Bitcoin Forum
February 22, 2020, 07:36:03 AM *
News: Latest Bitcoin Core release: 0.19.0.1 [Torrent]
 
   Home   Help Search Login Register More  

Warning: Moderators do not remove likely scams. You must use your own brain: caveat emptor. Watch out for Ponzi schemes. Do not invest more than you can afford to lose.

Pages: « 1 ... 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 [831] 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 ... 1348 »
  Print  
Author Topic: ASICMINER: Entering the Future of ASIC Mining by Inventing It  (Read 3901525 times)
blunthammer
Newbie
*
Offline Offline

Activity: 38
Merit: 0


View Profile WWW
January 23, 2014, 07:02:26 PM
 #16601

New to the thread. I am a data center guy. I know containers and sold one to NASA in 2009. I know data centers and the companies who own and run them. I have started two of them, one funded the other not.

I wanted to interject something as it relates to the overall solution - finding a place that can and will take a container, back it with an SLA you can get at another 'traditional' facility, and deploy it is VERY tricky. I know this because I spent most of 2013 building the business plan, team, four sites, and facilities to build a container data center. I ran out of money before we got live, but here is what I can share based on that year long experience:

The dividing line between a high density facility and 'you're full of sh*t when you say you're that dense' is 30Kw/cabinet. I spent $10,000 to have a third party who does a lot of market research for Dell validate this with 26 'high density' providers in 4 states where I wanted to put these facilities.

Liquid cooling is absolutely an option in the short term, but the back of house operations, maintenance and servicing the tanks increased the points of failure (risk). Techs don't like dressing up in aprons, gloves, and getting greasy and walking on slippery soles either.

The solution we designed cost $6M/MW (building, mech/elec, container, TI's all in) and cool 34KW/rack and 600Kw of actual critical load (750Kw capacity) on a site with a 1.3 PUE. Cooling towers + containers, traditional UPS. Concurrently maintainable.

The customer base were movie studios, geospatial, and HPC customers. About 2% of the overall data center market. I had not looked at bitcoin miners yet.

I love that there are companies building out integrated stacks (hardware, software, containers), the challenges will be delivering the reliability we have come to expect from a Tier III/concurrently maintainable facility:

Carrier density to mirror other transactional standards (<100ms)
Security - physical
Compliance - customers buy risk, compliance (right or wrong) provides standards
Reliable efficient power acquisition and distribution within the site
Managed buildings/sites who understand containers (offloading, landing, testing, commissioning)
Lease terms flexible enough to handle increasing or varying densities, dedicated infrastructure adds, moves, etc. to align with hardware cycles
Geographic diversity with a vendor who understands the above

You may think 'all of this non technical stuff doesn't matter' but  has for 30 years in the data center industry and while there may be a higher risk tolerance at the outset, that will evaporate with the first financial or technical hiccup or crater that happens. Don't fight the mindset, exploit it.

The end game is a well designed and optimized facility or facilities that can handle the flexibility and changes that are inherent with the guts of a miner - the hardware and the things that happen on it.

mark at blunthammer is the best way to get me if you want to do a deep dive on any of this.

Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1582356963
Hero Member
*
Offline Offline

Posts: 1582356963

View Profile Personal Message (Offline)

Ignore
1582356963
Reply with quote  #2

1582356963
Report to moderator
1582356963
Hero Member
*
Offline Offline

Posts: 1582356963

View Profile Personal Message (Offline)

Ignore
1582356963
Reply with quote  #2

1582356963
Report to moderator
sngwinner
Full Member
***
Offline Offline

Activity: 433
Merit: 100



View Profile
January 23, 2014, 09:43:40 PM
 #16602

"I love that there are companies building out integrated stacks (hardware, software, containers), the challenges will be delivering the reliability we have come to expect from a Tier III/concurrently maintainable facility:"

I'm an AM bull, but the operations has been pretty reliable as is with the "data center" being what looks like nothing more than a few shelves in a broom closet.

I don't know what a tier III facility is but it sounds costly and is it absolutely necessary?

ETERBASE | TRADE WITH NEGATIVE FEES
xbase ▬▬▬■▌[SIGN UP NOW]▐■▬▬▬ xbase
ANN THREAD   |    TELEGRAM    |    FACEBOOK    |    TWITTER
blunthammer
Newbie
*
Offline Offline

Activity: 38
Merit: 0


View Profile WWW
January 23, 2014, 10:35:49 PM
 #16603

How expensive is the hardware?
How expensive is it to burn the building the closet is in down because there is no way to mitigate heat that is generated from the hardware?


