Topic: [PicoStocks] 100TH/s bitcoin mine [100th] - page 39. (Read 470088 times)

Thank you for providing more fundamental figures, so the ASIC chips themselves ain't the problem (don't know queues and production timelines at the foundries to get these small projects done), but getting them deployed to hashing in fully functional data centres will be.

100th bumped into some technical limitations at the pool side as described by Dave. If you run your own bitcoind at the mining facility, this probably won't be an issue.

Is there any problem to be expected in distributing the work to be done at really high hashing speeds. I mean, every calculation to find a hash should be an unique combination within the same mining pool (noob assumption). At which hashing speed of a mine will internal distribution of the work to be done become a problem? Or do we talk infinity then...

Hard to scale a mining operation is good for decentralization of mining!

Maybe fixed rate mines like 100th are actually quite good for the distributed network.

Another 10TH drop in hashrate. Looks like deploying big farm is a big chalenge. AM also have problems to sustain stable hashrate.

run this script, it  shows the error:
But I already install this modue by running:

Thanks!

Just mentioning "Moore's law" does not prove your point. Numbers have been presented in the last few posts that support my "100x diff within a year" argument. Maybe you shows some numbers as well.

Friend of mine, WMP, has created a perl script, to count total and daily dividends of each account.

Link: http://wklej.org/id/1128634/txt/

To use it, you need to install:Under Linux, you can use this command:
Edit script and type your account number in line 8:

Enjoy.

Oh well, pity it wasn't cgminer ...

Pictures ?

We've had two problems with the pool since Friday - we are starting to push the design limits of some of these parts of the system.  On Friday we had a brief outage that was quickly fixed by Eleuthria, related to hitting the upper limits for shares accepted for a single worker.  Another issue cropped up Saturday, in which a portion of the rigs were getting an error returned by the stratum server, which wasn't coming from the pool.  This issue seems to have been related to share difficulty returned through stratum(s) to the rigs.  We've alleviated the problem by balancing the rigs/hashrate across workers, while also managing worker diff settings.

We have received more hardware today that will allow us build another 20 - 25TH.  We continue to work towards optimizing the rigs and getting everything run at top speed...

Well, imho, all vaporware of new companies could be getting very slowly to the market. Every new company with a new ASIC will get setbacks to mass produce fully functional hardware. Technical and organisational.

Even the succes stories ain't spotless: ASICminer doesn't show their projected 200 TH which, iirc, they would have deployed at the start of last summer. Bitfury ASIC retail miners had to rerun parts of their production. 100th mine has been several months delayed and then you also have the usual suspects BFL and Avalon. Completely normal in every starting business and to be expected also of all other newcoming companies, and BFL again for sure, claiming to not need a prototype for their 28nm, since their models are now superior.

The incentive to invest in and to distribute retail products is also dwindling, since one has to compete with mines like Ghash.io, which runs the most efficient chips atm against manufacturers cost, which could be under $1 per Ghash/s according to your assumption. But they probably won't compete with themselves, just building and maintaining their share of the network. So they will follow retail and competitors, not lead them, but they are king for now.

Retail will slow down its pace first (happening right now), due to inherent high cost structure and lower than expected returns of their capital already fixed in miners and preorders. Generated income will probably hardly be reinvested after being scared in these first ASIC production rounds and by current jumps in difficulty. And daily amounts per individual miner are relatively small and they have to save many days to buy a (then hopefully cheaper) miner. Slowing the growth pace more and more, since they earn less and less and are mostly still in the red for their earlier investments.

Public companies like ASICminer come next, due to 'dividend losses' and losing the energy race accelerated by a shrinking share of the pie. Unless they (have) invest(ed) sanely (when they ruled the network) and some monkey comes out of their sleeve on one of these days. AM still makes a lot of investment capital every day, but hardware production and growth of mining capacity is slow compared to demand up till now.

Large private mines will rule anyway and the forementioned Ghash.io is the only serious one right now and they soak up around 20% of the daily production out of the market which is hardly needed for their expansion (they mine per day a private gross investment capital to add ~100 TH mining capacity) and which is not available for other parties to reinvest.

Even if the cash to invest is available (current pre-orders don't run as fast anymore it seems, look for example at HashFast 1st batch Babyjets), the production capacity could very well being the factor restricting a continued high growth pace.

I can't predict the future however and only expect a bumpy road to this hundredfold. Price drops can slow down growth also significantly and the other way around. But I do hope for a 100 Phash/s network rather sooner than later, since it is a positive sign for bitcoin. But not so much for this mine.

Diff will follow Moore's law in that it will increase up to and never beyond what the latest gen ASIC can sustain with manufacturer  profitability. This will be FAR below 100x current diff.

You assumed prices of $10 per Ghash/s. It is well-known current ASIC vendors have insane margins on the raw silicon. As a matter of fact, Cointerra is already selling pre-orders $3 per Ghash/s. Watch the price drop to $1 or less per Ghash/s over the next 12 months. Therefore a $100 million investment in mining hardware is all you need for the network to grow to 100 Phash/s. This will happen by September 2014. Watch.

These examples are a drop in the sea. You can buy miners with more than 1000x the hashrate of USB-Miners and 40x the blades' one today.
Who cares if thousands of them are sold? The hundreds of 400GH/s bitfury miners sold/selling/installed in mines by professional miners make their impact irrelevant.
At one point the professional miners will stop buying if it isn't in their interest anymore. Today even 1 million USB-miners wouldn't even move the difficulty more than what we are accustomed to, how much will they matter in a couple of months?

100 times difficulty ~ 100,000,000 GH/s ~ 1,000,000,000 USD investments in mining gear

Who's gonna pay that?

How could ASICMINER sell their first blades?
Why are people buying dozens and dozens of USB-Miners, which will never ROI according to current parameters?

Because
- People want to mine and don't care about ROI. Remember, people are doing Folding@Home and other "zero-ROI" stuff.
- They use linear difficulty extrapolation at best

Difficulty will be >100x what it is today in less than a year. (No, I'm not taking bets. That's my personal opinion. Share it or don't.)

So you say that people would buy ASICs for hundreds or thousands of $/€ even though they would only give a 100% ROI in decades? In my opinion only pre-orders can put a meaningful amount of people in this position.

From GPU to ASIC mining miners have always avoided hardware with more than 3-6 months of 100% ROI. Given the current rate of difficulty increases they are probably more shy than ever. So most won't "buy hardware as long as it's profitable" they will buy until 100% ROI is so far away in the future that it's too much a risk for them.

Of course. And people are going to buy hardware as long as it's profitable. And maybe even longer.

So an increase of 100x, or even 1000x is everything but unrealistic. When calculating ROI it would be stupid to use anything less than an exponential difficulty increase.

According to my calculations, taking into account my electricity rate and ignoring the initial investment (because it's small compared to the running costs), "creating" one bitcoin costs as much as

- 500€ with a GPU
- 50€ with a FPGA
- 1€ with an ASIC (BitFury)

This means difficulty needs to climb by a factor of 100x until ASIC mining is no longer profitable (current exchange rate). Of course, at an 90x increase, even though you are still profitable, you are going to mine dust.

But then there are countries that have an electricity rate ten times lower than my 0.25€, so for them difficulty can rise by a factor of 1000 (!).

I've already sold. But its profitability goes below .005 per share in monthes even with default optimistic thegenesisblock monthly increase (it shows around 80%, while nowadays it increases 120%, and yet some better/cheaper devices did not hit the market).