Topic: Cointerra AIRE Miner 16nm PreOrder (Read 19826 times)

The good thing is none of this matters any more since Cointerra filed for bankruptcy   http://www.coindesk.com/bitcoin-mining-firm-cointerra-files-chapter-7-bankruptcy/


The preorder was obviously an attempt to try to save themselves. Guess the community choose to spend their money else where . Hopefully all manufactures that have screwed and continue to screw over their customers have a similar fate [KNC]

i think you've misunderstood something.   the size of the asic doesn't change its power consumption (watts per gigahash/s).  it doesn't change its efficiency.   The cooling of the system has to be designed to be appropriate for the heat output of the asic.

its the design of the hash engine, and process technology used that largely dictate the power consumption (and of course, the voltage its run at, which is clearly a design and process issue)

the hash engines are small and highly repeatable.  there are tens to hundreds of hash engines in most bitcoin mining asics.  perhaps even thousands in some.

the actual size of the asic - i.e., how many hash engines you put in the asic, makes almost no difference to the power consumption (on a w/gh/s basis).

so its up to the asic designer to put the right number of hash engines into the asic that they can efficiently cool the part.

however, the size of the asic does have an impact on system cost.  since the packaging and pcbs costs are directly related to how many parts (packages) you have, there is a cost advantage in going for larger packages that pack more hashing engines into bigger asics that consume less board space.  also, the on-chip circuitry is more efficient and less losses than the off-chip (on board) circuitry, so minimising the latter is desirable.

so in an ideal world, you would put as many hash engines as possible into a package to reduce the number of packages and keep system costs down.   however, as you rightly say, there is a cooling impact.   when you multiply the power consumption of each hash engine by the number of engines you have, you get your power per package.   there are different ways of cooling different amounts of power.

for tiny dies in small packages, like what bitfury and asicminer use, they have very few hash engines in each die... and they keep the power per package to around 2 watts, which is air-coolable without heatsinks, just the packages and board itself become the heatsink.  its cheap and easy to build.. however you need to have a huge number of asics and very large boards, and lots of boards, to produce the same hash power as a more densely packed asic that has more hash engines and runs hotter.  its also potentially more expensive, as the (asic) packaging cost becomes higher (so many packages for a given amount of hash power adds to the cost).  the design process for designing smaller asics is easier, but it doesn't cost much less, as the NRE is the same no matter how big the asic.  And the production is wafer based.  In theory you would have better yields with smaller dies, but in practice, bitcoin mining asics have very good yields as they have so much repetition on the die, that any individual defect doesn't make the asic useless (like it would with an Apple A7 asic)

thus the next step up is the spondoolies style, where they run the asic at probably around 50 watts, and use heatsinks to keep it cool.. but they have a lot more hash engines per package than asicminer/bitfury type design.

cointerra's 16nm design was similar this.   its a good balance of asic density versus power and cooling requirement.  the goal to bring the overall system cost down.

then you have the kncminer approach which is to pack a lot of engines onto one package and have a lot of cooling on each package.  probably 200+ watts per asic?  with huge pc-style cpu coolers.  luckily this method works because pc cpu coolers are plentiful and there's good ones out there that aren't expensive.

and lastly you have the cointerra and hashfast (late 2013) approach, which is to have an enormous number of hash engines on an asic, and run it very very hot (400 watts) and require exotic cooling technologies (liquid cooling) to extract the heat.  it was this approach which made cointerra's systems unreliable because they relied on a cooling tech that had a lot of moving parts (liquid, pumps, fans), and the cooling tech wasn't anywhere near as reliable as it was claimed to be.  and they had four of them in each system, which meant much higher than acceptable failure rate.  not of the asics, which was reasonable.. but for the cooling parts )CoolIT), which let them down badly.

anyway, sometimes you've got to try new techniques to find out that earlier techniques were better all along, and cointerra's 16nm system design didn't share any of the design elements of their previous one.  they learned a lot from their experience.  the 16nm system had well designed asics, well designed systems, and well designed cooling, and it was optimised to be low cost and high reliability.  unfortunately, they didn't raise enough cash to build any of them, even though the design was complete and ready to push the button into manufacture.

i don't have a cloud mining operation.  i was hosting several hundred TH of hashfast and cointerra gear when i was paying $150/KW/Month.  when i was using kncminer gear at their hosting facility i was paying much more.

the 'barebones' data centres ('shed' with power) in washington and iceland are a recently available phenomena... and weren't available to me last year.

I host under $100KW/month, but no by much, BUT I use a lot less power than a cloud mining operation so that's why I'm amazed that you got those rates.

Also in regards to the actually chip-set, my issue isn't the chip itself J/GH, its specifically with the cooling of that chip and the issues you have with maintain the efficiency as transistors get hot.  Lets put it this way, designing something in a computer model or simulation is much different than actual / practical numbers. This is one thing all engineers learn when they leave the classroom (good ones anyways).  How many products have we seen that 'look great on paper' but completely fail when it comes time to deploy?  Jumping to 16nm is very risky for any ASIC company, especially because other companies haven't paved the way much.

