Yeah but you would have paid as much as for another sierra and the MPP modules would only arrive 3+ months later. Possibly a lot later than that, HF is totally mum on when they will ship
For the same money, would you rather have 1 Sierra in november december, or 3 Sierra's in March/April/June if network doubling continues ?
Protip, if doubling continues, a Sierra in March will earn 1/8th of a sierra in December.

I beg to differ. Market value of those asics is directly and proportionally related to difficulty and therefore to your time to market. Whats true for an individual miner is just as true as for a company selfmining or selling gear. You go tell a KnC or HF customer it doesnt really matter when they get their gear.

And if you meant to say they lost the race anyway and it doesnt matter anymore if its February or June, I also beg to differ -except for the part where they lost. This race is far from over; in fact it has barely even begun yet and it will continue until next summer at least. Every month you are late your margins are almost cut in half. KnC very much did the right thing by "rushing" to market with a far from perfect product. Their profits from their october shipments alone would pay for another, more perfect design.

Duh. Of course. But you dont seem to understand that your upgrade is limited to 4x. Difficulty increase over 3+ months is not so limited.

With MPP you get 2x the GH/$ (4x the original hashrate but you pay ~2x as much to have MPP) 3 months later.
So how does that "guarantee" work as long as the network speed keeps growing much faster than doubling every 3 months ? Its been doubling every 3 weeks.


Something to chew on: ordering a prospero X3 (2TH @ $4900 promised for late February) is probably a better "guarantee" than a MPP for the Sierra.

So you got the wafers, 11-21 days after your promised shipping date, and all the while you have been saying the silicon wafer production was "on track".
The substrate issue aside, how did you ever expect to ship those miners 2 weeks before getting silicon?

If you are sending your miner for collocation, there shouldnt be an import tax levied, since the hardware doesnt change owner. In reality in most countries this is quite a battle, and you may get charged import taxes even on RMA items, that you have to claim back,  but with the right paper work (read: get a specialist lawyer), it can be avoided. Its one of the things Im looking in to.

IM no icelander, but Im considering offering a solution. Check my sig.

Yeah, let the scammer hand out some crumbs of the loot to bagholders that bought the stock for pennies on the dollar, and all will be forgiven by the IPO investors Im sure.
You want to buy some Pirate debt too at 1% of face value? You know, just in case...

People would send their own miners there. They wouldnt even have to trust me, they would be sending their hardware straight to the datacenter. For payment Im sure we can work something out with escrow/multsig and the DC is open to using bitpay.

Its just that they dont like to be bothered with dozens of individual customers ordering 4U at the time, hence an organized group buy. Once they are in Iceland, its also a lot easier to move them to another premise.

(btw, iceland is a small island. You are not going to run very far with a container full of stolen hardware).

KnC hasnt been shipping for 2 weeks and will only resume next week I believe
HF isnt shipping anything yet and wont until december
CT isnt shipping anything yet and wont until december or january
Actm isnt shipping yet and wont until december or january
Bitmine isnt shipping yet and wont until december or january
Black arrow isnt shipping yet and wont until February
BFL Monarch isnt shipping yet and, well, who knows really.

You get the idea.
The only thing thats still shipping right now (allegedly) is 65nm BFL miners, but they dont have much bite anymore, and perhaps some bitfury boards and USB miners.

List of rack units of potentially interested miners:

- 6x 4U CT anonymous via PM
- 2x 4U CT anonymous via PM


Total: 8U

Hi all,

Today, few among you will care about electricity price. Its only a tiny fraction of your mining investment and revenue, and your ROI is determined almost solely by when you receive your hardware. I dont think many people would be interested in shipping their beloved hardware far abroad right now, I guess many would rather sleep next to it

However, this will change. In 6 months or so, profit margins will dwindle and at some point, most of our revenue will be eaten up by electricity (and hosting) costs. Many among us who pay residential electricity prices will not be able to compete with large scale mining operations that get their electricity far cheaper. So, lets solve this

Im considering setting up a company to provide bitcoin collocation services in Iceland. Why Iceland?
- $0.043 per KwH for industrial contracts, guaranteed for up to 12 years.
- 100% green energy from hydro and geothermal power plants.
- attractive business climate, unlike most other countries that offer such low electricity cost (Iran, Venezuela, Kuwait,..).
- not too far from either the EU or US.

The plan currently is simple and consists of 2 phases. First phase is simply a group buy to rent rack space at an existing tier 3 or 4 datacenter in iceland.

