Topic: Cointerra AIRE Miner 16nm PreOrder - page 5. (Read 19889 times)

Fancy and complicated answer for a simple question. How low can TerraMiners go with efficiency was my question. From your answer I understand that they can go as low as 0.23W/GH which I don't believe it. Maybe you were referring to the chip efficiency, not the whole system, but even so I still find it hard to believe it. I want to see it!


That's a total of 7MW on top (or not?) of their 20MW facility in Canada. I am sorry, but I am not impressed. Bitfury can do 20MW every 2 months. That's impressive! Remember that they still need a couple of millions on top of NRE costs in order to build their new miners. Also we must not forget that HF defaulted with how many tens of millions in pre-orders? 30M$? 50M$?

You should be impressed because they have at least had 1MW added every month since April, which got them a good amount of profit already.
They also sold much of that power at a massive profit to their cloudhashing customers, so instant profit there.

I guestimate that it is not running much better than 1W/GH but their operational cost should be quite low in terms of $ /kWh due to economies of scale.
(Then again, only their newly built 20MW facility in Canada, which isn´t finished yet, should have the very lowest cost per kWh).


This recent option to pre-order their newest miner is only a way for them to recover/finance the MASSIVE NRE (which already is around $7-15M for 20nm development) faster.

Actual internal cost of these units could be around half of the current 28nm cost, which is slightly below 0.4$/GH on good 28nm designs.
(Assuming cheap labour costs and very high production volume, a little higher for water-cooled setups).

in the web management pages, users have the ability to tell it how much power to use (power stepping mode), and it will scale the performance accordingly.  this should allow people to choose the optimum power and performance for their situation, if they're paying a lot for their power.  of course, with many miners now having moved or are moving to cheaper places to host, like washington state, where the cost of power is less than half the cost of power in the rest of the US (and 1/10th of the cost of power in California)... there will still be a few more months of profitable mining for many miners.   but yes, the writing is on the wall.. and the current generation of miners is surely on their last legs and need to be replaced with more power efficient miners.   cointerra's 0.23w/gh miner is one such but I'm sure there will be others

luckily, difficulty has slowed right down and the next step will probably be the smallest difficulty jump in a very long time - which could allow bitcoin miners to hang in there awhile longer.   and who knows, difficulty might even go backwards now that the price of bitcoin is so low and the cost of mining is relatively speaking, so high.

How low can the TerraMiner go with efficiency? I remember them being over 1W/GH so having 20MW of that efficiency is close to losing money at this point so I'm not impressed of anyone having so much speed deployed if they are close to no profit.

clueless newbie - saying that you spoke to an actual engineer isn't quoting a source.  to do that you need to say exactly who you spoke to.

anyhow... thats ridiculous to say 16nm isn't ideal for bitcoin chips and only useful for mobile chips.  what a ridiculous thing to say.   the reality is that anything thats good for mobile chips, which are low voltage and low power, is exceptionally good for bitcoin mining chips as well since they too need to be low voltage and low power.  in fact, i think you'll find that bitcoin chips run at even lower voltages, on the whole, than mobile phone chips do..!

one thing I'm pretty sure of, is that in 2015, a large amount of bitcoin mining will be done on 16nm chips, and by 2016, probably most of it will be 16nm.  its just so much better than 28nm (and yes, even 20nm) that there's no reason not to use it for everything.

btw, nvidia has already announced they're using 16nm for their next graphics chips.

cointerra may have the first 16nm to be ready for tapeout, but i guarantee within the space of 6 months, a lot more 16nm are coming.  its the obvious next step for the majority of the asic companies that want the lowest possible power and smallest possible silicon area (= more dies per wafer).   sure, some smart companies can eek out a bit more juice out of 28nm and survive for one more generation in 28nm, but ultimately in 2015 they will all be using 16nm,.. cointerra's just the first, thats all.  they might be a quarter or two ahead of the pack, but the whole pack is going there!

(btw, 16nm tsmc and 14nm samsung/globalfoundries are pretty much similar in specs, so for our purposes are interchangeable)

To be fair, 16nm process should actually be suitable for power efficient mining chips (unlike 20nm FinFet, which is really most useful for mobile applications).

If Cointerra uses enough silicon, it should be both power efficient and achieve a massive improvement in density, presumably only limited by cooling ability.

I have to agree that standard air cooling is most likely not practical using 16nm high power (density) architecture.


However, closed loop water cooling systems would likely solve most of the thermal issues that come with this chip design. (Similar to the current Cointerra miners).



While I wouldn´t advise "normal" people on pre-ordering this product, it will most likely be the only chance of getting your hand on these specific miners at a ?reasonable? price.
Additionally, Cointerra is by now a bigger player in the ASIC mining industry, with around 20MW of own mining operations, and a good amount of sold product, so the chance of them defaulting over this product is smaller.

Entertaining to say the least-

Fact's-

Only devices using 16nm technologies are cell phones, using TINY amounts of voltage and amps, thus, low heat.

Intel's 14nm processor is suffering from massive overheating issues, they are now triyng to work in 3 dimensional space, as the 14nm / 16nm resistors generate so much heat that it melts the wafers.