The dilemma you face is this, lets say you get 1,000 16nm chips, each chip can do 10gh, they are small chips, they will probably run super efficiently, stay cool, but how many does it take to make a complete competitive miner / operation?  As you increase the die size, you save some money, say 500 16nm chips at 18gh a piece.  Now you have more heat issues, your efficiency drops, but it costs less.  Same thing with larger runs and so on.  More circuits, less circuits, more complex boards, boards melting, these are not easily 'modeled' on a computer and often require samples.

To be honest, this is why chips haven't been dropping in size as quickly recently, the efficiency gains are getting less and less, and the cost benefits are questionable - even for the mighty Bitcoin ASIC's.

The list goes on and on for the complications you find with the smaller and smaller chip sets.  And at the end of the day, if the ROI of the equipment will take someone 1-2 years to return even @ lowest from the wall costs, whats the endgame??

We have a facility mining in the 50's, warehouse / cheap style, not even in Washington state, granted with build-out, its more like 65 for the first year, then drops to $50 per month once TI is paid (includes staffing), power, water, PUE etc. 

We also have pricing in the 60's in Iceland and Washington state, in full DC environments (for spoons), not warehouse style (for bitmain) and get discounted prices for bulk from manufacturers.  Will be able to turn a profit while other mining companies fail and drop out of the pictures.

Right now we're just dealing with large scale investors, but I've been working on getting them to release an actually profitable venture in terms of cloud mining to the masses by splitting the profit and mitigating some of the risk.  I mean we're fine either way, but would rather not see just a few major mining companies own the entire network.

i said it was back of the envelope calculations.  please go ahead and improve the accuracy and show your own calculation.  perhaps we can crowdsource the accurate figures.  but the principle is the same and the numbers are broadly in line, don't you agree?

do you currently host for significantly less than $100/KW/Month?   I'm pretty sure when i got quotes from iceland and Washington, the prices were in the $70-80 for the lowest cost places, and most of america was in the $100-150 range (i myself was paying $150/KW/Month which is why i shut down my mining in october)

That's damn expensive power for a cloud location.

you didnt include the cost of making the asics, of making the wafers, of building the systems, of shipping the systems, of employing the staff to do any of that.

once you add in all the costs, its an unprofitable exercise when btc = $230.   once upon a time, when cointerra started, btc was higher.. and im sure they sold most of their systems when btc was $1000.   the costs didnt change but the revenue did.  the main thing that screws up mining is that the cost of btc is now too low to justify mining.  it dropped a lot in the last few months, making any deal cointerra had completely unworkable and unprofitable.  their mistake was that their entire business plan was predicated on btc not going down to the levels it now has.   but unfortunately for all of us, every mining hardware company is in a similar position.  they all have the same economics.  the costs to make new silicon are the same for all of them, and the potential revenue is the same too.   kncminer and bitfury were lucky they raised some cash last year but with the price of btc where it is, the economics look unattractive to be mining, no matter who you are.

its easy for you to throw stones.  i dont think ive been deliberately arrogant.  sure, ive had the odd argument.  but ive always tried to be balanced, and ive never hidden my bias when i had one.  im also one of the few people on this forum to use my real name, so i can stand behind anything ive said in a way that most wouldnt.  i also show my face at conferences  and repeat my views in panel sessions and take questions etc.  in short, im a very open person and havnt ever hidden my investment in cointerra.  ive also mined with bitfury, kncminer, asicminer, hashfast and cointerra gear.  when it comes to bitcoin, ive always been a miner first, and investor second.  but i switched off all my mining gear in october, when it became unprofitable for last year's hardware to mine.  so now im just an investor.   and the investments ive made so far are very hit and miss.  and mostly miss.  investing in bitcoin companies is extremely high risk!

anyway, i dont agree with your statements.  cointerra HAD finished their 16nm design and it was a good design.  they were way ahead of any other 16nm design at the time (because they had early access and no one else did).  the design was done and dusted in october but they tried to raise cash from investors to tape out at a time when bitcoin price was dropping and investors were skittish.  ct couldnt raise anywhere near enough for tapeout (they needed some $20m, which is seriously what it costs to tapeout 16nm, make first batch of production wafers, and build 1st batch of systems)  the market conditions werent conducive to raising cash for mining companies at that time (and may never be again), and im sure their prior reputation of under-delivery and unhappy customers didnt help one iota.

you claim to be an engineer yet your assumptions about 16nm and asics in general show that you have a superficial and incomplete grasp of asic technology.  the way you write also doesnt inspire confidence that you know anything about asics.  (Die is not spelled Dye, for instance)

however all that aside, the mining business sucks when the price of btc is low - and its super low today.  its pretty much unprofitable for anyone to mine at this time.  its a serious problem for bitcoin in general if miners have no business.  mining is necessary (up until the tech gets redesigned to use a different method of securing the network against fraud and double spends).  even the biggest miner on the network, cex.io (same company as ghash) has now stopped cloud mining.   at the price they can mine with their huge number of bitfury boxes, it has to be unprofitable for almost every gigahash at this point.

