ASICMINER: Entering the Future of ASIC Mining by Inventing It

[Chris_Sabian]

and havelock crashes.

Of course.  That happens alot though unfortunately.  I just hope that they don't pull a MtGox

[hdbuck]

Sorry folks,

FC talks and we crash 

Working to bring it back up...

Lmfao

[whalezy]

Crash on a 200 shares buy?? 

Exactly the opposition I think.

[friedcat]

Development
The 28nm BE300 engineering batch tapeout was in September 16 with TSMC.
The power efficiency is 0.225W/G to 0.343W/G ranging from core voltage of 0.55V
to 0.70V.

Huh, interesting: Initially the low-power-mode of gen 3 (BE200) was supposed to be <0.2W/GHs, now the planned efficiency is only 0.225W/GHs with the BE300s... How come?
When will the first physical chips be tested, when will we get to know if the power consumption is fine?

Thanks a lot for the update, friedcat!

BE200 was supposed to be <0.2W/GHs because we planned that it has some large room of downvoltage. However most room was
eaten by the mistake we made in chip and package design.

For BE300 in theory it is possible to get lower than 0.225W/GH by stressing the voltage down below 0.55V, but we do not have
solid simulation data yet, so let's see after the test chips are out.

December 16 is to be expected for us to get the chips. Testing time varies at 3-10 days since we had much more preparation work already done this time.

[ujka]

Thanks on update. BE300 seems like it's coming as planned.

[rudi]

Friedcat: Thanks very much for the update, and congratulations for the good work.

[FUR11]

December 16 is to be expected for us to get the chips. Testing time varies at 3-10 days since we had much more preparation work already done this time.

Alright, great! Please let us know immediately as soon as the initial and most important tests are concluded!!!

[Blazed]

Wow Havelock sucks...down again. Any chance of moving AM to something more reliable? Any thoughts on some type of non centralized platform (counterparty/ colored coins / etc...)?

[jdany]

FC -

Your candor is appreciated.
When will we get a glimpse of the closing quarter's financials?

Can we get an idea what self mining looks like right now?

[whalezy]

Update

Development
The 28nm BE300 engineering batch tapeout was in September 16 with TSMC.
The power efficiency is 0.225W/G to 0.343W/G ranging from core voltage of 0.55V
to 0.70V. The two main mistakes made in BE200 were dedicatedly addressed and
avoided in the whole design stage so we do not expect surprise in frequency and
power consumption in real products.

14nm/16nm projects are at pre-evaluation and pre-design stage. It has to be done
in 2015 but the starting time of placing orders depends on overall gain vs (NRE/R&D/risks).

Led by Block Erupter Prisma, a series of efforts to reduce component-wise
cost and system-wise power efficiency were made.

Production
As known by OEM producers, several large batches of BE200 had problems with
popping chips affecting whole boards due to the misoperation from the packaging
company. It hindered afterwards chip sales and delayed the starting time of
Tube/Prisma sales. Later supply of BE200 has no such problem and filtering process
on old BE200 is also underway.

Because of the non-open-source status of Prisma and high cost of all other designs, BE200
sales were stopped. Rest BE200 chips will likely be consumed by Prisma and succeeding revises.

The mass production time of BE300 in terms of chip-out date is February to March, 2015.

Operation
As sales were seriously hindered by unexpected yield issue of BE200, we had missed many
dates with financing the big mining farm of our own. It is still hopeful that all good BE200
will be turned to hashrate before too late and we have liquid profits again before BE300, but
from the big picture, the missing of target on BE200, the mis-judgment on OEMs and later
yield/financing problem, limited our expansion and profit in this year. But luckily R&D on later
products is not much affected.

AM will continue to develop new products, expand its mining farms, deliver machines to miners,
as well as make money for shareholders.

There is no any foreseeable plan from AM side on force buy-back, dumping, or any
activities on the open stock market. There is no plan of privatization either. The management,
R&D, sales as well as marketing is being more and more separated and dedicated to professionals,
but the founders will continue to be heavily involved and drive AM in the more and more competitive
market. Scenarios of BE100 vs graphics card may never happen again, but we need catch
any opportunity on timing window of leads at either technology or deploying/financing resources
to win the last battle when mining is still an under-capitalized game.

Any way, thanks, FC. Although I lost a lot of money on AM shares, I'm glad to hear departments of AM is aiming to be more separated and professional. Packaging problem is obvious due to the absence of quality management system. If AM sent engineer to package factory, this failure would be avoided.