Slow hardware=less heat

Powerful hardware=more heat

Computers turn electricity into heat no matter what. If the hardware needs to crunch numbers and algos faster they will run hotter. CME, NYSE, and the racks that house the rendering software for motion pictures all run 25kw+ because they get data crunched faster. The challenge every one of them has is heat rejection.

So if I have $3M in hardware in a rack running 30Kw means it's costing $100,000 per Kw of processing. You'll pay $10k a month to put that rack in a secure building where you have a contract that says you'll have less than 5 minutes of downtime per year. So if it costs you $100,000 a KW one time, and $120k a year to house and cool that hardware, you need to clear $240k to break even.

How much can you mine a year?
webbrowser
Full Member
***
Offline Offline

Activity: 215
Merit: 100


View Profile
January 24, 2014, 04:54:22 AM
 #16604

?  If hosting costs USD 10k/mth, at current exchange rates, that is 12.5 BTC and can be recouped using half of the current block reward.
burger
Full Member
***
Offline Offline

Activity: 195
Merit: 100


View Profile
January 24, 2014, 06:17:12 AM
 #16605

....
Slow hardware=less heat

Powerful hardware=more heat

....

I would say that my old gfx card (about 700 Mhash) that is a slow hardware generates more heat than a 5Ghash BFL ASIC. So

Slow hardware = more heat

Efficient hardware = less heat

ineededausername
Hero Member
*****
Offline Offline

Activity: 784
Merit: 1000


bitcoin hundred-aire


View Profile
January 24, 2014, 11:42:16 AM
 #16606

friedcat did tell me on the 14th that tapeout was still on track for the 20th, so I'm pretty confident in a tapeout at least this week.

(BFL)^2 < 0
empoweoqwj
Hero Member
*****
Offline Offline

Activity: 518
Merit: 500


View Profile
January 24, 2014, 12:17:37 PM
 #16607

friedcat did tell me on the 14th that tapeout was still on track for the 20th, so I'm pretty confident in a tapeout at least this week.

Where did he tell you that?
pierrejo
Donator
Sr. Member
*
Offline Offline

Activity: 290
Merit: 250



View Profile
January 24, 2014, 12:48:06 PM
 #16608

friedcat did tell me on the 14th that tapeout was still on track for the 20th, so I'm pretty confident in a tapeout at least this week.

Where did he tell you that?

Just like there are exchanges necessary to move shares around, some questions are likely to find their way in said exchanges - I would bet that "are we still on track for a Jan 20 tapeout?" is one such question.
minerpumpkin
Hero Member
*****
Offline Offline

Activity: 686
Merit: 500


A pumpkin mines 27 hours a night


View Profile
January 24, 2014, 01:33:27 PM
 #16609

friedcat did tell me on the 14th that tapeout was still on track for the 20th, so I'm pretty confident in a tapeout at least this week.

Where did he tell you that?

Just like there are exchanges necessary to move shares around, some questions are likely to find their way in said exchanges - I would bet that "are we still on track for a Jan 20 tapeout?" is one such question.

Possibly. When I read 'friedcat did tell me...' I just thought 'haha, troll elsewhere' But then I realised this wasn't a newly registered kid.

So... really?  Smiley *hopesUp*

I should have gotten into Bitcoin back in 1992...
Dargumin
Member
**
Offline Offline

Activity: 107
Merit: 10


View Profile
January 24, 2014, 11:35:14 PM
 #16610

Everyone too busy dumping shares to post here? Share price is tanking hard Sad

Was my post useful to you?  Tips graciously received!
1MZXNoqzRgNp8rA7DjPVV6R21kcTKGt31T
empoweoqwj
Hero Member
*****
Offline Offline

Activity: 518
Merit: 500


View Profile
January 25, 2014, 02:09:59 AM
 #16611

Everyone too busy dumping shares to post here? Share price is tanking hard Sad

Not exactly a shock. If people were buying on the "20th January" tapeout date and apart from a "rumour" posted last night, absolutely no news from FC about it, the share price was always going to head south again. Dividends almost evaporating this week didn't help.