do the math, hosting in most places costs more than $100/KW/Month.  In the cheapest places (Washington, Iceland) its a little less, but most of the network isnt in Washington or Iceland today and there are high moving costs to ship it there (as well as high installation costs to deploy it

All the math here is back of the envelope and inaccurate, but, lets play with some numbers and highlight the issue.  Most miners on the (circa 300 PH) network today consume in the range of 0.6 and 1.2 Watts per GH/s (aka J/GH).   lets assume the average efficiency on the network today is 0.8 W/GH/s (its probably much worse as only the recently available hardware is better, but most thats out there is older hardware).  There are 24 hours a day, 30 days in a month, thus each TH of hashing power consumes 0.8 W/GH/s x 1000 watts x 24 (hours) x 30 (days) x $100/KW/Month = $576/TH/month to run.   And at the current network hash rate, each TH of mining power will generate less than 1 btc in a month which at today's value of BTC=$230, means that it costs you significantly more to mine that it generates in revenue.  You can BUY the btc for almost half the price that it costs you to mine it.   And i havnt even included the cost of buying the miner in that calculation, which adds insult to injury.  (as mentioned, all math is uber approx).   This is why no one is selling miners any longer and why there was a transition from people selling miners to people offering cloud mining services (its cheaper to deploy and easier to sell).. but even that, now, is unprofitable for all concerned.

almost no one, no matter where they are, can host cheaply enough to make bitcoin mining be profitable, at the present btc price (and electricity price).   Even with Icelandic hosting prices ($70/KW/Month) or bare bones washington state prices ($50-60) its still unprofitable to mine, today.   note, hosting cost is more than just electricity cost, as you need a building, some wiring, some cooling, some transformers, some minimal staff etc.  but even at raw electric cost, its probably unprofitable right now, too.

Man what crooks, i guess thats the silver lining to these low BTC prices its putting alot the high end of ponzi/scammers/crookded manurfactuerers out of business

let hope bitmain and KNC are next up to bat

Not to mention cointerra probably planned this.  They were probably unprofitable mining several months ago, lets do the quick math:

if you look at the lawsuit from the C7 filed against them, they were paying close to $101 per KW, assuming they were perfectly using 79.99% of each circuit @ 100% up-time.  Realistically, $105 per KW, completely ignoring the IT personnel you need to maintain the equipment, the actual costs break down as:

Cointerra = 2kw for 1.6tera
1 kw = .8tera hash
@ 1 mw, they were hashing at 800 tera for 105k per month (these numbers are based on the 1 MSA, they may have purchased additional services in the future, but scale wouldn't change)

with Bitcoin @ $300.00 dollars, and selling all coins about 132k net, and 30k dollars profit per month.  Instead of paying to innovate and work on realistic / competitive miners with a 20nm or 28nm process, they pocketed profits.

They decided, hey, lets mine for free until they shut us down, then file chapter 11 as its more cost effective to do it this way than paying our DC $200,000 dollars+ we owe them....

my problem was the arrogance you showed when you showed to the situation unfolding.  I come from an engineering background and you dismissed me like some idiot and raved at how 'wonderful' Cointera was when multiple issues were staring you in the face.

Humility is something arrogant people find only after a massive slap to the face, its nice to see you found some. I still believe there are major technological hurdles with the 16nm process that none of these companies have truly addressed, as is the cost.  For now, I would advice staying away from all 16nm processed ASIC chips until we have some actual engineering data from them...  Perhaps small Dye 16nm may be able to handle the heat dissipation load with adequate cooling, but can and will it justify the costs of this process and reach efficiency promised?  Not for the first production runs would be a fair guess.

Quarter 4 this year we may see them, but not with the way the Bitcoin price is right now.

Great point, and I will say the primary hosting providers they used in the short term for their cloud mining left them with serious obligations.. I can't speak to specifics due to NDAs but let's just say they stopped being profitable some time ago even with cloud mining contracts..

the misfortune of others is so funny, graymatter.  hilarious

dang it you beat me to it!  I was just about to post this!  lolz, where's aerobatic now? Hiding in a closet after realizing how TERRIBLE of an investment coin terra was....

http://www.coindesk.com/cointerra-lawsuit-5-4-million-damages/

LOLZ!

http://www.coindesk.com/cointerra-lawsuit-5-4-million-damages/

So at this point I am sure that there will be no TO considering the lawsuit.

If you make 10m a year there is usually a way out of this sort of thing.

A default does not comment on a company's profitability directly as its a hallmark of a company's liquidity. They could be making $10M a year but it would be meaningless if they can't pay their short term liabilities at the right time.

I found this:

http://www.reddit.com/r/Bitcoin/comments/2rkevr/cointerra_is_in_default/

It's weird that they couldn't make money with cloud mining, but their timeline for the Aire was extremely optimistic.