[elasticband]

Update

Development
The 28nm BE300 engineering batch tapeout was in September 16 with TSMC.
The power efficiency is 0.225W/G to 0.343W/G ranging from core voltage of 0.55V
to 0.70V. The two main mistakes made in BE200 were dedicatedly addressed and
avoided in the whole design stage so we do not expect surprise in frequency and
power consumption in real products.

14nm/16nm projects are at pre-evaluation and pre-design stage. It has to be done
in 2015 but the starting time of placing orders depends on overall gain vs (NRE/R&D/risks).

Led by Block Erupter Prisma, a series of efforts to reduce component-wise
cost and system-wise power efficiency were made.

Production
As known by OEM producers, several large batches of BE200 had problems with
popping chips affecting whole boards due to the misoperation from the packaging
company. It hindered afterwards chip sales and delayed the starting time of
Tube/Prisma sales. Later supply of BE200 has no such problem and filtering process
on old BE200 is also underway.

Because of the non-open-source status of Prisma and high cost of all other designs, BE200
sales were stopped. Rest BE200 chips will likely be consumed by Prisma and succeeding revises.

The mass production time of BE300 in terms of chip-out date is February to March, 2015.

Operation
As sales were seriously hindered by unexpected yield issue of BE200, we had missed many
dates with financing the big mining farm of our own. It is still hopeful that all good BE200
will be turned to hashrate before too late and we have liquid profits again before BE300, but
from the big picture, the missing of target on BE200, the mis-judgment on OEMs and later
yield/financing problem, limited our expansion and profit in this year. But luckily R&D on later
products is not much affected.

AM will continue to develop new products, expand its mining farms, deliver machines to miners,
as well as make money for shareholders.

There is no any foreseeable plan from AM side on force buy-back, dumping, or any
activities on the open stock market. There is no plan of privatization either. The management,
R&D, sales as well as marketing is being more and more separated and dedicated to professionals,
but the founders will continue to be heavily involved and drive AM in the more and more competitive
market. Scenarios of BE100 vs graphics card may never happen again, but we need catch
any opportunity on timing window of leads at either technology or deploying/financing resources
to win the last battle when mining is still an under-capitalized game.

[elasticband]

BE200 was supposed to be <0.2W/GHs because we planned that it has some large room of downvoltage. However most room was
eaten by the mistake we made in chip and package design.

For BE300 in theory it is possible to get lower than 0.225W/GH by stressing the voltage down below 0.55V, but we do not have
solid simulation data yet, so let's see after the test chips are out.

December 16 is to be expected for us to get the chips. Testing time varies at 3-10 days since we had much more preparation work already done this time.

This will be the most important 2 weeks in AM history. Great work getting us this far and keeping us on track for the future!

[MilkyLep]

friedcat do you feel the BE300 is already highly optimized (minus futher undervolting tests) enough to not make revisions to the chip/package etc itself?

Would you also be able to elaborate more on the operations side relating to DataTank or even AMHASH?

Any chance of an estimated date for financials?

[Bonam]

Definitely nice to see an update from FC. Reading between the lines, it seems that dividends are unlikely until Spring 2015, contingent on gen 4 being profitable.

The post clearly states that there are no "liquid profits" presently due to the gen 3 issues, and that there is only a "hope" that BE200s might be put to use mining profitably before that.

[jimmothy]

Anyone know what these are? Giant miners?

[novusordo]

Anyone know what these are? Giant miners?

"Enhance!"

[nycgoat]

Anyone know what these are? Giant miners?

Potentially immersion cooling.  I saw these and a bunch of other pictures on Instagram today.

http://instagram.com/asicminerhash/

[teek]

i think its compressor based cooling of some kind..  kinda like industrial style portable a/c?     def not immersion cooling as someone stated.

[hdbuck]

friedcat do you feel the BE300 is already highly optimized (minus futher undervolting tests) enough to not make revisions to the chip/package etc itself?

Would you also be able to elaborate more on the operations side relating to DataTank or even AMHASH?

Any chance of an estimated date for financials?

[Chris_Sabian]

Definitely nice to see an update from FC. Reading between the lines, it seems that dividends are unlikely until Spring 2015, contingent on gen 4 being profitable.

The post clearly states that there are no "liquid profits" presently due to the gen 3 issues, and that there is only a "hope" that BE200s might be put to use mining profitably before that.

I would tend to agree with that statement.  I sounds like all of the money is being used to fund gen4