A post from FC is getting rarer than a Man U win these days Wink
tomywomy
Jr. Member
*
Offline Offline

Activity: 57
Merit: 27


View Profile
January 25, 2014, 02:24:50 AM
 #16612

How much hashing power would Asicminer have to have to justify this share price?  There are 400,000 shares, so at 0.5 BTC per share we are looking at a value of 200,000 BTC.  If Asicminer discovered 8 blocks a day they'd make 200 btc a day in revenue.  That's 3.3% of the network.  Even if 50% of that was profit for distribution, that would mean 100 btc a day.  It would still take 2,000 days, or 5.5 years at that rate to generate 200,000 BTC of profit and get the shareholders buying in at this price to break-even.  

With only 4 blocks a week being found at the current hashing power, we are talking about increasing hashing power by 14 times to go from 4 to 56 blocks a week, or 8 blocks a day.  Does Friedcat have access to 14 times the data space he is currently using, 14 times the electricity, 14 times the chips?

What would it genuinely cost to maintain 3.3% of the entire hashing power?  Right now the network hash rate is about 20 PetaHash. 1% is 200 TeraHash and 3.3% is 660 TeraHash.  

This isn't just about deploying some faster technology.  The network hash rate has grown 500 times (50,000%) since May, when Asicminer ruled the mining world.  Do you really believe Asicminer can produce 50 times the hardware they had been (to end up at 3.3%) and keep that growing at the network growth rate, which has been 100% per month for many months?

I'm wondering if someone who is a believer here can quantify what they think Asicminer will be producing that will justify the value currently being assigned to these share.
jimmothy
Hero Member
*****
Offline Offline

Activity: 770
Merit: 500



View Profile
January 25, 2014, 02:31:46 AM
 #16613

How much hashing power would Asicminer have to have to justify this share price?  There are 400,000 shares, so at 0.5 BTC per share we are looking at a value of 200,000 BTC.  If Asicminer discovered 8 blocks a day they'd make 200 btc a day in revenue.  That's 3.3% of the network.  Even if 50% of that was profit for distribution, that would mean 100 btc a day.  It would still take 2,000 days, or 5.5 years at that rate to generate 200,000 BTC of profit and get the shareholders buying in at this price to break-even.  

With only 4 blocks a week being found at the current hashing power, we are talking about increasing hashing power by 14 times to go from 4 to 56 blocks a week, or 8 blocks a day.  Does Friedcat have access to 14 times the data space he is currently using, 14 times the electricity, 14 times the chips?

What would it genuinely cost to maintain 3.3% of the entire hashing power?  Right now the network hash rate is about 20 PetaHash. 1% is 200 TeraHash and 3.3% is 660 TeraHash.  

This isn't just about deploying some faster technology.  The network hash rate has grown 500 times (50,000%) since May, when Asicminer ruled the mining world.  Do you really believe Asicminer can produce 50 times the hardware they had been (to end up at 3.3%) and keep that growing at the network growth rate, which has been 100% per month for many months?

I'm wondering if someone who is a believer here can quantify what they think Asicminer will be producing that will justify the value currently being assigned to these share.

AM will deploy 2-20ph during what I predict to be a 40-400ph network hashrate.

So that would give AM from 0.5% - 50% of network hashrate.
empoweoqwj
Hero Member
*****
Offline Offline

Activity: 518
Merit: 500


View Profile
January 25, 2014, 03:41:32 AM
 #16614

How much hashing power would Asicminer have to have to justify this share price?  There are 400,000 shares, so at 0.5 BTC per share we are looking at a value of 200,000 BTC.  If Asicminer discovered 8 blocks a day they'd make 200 btc a day in revenue.  That's 3.3% of the network.  Even if 50% of that was profit for distribution, that would mean 100 btc a day.  It would still take 2,000 days, or 5.5 years at that rate to generate 200,000 BTC of profit and get the shareholders buying in at this price to break-even.  

With only 4 blocks a week being found at the current hashing power, we are talking about increasing hashing power by 14 times to go from 4 to 56 blocks a week, or 8 blocks a day.  Does Friedcat have access to 14 times the data space he is currently using, 14 times the electricity, 14 times the chips?

What would it genuinely cost to maintain 3.3% of the entire hashing power?  Right now the network hash rate is about 20 PetaHash. 1% is 200 TeraHash and 3.3% is 660 TeraHash.  