I requested a few quotes, and what I got so far is substantially lower than the hosting that Cointerra advertises on their website.
Cointerra charges $2599 for 6 month hosting of a terraminer 4, or $433 per month.
What I got so far works out to about $200 per month, less than half Cointerra's price and competitive with many European domestic electricity rates alone.
Remember, this is for hosting in a tier 4 datacenter with all provisions, like UPS, diesel backup, multiple backbone connections, fancy access control, fire suppression, on site permanence etc. Price is based on the assumption the machine will use ~1400W and if we can get enough interest to fill at least 5 racks.

In a second phase, if our margins drop even further and $200 per month per 4U miner begins to hurt,  assuming we can get enough miners to fill 20 or more racks, I am looking in to buying one or several modular datacenters (fancy word for a datacenter container) and putting them in Iceland. Something like this:



(And yes, I will be looking at asicminer's submersion cooling solution if/when it becomes available, although I have my doubts it will be economically more viable).

These systems would have their limitations, for instance I would likely not be providing UPS power and diesel backup, but at a later stage in the mining game, if our margins are that thin, I dont think those costs are warranted. If you only pay for the electricity you use, downtime would be "cheap" because our margins would be low anyway. I may reconsider that, but thats my current thinking. I cant give a reasonable price estimate for those containers yet, but obviously the idea is to be significantly below that of tier 3/4 datacenters. I suspect we may be able to cut that price by another 30-50%, but thats only a rough guess.

Before anyone asks, this wont become a bitcoin security that you can invest in, nor will it be a preorder scheme, if I decide to go ahead I will fund it myself, or at most accept private investors.

Time frame I have in mind, is to start renting the racks in spring 2014, meanwhile prepare everything for our own modular datacenter and depending on market conditions, likely implement that by summer/autumn 2014.

Right now, I would just like to hear from people if they are interested, and how many U you may want to rent and for what timeframes. Feel free to PM if you dont want this in public.  IM also open to suggestions and criticism of the concept.

Where the crime took place is actually a legal quagmire. I suspect legally it took place in every country where someone bought shares.

Yep. I was looking in to similar solutions to put somewhere with cheap electricity.
There are many suppliers, prices are harder to find. But this caught my attention, just to have a reference point:
http://www.theregister.co.uk/2011/09/15/data_centre/

The smallest SGI solution costs $100K and provides room and cooling for 204 U and 148KW or 51 Terraminer IV's.
Thats actually fairly affordable.

By comparison, grcooling offeres a submersion cooling container for $2/W
http://www.grcooling.com/docs/2013-07-09-Containerized-CarnotJet-Systems-Now-Available-from-Green-Revolution-Cooling.pdf
 
$2 * 148K = ~$300K

3x more expensive for the same capacity.

Now Im not saying it cant be done cheaper for either approach, but Im not so convinced submersion is the way to go.

Are you reporting this to the italian police, or the HK police, or both?

I was thinking the same. Plus power efficiency is (or will) be even more crucial than with almost any other application and sales margins -for the time being at least- are higher than any other asic I can think off. The key question is what margins (for a more energy efficient chip) will be like next year, but if they can sell them soon enough, Im also inclined to believe that it can happen.

There are no 60 days left this year. So much for December delivery.

yeah just a $3000 container and some  bottles of fluorinert. lol
for the record, similar solutions exist commercially, and are sold (in containers) for $2.per watt.  that's more affordable than a brick and mortar datacenter but the price of the container itself doesn't really matter

I would also dispute that datacenters suitable for mining  don't exist  the requirements for hpc and gpu compute are quite similar.

Sure, but it would still only be 1.58% of the power consumption as long as you dont need AC. Which of course isnt likely, but if you followed the thread, that was my point.

How much do you think the low speed fans on a KNC Jupiter consume ? 2W per fan? Total of 8W. 12W if you count the case fans which KnC users disconnect because they (weirdly) get better performance at higher temperatures. 12W out of a total of ~650W at the wall is 1.58%. Thats not massive in my book, thats statistical noise.


Yes, I was confusing with hashfast which claim 400GH per chip. Not that the comparison with KnC's 100GH chip looks so good.


They all use the same chip. Its not because a blade contains 32 asics that the per asic performance is somehow better, nor that 32 asics would cost less than 32x the price of 1 asic.


Who cares? Someone buying it may care, but as a sharesholder dont you understand your company is selling 96 asics (+PCB, assembly, etc) for ~$500 ? <$5 per chip whereas hashfast is charging $1500 for a module.  How much do you think these chips cost to produce?


Hu?