No 16nm chip produces has had the type of heat running through it like these ASIC's will produce (unlike 20nm, use in processors, and GPU's)


Speculation-

ITS VERY BAD, that no processors, or GPU chips (high heat) have been made with this technology.  This chip will fail considering intel, with 2 years, and 50million dollars invested haven't been able to produce an efficient 14nm processor chip, I doubt these under qualified engineers have a chance in hell at doing anything that will work. 

Do NOT PRE-ORDER, it may take several months to a year before 16nm chips are ready for ASIC production.

Sources - ACTUAL ENGINEER, consulting with engineer friends who work on the 14nm intel chip production plant in Chandler, AZ....

I used to work for Cointerra. I didn't like the way they handled things there so I left. It was utter chaos in the beginning with people not sure what to do.

Many months later, I brought a Terraminer and it died on me. Cointerra wanted to charge me almost half of my machine to fix it. I should have listened to my instincts.

DON'T GIVE THESE PEOPLE ANY MONEY.

^^^  we want PRE-DELIVERY!!  .  Yup let them pay for it themselves.

jez,
I like you and cannot fault you for trying to get a return on your shares in Cointerra.
But Cointerra has proven that their first priority was to build a mine and that many of the customers that financed their buildout and enabled them to build a mine, are very unhappy.   I think that if Cointerra now wants to get a better chip for their own mine and equipment that requires less power (for their own mine), then they should pay for it themselves.
As I have said before, the only way that I will ever do business with Cointerra is IF COINTERRA PRE-DELIVERS.   That is simply, that Cointerra sends me the machines FIRST and when (and if) I mine enough to pay for the machines, I will pay for them.

This seems VERY fair to me since we trusted them and financed their company with the communities pre-orders, they can now trust us and PRE-DELIVER.

All it takes is that none of the big driver/mining software makers outthere will adapt the miners to use the cointerra chips. Hopefully someone puts this thing together, else all this talk is for nothing .. you still need to build your rig to drive these units Happy mining!

BBB investigation on CoinTerra: http://www.bbb.org/central-texas/news-events/bbb-warnings/2014/06/bbb-investigation-cointerra-sells-bitcoin-mining-computers-that-consumers-allege-dont-meet-promised-specs/

16 nm sounds awesome.
Cointerra sounds terrible.
No thank you!

He is used with AM products.

investor = free R&D loan.

They never said they will be shipping in 3 months or have efficiency below 0.3 w/gh.

They advertise 0.33 w/gh and q1 shipping means march 31 at 11:59pm not January 1st.


Did they not advertise 0.6w/gh for their first gen which turned out to be 1.2 w/gh?


What makes you think their new architecture is less prone to failure/underperformance?

Yes, you can't currently buy more efficient hardware but as with all preorders there will be more efficient hardware by the time they ship.

Bitfury announced 0.2w/gh hardware to be released this year and 0.1w/gh in 2015.

so youre saying instead of 3 months they will be 9 months for delivery... and instead of sub 0.3 J/GH, they will be 0.6 J/GH ?

on what basis can you make those claims?   these guys may not be popular but theyre also not the epic fail that you like to make out.

i realise its now cool to knock them and since they dont monitor the forums they never get a chance to reply.. but cointerra arent in the same league of 'badness' as some make out.

1.  they shipped!  For that alone, they deserve a prize since so many other asic companies DIDNT ship or shipped months late.. and for most of ct's customers they shipped pretty much on time (they actually shipped most of their batches in the month they were due).
2.  their hardware was 20% below performance target, and they offered anyone who didnt wish to take delivery a refund (and a number of their customers did take that refund) - the customers that did decide to take the miner were given a discount off a future order.

there are many other bitcoin asic companies out there that missed their publicised performance and didnt offer anything at all...  or who were grossly late that the rise in network difficulty made anything that was eventually received worth far less than cointerra's boxes, which in my case, have definitely roi'd for me (as have my knc boxes, but clearly my bfl and hashfast didnt roi).

lets put a bit of perspective on this.. cointerra is undeniably one of the bigger bitcoin mining asic companies.  they currently (ie: today..) are running one of the largest cloud mining operations - running over 10 Petahash using their own boxes.   there's only a small number of other hardware companies that have produced hardware capable of doing that - and the biggest by far is BitFury and the other (same PH's as cointerra) is KnCMiner.  If cointerra's boxes were 'garbage' they wouldnt be able to run that scale of operation - and i believe they run pretty well autonomously.. ie, there's little to no staff holding their hands and giving them constant tlc.  these things are working 'at scale' as well as anyone else's boxes.

so to summarise.. you really think they will be grossly late and extremely under performing.   And youre entitled to your opinion, but from what i know, and also from the experience gained by them last time, i believe they will do a better job this time (but the reality is that the job they did last time wasnt perfect but it wasnt terrible either).  their architecture this time is more resilient to many of the issues they experienced last time.  and i cant buy any other system thats lower power than this one, at this time.   low power is the best way of having an roi positive mining operation.

Will you be satisfied 9 months from now when you are finally delivered your 0.6w/gh miner with no compensation?

I'm both a customer and investor

so you a buyer(pre-order)