This isn't just about deploying some faster technology.  The network hash rate has grown 500 times (50,000%) since May, when Asicminer ruled the mining world.  Do you really believe Asicminer can produce 50 times the hardware they had been (to end up at 3.3%) and keep that growing at the network growth rate, which has been 100% per month for many months?

I'm wondering if someone who is a believer here can quantify what they think Asicminer will be producing that will justify the value currently being assigned to these share.

Read all about ASICMINER's "vision of the future" from their post on 10th December. The only question is whether it happens or not ...
ninjarobot
Hero Member
*****
Offline Offline

Activity: 761
Merit: 500


Mine Silent, Mine Deep


View Profile
January 25, 2014, 05:54:40 AM
 #16615

Oh, how the mighty have fallen! How did we get here and will we ever see those glory days again?

I'm still holding AM, mainly because I am too lazy to sell. But if Gen 3 fails, I fear all is lost.

FC mentioned that he does not like to work for others, what is preventing him from neglecting AM and letting it die a slow death while rebooting privately? His identity is still unknown so he can do whatever he pleases. Keeping my shares on faith alone is tough.
Entropy-uc
Hero Member
*****
Offline Offline

Activity: 756
Merit: 501


View Profile
January 25, 2014, 06:00:26 AM
 #16616

Oh, how the mighty have fallen! How did we get here and will we ever see those glory days again?

I'm still holding AM, mainly because I am too lazy to sell. But if Gen 3 fails, I fear all is lost.

FC mentioned that he does not like to work for others, what is preventing him from neglecting AM and letting it die a slow death while rebooting privately? His identity is still unknown so he can do whatever he pleases. Keeping my shares on faith alone is tough.


A study of Discus Fish would lead most to think that your scenario is already underway.
empoweoqwj
Hero Member
*****
Offline Offline

Activity: 518
Merit: 500


View Profile
January 25, 2014, 06:04:53 AM
 #16617

I think "if Gen 3 fails, I fear all is lost" is a completely fair statement. But if it succeeds, a lot is won at current share prices. That's why I've not sold out.

But the lack of communication from FC is getting very disturbing now. This must be  the longest ever gap between his posts since IPO. If he doesn't announce something real soon, I'll be leaving the ship, whether its sinking or not. Can't make investment decisions in a vacuum.
BitHub
Sr. Member
****
Offline Offline

Activity: 253
Merit: 250


View Profile
January 25, 2014, 06:31:07 AM
 #16618

there is a reason why FC isn't saying anything because gen 3 wont happen in time. Get out now while you can. Even if they put out gen 3 its already too late. FC doesn't care about asicminer anymore for months they been working on something else that doesn't require investors. I feel sorrry for those still holding on but the risk is sooo huge dont let 10btc turn into 1btc please. I'm trying to sell off my last 100 shares very slowly at a lost but good to get something back.
jimmothy
Hero Member
*****
Offline Offline

Activity: 770
Merit: 500



View Profile
January 25, 2014, 07:13:43 AM
 #16619

there is a reason why FC isn't saying anything because gen 3 wont happen in time. Get out now while you can. Even if they put out gen 3 its already too late. FC doesn't care about asicminer anymore for months they been working on something else that doesn't require investors. I feel sorrry for those still holding on but the risk is sooo huge dont let 10btc turn into 1btc please. I'm trying to sell off my last 100 shares very slowly at a lost but good to get something back.

You might want to make a new account. Nobody here takes you seriously even for a fudster.
jimmothy
Hero Member
*****
Offline Offline

Activity: 770
Merit: 500



View Profile
January 25, 2014, 07:16:44 AM
 #16620

I think "if Gen 3 fails, I fear all is lost" is a completely fair statement. But if it succeeds, a lot is won at current share prices. That's why I've not sold out.

But the lack of communication from FC is getting very disturbing now. This must be  the longest ever gap between his posts since IPO. If he doesn't announce something real soon, I'll be leaving the ship, whether its sinking or not. Can't make investment decisions in a vacuum.

Does FC need to hold your hand too when making investments? Updates come monthly and its been like 1 month and a few days since last update. But your right AM is totally doomed and we should all jump ship right before next gen.
Pages: « 1 ... 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 [831] 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 ... 1348 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!