity prices. I can't see how they will remain profitable in 2014. Plus I think 10% of the global hashrate is itself an optimistic estimation for 2013.

This is why I think 0.6 BTC is overpriced for a share that definitely will not return more than 0.25, maybe 0.3 or 0.4 BTC if being overly optimistic.

The only possible justification for a price of 0.6 BTC/share is if you assume that ASICMINER is going to plan to build a more efficient chip, and if they are going to succeed, and if they are going to deploy and ramp it up as fast as (or faster than) the competion. But at this point it would be a very speculative gamble. There are too many "ifs" for me.

Joking of course

See https://bitcointalk.org/index.php?topic=152030.0

Do it, 32GB only costs $160, assuming you already have a machine with a motherboard supporting 4 x 8GB DDR3.
But don't buy hardware ram drives. They are all horribly overpriced, as in 10x more expensive per GB, or more.